Embed Size (px)
Citation preview
National Bank of Poland
THE ECONOMIC I NSTITUTE
IN COLLABORATION WITH REGIONAL BRANCHES
Report on the situation in the Polish
residential and commercial real estate
market in 2011
WARSAW, AUGUST 2012
2
The Report has been prepared at the Economic Institute in collaboration with sixteen
Regional Branches for the purposes of NBP authorities and presents the opinions of the authors. The document should not be understood as an advisory material or a basis for investment decisions.
3
Authors: Part I Augustyniak Hanna Economic Institute Gajewski Krzysztof Economic Institute Łaszek Jacek Economic Institute Olszewski Krzysztof Economic Institute Waszczuk Joanna (technical support) Economic Institute
Part III Baldowska Grażyna Regional Branch in Warsaw Barska Ewa Regional Branch in Bydgoszcz Białach Ewa Regional Branch in Lublin Borzym Henryk Regional Branch in Olsztyn Broniecki Waldemar Regional Branch in Olsztyn Czapka Izabela Regional Branch in Katowice Czechowski Tomasz Regional Branch in Zielona Góra Czekała Magdalena Regional Branch in Wrocław Decyk Paweł Regional Branch in Gdańsk Gałaszewska Krystyna Regional Branch in Gdańsk Hulboj Izabela Regional Branch in Zielona Góra Jung Katarzyna Regional Branch in Wrocław Kiernicki Jarosław Regional Branch in Bydgoszcz Krzyżanowska Kinga Regional Branch in Kraków Książczyk Jolanta Regional Branch in Łódź Lekka Marta Regional Branch in Szczecin Leszczyński Robert Regional Branch in Białystok Leśniewicz Artur Regional Branch in Poznań Mach Barbara Regional Branch in Rzeszów Mach Łukasz Regional Branch in Opole Markowska Janina Regional Branch in Wrocław Masiak Małgorzata Regional Branch in Wrocław Mikołajczyk Łukasz Regional Branch in Opole Misztalski Maciej Regional Branch in Wrocław Myszkowska Barbara Regional Branch in Warsaw Opioła Zbigniew Regional Branch in Katowice Orliński Sławomir Regional Branch in Kielce Osikowicz Grażyna Regional Branch in Kraków Owczarek Ewa Regional Branch in Szczecin Perczak Jacek Regional Branch in Kielce Piwnicka Małgorzata Regional Branch in Poznań Tomska-Iwanow Anna Regional Branch in Szczecin Tyszkiewicz Robert Regional Branch in Łódź Warzocha Jolanta Regional Branch in Rzeszów Zadrożna Iwona Regional Branch in Gdańsk
4
Summary ................................................................................................................................... 5
Part I The real estate market in Poland in 2011.................................................................... 8 1. Key issues in the analysis of the residential real estate market .......................................... 8 2. Trends in the private residential real estate sector in Poland in 2011 ............................... 11
Housing stock ............................................................................................................. 11 Situation in the residential market in Poland .............................................................. 14 Residential sector environment .................................................................................. 19
3. Relations between the financial sector and the residential real estate sector .................... 21
Housing loan risk faced by banks ............................................................................... 25 Factors affecting demand for housing loans ............................................................... 31 Box: Analysis of particular elements of risk inherent in residential and commercial
real estate investment, including the risk run by the lending bank .................................. 33
4. The real estate development and construction sector ....................................................... 36 5. Commercial real estate in Poland ..................................................................................... 45
Office space ................................................................................................................ 46 Modern retail space – shopping centres ..................................................................... 49 Other commercial space ............................................................................................. 51 Warehouse space ........................................................................................................ 52
Glossary of terms and acronyms........................................................................................... 53
Part II Analytical research papers ....................................................................................... 57 A1. Development trends in the local markets (comparative analysis of 16 cities in Poland)
..................................................................................................................................... 57
A2. Modelling of cycles in the residential real estate markets – interactions between the primary and the secondary market and multiplier effects. .......................................... 86
A3 Real estate development enterprises in the Polish market and issues related to its analysis ...................................................................................................................... 111
Part III. Monographs of 16 cities in Poland ....................................................................... 118 1.1 Main developments in the Warsaw housing market ..................................................... 118
5
Summary
The results of the studies presented in this Report lead to the following conclusions: • The year 2011 saw light, in nominal terms, and somewhat higher in real terms, falls in
home prices. Following price surges in the years 2005-2007, the adjustment process spread over time, gradually reduces the risk of drastic price shifts in the housing real estate market in the near future. Yet, as a result of rather rigid home prices in the primary market, driven by high output concentration and upfront financing system the six largest cities saw growing supply of unsold home construction contracts. These effects were enhanced by accelerating home constructions in response to the entry into force of the Act on the protection of home buyers’ rights. Moreover, an import reason behind it was the generally weak situation in the remaining sections of the construction industry.
• Housing policy continued to be focused on supporting home-ownership. This led to persisting demand for housing loans. Growth in PLN lending, despite concerns that this might generate risk to the banking sector, was fuelled by the government-subsidized housing program “Family’s own housing” (Rodzina na Swoim). • In 2011, housing loans were the key element of banks’ credit portfolio (approx. 36%). A rise was noted mainly in the loan debt of individuals resulting from housing loans in PLN. Yet, the annual growth in PLN loans was lower in comparison to the 2010 growth, which is indicative of the downward trend in lending. In 2011, the volume of housing loans granted in foreign currencies was limited and in 2012 Q1 a fall in foreign currency debt of individuals resulting from housing loans was noted. • As a result of the currently observed and further anticipated declines in home prices some credit portfolios are not fully hedged. Also household burden resulting from loan repayment is on the rise. Yet, until now, these phenomena have not threatened the banking sector’s stability. • The recently observed progress in the implementation of new prudential regulations for the banking sector, has been translated into more prudent bank lending policy; yet, no progress has been noted in housing policy regulations. • The real estate development sector, which was one of the main beneficiaries of the housing boom, in currently undergoing a restructuring. Consequently, real estate development companies should reckon with higher real estate development risk. Although the previously observed fundamental factors for those enterprises such as the profitability of their production, did not raise any major concerns, that their situation may deteriorate. • Real estate development involves considerable risk, such as the cyclical nature of the market, risks inherent in general business conditions, political and legal risks, not to mention construction or natural disasters. As a result, with allowance made for risk premium, profits are generally higher than in other industries. • The discussed phenomena were observed in sixteen local markets. Despite differences in market developments and price levels, growth rates of those prices did not differ considerably across those markets. • Cycles in the commercial real estate market in Poland are related to space demand, which is largely dependent on the overall economic situation and the specific character of particular sectors of the market. In Poland, this sector has greatly relied on foreign capital and has been dominated by international investors. More investment projects may be expected to be carried out in small and medium-sized cities financed with Polish capital and through banks operating in Poland. Consequently, it will be necessary for the financial supervision authority to adjust the prudential procedures accordingly.
6
Introduction The Report aims to provide concerned economic agents, including real estate market
participants, with fairly complete, reliable and objective information on the situation in the residential and commercial real estate market in Poland in 2011. The Report focuses mainly on the 2011 phenomena which directly impact current situation. Yet, whenever justified, backward-looking insight was also provided.
Due to local nature of housing markets, similarly to the previous editions of the Report, sixteen markets of voivodship cities are the object of an in-depth analysis, yet, in various outlays. The Report is based on the data derived from various sources. The main source of information on housing prices and structure have been data originating from the Real Estate Market Database (BaRN) created by the NBP for analytical purposes1. The study also relied on a database prepared and updated by PONT Info Nieruchomości containing data on offer home prices, the SARFIN database of the Polish Bank Association containing data on housing market financing and AMRON database containing data on housing appraisal and transaction prices of loan-financed housing. The authors have also drawn on the reports issued by the Polish Financial Supervision Authority (KNF) as well as aggregate credit data released the Credit Information Bureau (BIK). The statistical data published by the Central Statistical Office (GUS) and analyses including sectorial data2 have been used in the structural analysis. The authors have also made use of the findings of the “Social Diagnosis” survey. The information about the commercial real estate market is based on data provided on a voluntary basis by commercial real estate brokers, as well as real estate management and consulting companies. The analysis has been supported with knowledge of experts of the particular agencies3.
Although many sources of information have been used, missing data or insufficient quality of data have proved a significant barrier. In such situations, the authors have relied upon estimates verified on the basis of expert and specialist opinions. In drawing up the Report the authors have assumed that even estimates, verified in several sources, provide better information than general opinions.
Technical terms, defined in the glossary of terms and abbreviations following the first part of the Report, have been marked with’#’.
The Report consists of three parts. Part I presents some common processes in the real estate market in Poland, Part II consists of analytical studies, providing more insight to certain issues discussed in Part I. Part III is more detailed and includes appendices presenting data of each of the sixteen markets of voivodship capitals.
Chapter 1 in Part I presents key issues of the analysis of the housing real estate market.
1 Detailed information on the BaRN data base are presented in Annex no. A1. 2 This concerns, in particular, Sekocenbud studies on the structure and costs of construction, research conducted by the company Real Estate Advisory Service (REAS) on home prices and the real estate development sector, of the Polish Construction Research Agency (Polish: Polska Agencja Badawcza Budownictwa (PAB)) concerning the construction sector and many other entities and associations operating in this market. The most important ones included the Polish Association of Polish Banks (Polish: Związek Banków Polskich), Polish Association of Home Builders (Polish: Polskie Stowarzyszenie Budowniczych Domów), Associations of Employers – Producers of Construction Materials (Polish: Związek Pracodawców-Producentów Materiałów dla Budownictwa) and many others. 3 The authors relied on the data and information provided by the following agencies: CBRE, Colliers International, Cushman & Wakefield, DTZ, Jones Lang LaSalle, Ober Haus and the following associations: Retail Research Forum of Polish Board of Shopping Centres, Warsaw Research Forum and the data base: comparables.pl.
7
Chapter 2, due to the importance of the housing market, focuses on the six major urban markets (Warsaw, Cracow, Wrocław, Łódź, Poznań, Gdańsk, sometimes together with Gdynia which makes seven cities altogether). Annex 1 presents a comparative analysis of developments in the sixteen local markets in Poland and discusses the demographic and housing situation, economic factors and residential construction. It also analyses the data gathered in the Real Estate Market Database (BaRN). Annex 2 also presents an overview of models of cycles in the housing market.
Chapter 3 examines interactions between the financial sector and the real estate sector by analysing the risk of housing loans to banks, analyses factors affecting the supply of and demand for housing loans. The analysis of investment risk in housing and commercial real estate, including the risk to the financing bank is presented in a box.
Chapter 4 depicts the situation in the real estate development and construction sectors, analysing construction costs, real estate developers’ profits and the sector financing methods. The complimentary annex 3 highlights issues related to real estate developer costs and profits analysis throughout the whole construction process.
Chapter 5 describes the commercial real estate sector in Poland (office, shopping centres, warehouses), including the major issues concerning rents, vacant premises and capitalization rates in this sector.
8
Part I The real estate market in Poland in 2011
1. Key issues in the analysis of the residential real estate market
Real estate markets, including residential markets, are subject to cycles and are determined by local factors. This dependence is the result of local interactions of a variable demand and rigid short-term supply, which results from the relation between the real sector of the economy (real estate developers, construction companies, home buyers), the financial sector (providing financing for home construction and purchases) and the public sector (regulating the market).
In Poland in the years 2006-2008 such factors as a lending boom, income growth and demographic factors dramatically boosted demand, resulting in higher home prices. The developments which were observed in 2011 in the Polish residential real estate sector were largely the consequences of the sector reaching its new equilibrium, a process which has been under way since 2008. The current situation in the market, which is determined by the volume of housing production, interest rates and market home prices, features weaker demand and lower prices.
Another important group of processes is the effect of the natural development of particular local markets. As a result of those processes, disparities between the analysed cities diminished, yet, this was not a linear or unidirectional process. The phenomena observed in the past few years were analysed in detail in the previous editions of the Report4.
The residential real estate sector is analysed as a system composed of various economic
segments, pursuing a common economic objective, namely the generation of income from services provided by the housing stock. Housing is considered as a capital good generating a stream of services rather than a consumer good. Services may be sold to third parties or may be internally consumed. The main components of the housing market are:
• the housing stock, generating housing services for households,
• the financial sector enabling home purchases by changing capital into a stream of periodical payments and providing the sector with new capital inflow amidst growing demand,
• the residential construction sector, including, in particular, the real estate development sector transforming financial capital flowing into the sector, into fixed capital in the form of new housing stock,
• the external environment of the housing real estate sector, or the remaining part of the domestic economy with many sectorial interactions.
From this point of view, the analysis of the residential sector is the analysis of its components and relationships between them. As a result, there are new, important interactions affecting the components, the whole sector and its environment such as the domestic
4„Report on the situation in the residential and commercial real estate market in Poland in the years 2002-2009” (2010), „ Report on the situation in the residential and commercial real estate market in Poland in 2010” (2011).
9
economy. The sectorial equilibrium is the main focus of the central bank. Such an equilibrium is a state, under which different segments can generate goods and services in a regular way, thus economic aims are achieved without excessive risk accumulation.
The analysis of housing services involves mainly the analysis of the situation in the residential market, mechanisms of housing needs satisfaction and domestic housing policy from the consumer’s perspective. The housing policy or sectoral policy often has a strong impact on both the quality of housing needs satisfaction as well as the distribution of the housing stock and housing services.
The financial sector is the main driving force behind demand in advanced housing markets. The financial sector analysis involves both the examination of the assets of the banks’ balance sheets (impact on the housing market liquidity and on investment in residential construction) as well as their liabilities (analysis of the sources of financing, financing instruments, investors and deposit holders). This provides an answer to the following questions: a) whether the sector regularly offers financial instruments enabling liquid trading of the housing stock and its long-term financing without creating excessive tensions in this stock and b) whether the sector makes it possible to convert savings into housing capital goods. Borrowing costs and home prices should be adjusted to household income in the local market. A permanent provision of financial instruments means that they need to be adjusted to the needs of the primary and secondary markets. Other necessary conditions include sustainable and satisfactory economic performance of financial market agents as well as the prevention of the financial and market risk accumulation in the financial sector and residential market. The adjustment of the supply of necessary instruments to meet the market needs is a complex issue. Due to the cyclical nature of the residential market and its vulnerability to speculation and collective behaviour, the market and the financial sector, if unsupervised, may trigger cycles and crisis situations. As a result, it is necessary to manage housing demand in various ways. As experience shows, the most effective form of housing demand management is, apart from fiscal intervention, is a conservative behaviour of specialist banking. Moreover, the demand should be managed through systemic prudential regulations, market monitoring by financial supervision authorities and ad-hoc legal regulations.
The investment function of the financial sector transforms financial capital into housing assets and, as referred to in the previous point, in long-term financing of housing assets.
The analysis of the financial sector’s liabilities should account for sustainability of the sector’s financing – should ensure that liabilities match assets. This means that the assessment of sustainability of the sources of financing in the long-term and also the assessment of interests of capital providers i.e. rates of return generated in the long term should be analysed. The literature emphasises that sustainable systems have a diversified financing structure, thus financing origins from the deposit market and the capital market5. This ensures new loan financing and refinancing of the old portfolio even if one part of the system does not perform appropriately.
The investors’ analysis has to take into account their expectations and alternative investment options. This is of particular importance during the period of transition from the deposit-based system into capital market based system. In the long run, the financial instruments need to generate a positive rate of return.
5 E.g.. Lea M., Diamond D.B. (1995). Sustainable Financing for Housing. A Contribution to Habitat II. Fannie Mae Office of Housing Research.
10
Analysis of the construction and real estate development sector, thus concerning both the market and single companies, provides an answer to the question whether housing output is and will be profitable. A separate aspect of the analysis, aimed to assess output stability, is the assessment of the market situation, both from the demand-supply ratio point of view and the economic situation of firms.
The last element of the analysis are the relations of the real estate sector with other economic sectors. The residential real estate sector has strong interactions with the rest of the economy. From the point of view of economic policy, including monetary policy, the relations between mortgage banking and the rest of the financial sector, and between the residential construction sector and the rest of the economy are of particular importance. As monetary policy is pursued by influencing interest rate, and the residential sector is generally vulnerable to this instrument, the assessment of the impact of monetary policy on the situation in this sector is of particular significance. The reverse impact is equally important. As the bulk of factors affecting the sector development originate from the economic and social environment (for example income growth, foreign migrations and urban migrations), the analysis of the environment is essential for the assessment of the situation in this sector. An additional factor having a major impact on these relations is housing policy, in particular, policy supporting the sector. The coordination of monetary policy, fiscal policy and prudential policy are the key issues since, as experience shows, they are often contradictory and boost the sector’s inherent risk.
11
2. Trends in the private residential real estate sector in Poland in 2011
Housing stock Poland’s housing situation, as measured with the housing stock to population ratio is
adequate to the country’s level of advancement. It improved in 2011, which was mainly driven by expanding single-family housing construction and construction based on owner’s resources in smaller cities, towns and villages, and multi-family housing construction in the real estate development in the largest cities. Fading fundamental factors of housing demand suggest, that in a longer perspective, the new sector equilibrium will feature a smaller size of housing construction and lower home prices.
In 2011 there were no major changes in the housing policy. Since the beginning of the transformation period, this policy was unilaterally focused on owner-occupied housing. No serious measures were undertaken to relax the excessive, administrative protection of tenants, in advantage of an economically efficient protection, with no major disturbing effects on the market. The difficult budget situation forced the government to trim the scale of subsidies under the government-assisted housing scheme Rodzina na Swoim, which, from the point of view of the market situation, was a positive measure. It should be emphasized that in 2011 the long-expected Act on the protection of home buyers’ rights was put in place6.
At present, Poland’s housing stock comprises mainly multi-family housing located in the sixteen voivodship cities and some medium-sized cities as well as single-family houses in smaller cities, towns and villages (see: Figure 1). The growing share of owner-occupied housing is a persisting trend.
Figure 1 Structure of the Polish housing stock7 broken down into forms of ownership*/
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2002
2003
2004
2005
2006
2007
2009
2002
2003
2004
2005
2006
2007
2009
2002
2003
2004
2005
2006
2007
2009
2002
2003
2004
2005
2006
2007
2009
6 cities 10 cities rest of Poland Poland
owned by municipalities owned by housing cooperativesowned by state owned enterprices owned by private individualsTBS (social housing cooperatives) State owned
*/in 2009 the Central Statistical Office (GUS) changed the definition of ownership; since 2007 statistical data are gathered every two years; the 2011 data will be available in 2012. Source: NBP based on GUS data.
6 The Act of 16 September 2011 (Journal of Laws No. 232, item 1377) on the protection of rights of a housing unit or a single-family house buyer, defines the real estate development contract and obliges the entrepreneur to provide appropriate protection to the buyer. 7 Whenever the charts refers to the 6 cities we mean: Gdańsk, Łódź, Cracow, Poznań, Warsaw and Wrocław, whereas the reference to the 7 cities concerns the above mentioned cities and Gdynia, whereas the group of 10 cities includes: Białystok, Bydgoszcz, Katowice, Kielce, Lublin, Olsztyn, Opole, Rzeszów, Szczecin, Zielona Góra.
12
The steady growth in housing stock is the effect of a steady expansion of housing construction (Figure 2). The price boom observed in the years 2006-2008 led to a rapid increase in the housing stock, followed by its levelling-off and a slight rise in the subsequent years, also in 2011 (see Figure 3). As a result of a strong economic growth and slowing growth rate of home prices after 2007, the growth of the value of the housing stock was outpaced by GDP growth, and, as a result, its ratio to the GDP slid to approx. 180% (see Figure 4). The share of housing construction in total construction reached approx. 13% in 2011 and similarly to the whole construction sector, was subject to strong cyclical fluctuations, including in relation to the GDP. In 2011, the contribution of housing construction expenditure to GDP, as published by the Central Statistical Office (CSO) was low (single-family and multi-family housing construction of approx. 1.8% of GDP, including real estate development construction of approx. 0.9%) and close to the one recorded a year ago (see Figure 5). Yet, a rise was observed in the share of civil engineering construction, being the result, among other things, of programmes financed with EU funds, including, the Euro 2012 Championship projects (it accounted for approx. 8% and outstripped the one recorded in the previous year by approx. 1 percentage point).
Figure 2 Housing stock in Poland7 (in millions of square meters)
0
100
200
300
400
500
600
700
800
900
1000
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011*
6 voivodeship capitals 10 voivodeship capitals rest of Poland
Figure 3 Housing assets in Poland7 (in PLN billion)
0
500
1 000
1 500
2 000
2 500
3 000
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011*
6 voivodeship capitals 10 voivodeship capitals rest of Poland
*/ Figures 2-4 present own estimates for 2011. Source: GUS, PONT Info, Sekocenbud, NBP.
Source: GUS, PONT Info, Sekocenbud, NBP.
Figure 4 Housing assets in Poland7 in relation to GDP
0%
50%
100%
150%
200%
250%
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011*
6 voivodeship capitals 10 voivodeship capitals rest of Poland
Figure 5 Investment expenditure on construction in Poland in relation to GDP
1%2%3%4%5%6%7%8%9%
10%11%12%13%14%
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011*
civil engineering output non-housing output housing output
Source: GUS, PONT Info, Sekocenbud, NBP. Source: GUS.
13
Analysis of the housing situation conducted for many years by international institutions 8 show that this situation has been improving along with GDP growth (see Figure 6). However, when accounting for the impact of the last housing market boom, the regression shows that the previously observed trend had weakened and the correlation has reversed (see Figure 7). This has been driven by housing booms (rapid growth in housing stock) and simultaneous slump in GDP. Poland’s housing situation, albeit weaker than the EU average, reflects the level of Poland’s economic development (see Figure 8).
Considerable deviations from the above mentioned trends observed in particular countries, may be largely attributed to domestic housing policies, often fuelling the sector’s development at the expense of the rest of the economy.
The most favourable housing situation in Poland may be observed in the largest cities enjoying, in the past few years, a rise in housing quality as measured, for example, by the number of housing units in the housing stock per 1000 inhabitants or by the number of square meters of a dwelling unit in the housing stock per one person9 (see Figure 9 -Figure 11). The large mean usable area of housing in the countryside is related to single-family house construction, dominating this market.
Figure 6 Housing stock per 1000 inhabitants and GDP per capita (2005) in the EU Member States
R² = 0,4127
0
5
10
15
20
25
30
35
40
45
300 350 400 450 500 550 600
GD
P/c
(in
tho
usa
nd
s U
SD
, P
PP
)
Housing stoc k per 1000 inhabitants
PL
CHDK
NL
E
P
EESLO
Figure 7 Housing stock per 1000 inhabitants and GDP per capita (2010) in the EU Member States
AT
BE DKDE
ESFR IT
NL
PT
FI
SE
UKIE
EL
LU
CYCZ
EELT
LT
MTPL
SKSI
HU
R² = 0,0414
10
20
30
40
50
60
70
80
90
300 350 400 450 500 550 600
GD
P (
in t
ho
usa
nd
s U
SD
, PP
P)
Housing stock per 1000 inhabitants
Source: Hypostat. Source: Hypostat.
8 E.g. Hypostat, Eurostat. 9 This is discussed in more detail in Annex no. 1 Trends in local market development – comparative analysis of 16 cities in Poland and in Part III Monographic study of 16 cities in Poland.
14
Figure 8 Number of housing units per 1000 inhabitants (2010) in Europe
0
100
200
300
400
500
600
Figure 9 Housing stock per 1000 inhabitants in Poland7
300
350
400
450
500
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011*
hous
ing/
1000
peo
ple
6 cities Warsaw 10 cities villages Poland
*/Western Europe – selected countries, i.e. Austria, Denmark, Finland, France, Italy, the Netherlands and Spain; **/The number of housing units in Bulgaria is largely overstated as also recreational homes are classified as housing units; Source: Hypostat.
*/ Figures 9-11 present own estimates for 2011. Source: GUS.
Figure 10 Average usable area of housing in the housing stock in Poland7
55,0
60,0
65,0
70,0
75,0
80,0
85,0
90,0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011*
sq m
/ ho
usin
g
6 cities Warsaw 10 cities villages Poland
Figure 11 Average usable area of housing per person in the housing stock in Poland 7
20,0
21,0
22,0
23,0
24,0
25,0
26,0
27,0
28,0
29,0
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011*
sq. m
/ pe
rson
6 cities Warsaw 10 cities villages Poland Source: GUS. Source: GUS.
Situation in the residential market in Poland The trend of long-term reductions in home prices, both in the primary and secondary
market that started in 2009, continued in 2011. This phenomenon was observed in Poland’s major residential markets which experienced only slight declines in nominal prices (see Figure 12-Figure 15) and faster declines in real prices (see Figure 19-Figure 20), driven by rising inflation.
The analysis of offer prices (Figure 12) shows that during the boom period, prices in Warsaw were rising faster in the secondary market than in the primary market. During the downturn, the primary market showed a higher flexibility and prices fell, whereas owners of secondary market housing units tried to maintain prices at a high level. Concerning transaction prices (Figure 13), Warsaw was the only market where prices in the secondary market considerably exceeded those in the primary market. In other cities those prices were similar or the secondary market prices were lower. This can be explained by the considerable difference in the quality of newly constructed housing as compared to the housing stock quality in those cities.
In the 10 cities, which were less affected by the lending boom, the difference between the transaction and offer prices, in the primary and secondary market, is rather stable (see Figure 14). On the other hand, in Warsaw and in the seven cities, the level of offer prices in the
15
primary market during the boom period always exceeded the level of transaction prices. This resulted from excessive expectations of real estate developers. Yet, since 2010 real estate developers have made attempts to adjust the prices. This was also the case in 2011.
In the secondary market (see Figure 15) home sellers have always expected higher price that the achievable one.
The analysis of hedonic prices, reflecting „pure” price changes i.e. resulting from other factors than differences in housing quality, shows that prices in the markets remain stable, and the main factor behind short-term fluctuations is the change in the structure of the analysed sample (e.g. bigger or smaller number of housing units gathered in the base) and the market offer (e.g. bigger or smaller number of more expensive dwellings) (see Figure 16 – Figure 18). Since 2009 this index in the case of Warsaw, Cracow, Wrocław and Gdańsk was stabilizing, while in the case of Poznań it fluctuated a little. This is connected, among other things, with larger amount of comparable data on the former cities. We had a smaller amount of data on Poznań in 2010, thus the index was impacted by transactions deviating from the average.
Figure 12 Weighted average10 price of 1 square meter of housing, offers, primary market and secondary market
20003000400050006000700080009000
1000011000
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
PLN
/ m
2
10 cities offer PM 7 cities offer PM Warsaw offer PM
10 cities offer SM 7 cities offer SM Warsaw offer SM
Figure 13 Weighted average10 price of 1 square meter of housing, transactions, primary market and secondary market
2000
3000
4000
5000
6000
7000
8000
9000
10000
20
06
Q3
20
06
Q4
20
07
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
08
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
09
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
10
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
11
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
12
Q1
PLN
/ m
2
10 cities transaction PM 7 cities transaction PM Warsaw transaction PM
10 cities transaction SM 7 cities transaction SM Warsaw transaction SM
Source: NBP. Source: NBP.
Figure 14 Weighted average10 price of 1 square meter of housing, offers versus transactions, primary market
2000
3000
4000
5000
6000
7000
8000
9000
10000
11000
200
6 Q
320
06
Q4
200
7 Q
120
07
Q2
200
7 Q
320
07
Q4
200
8 Q
120
08
Q2
200
8 Q
320
08
Q4
200
9 Q
120
09
Q2
200
9 Q
320
09
Q4
201
0 Q
120
10
Q2
201
0 Q
3
201
0 Q
420
11
Q1
201
1 Q
220
11
Q3
201
1 Q
420
12
Q1
PLN
/ m
2
10 cities offer PM 7 cities offer PM Warsaw offer PM
10 cities transaction PM 7 cities transaction PM Warsaw transaction PM
Figure 15 Weighted average10 price of 1 square meter of housing, offers and transactions, secondary market
2000
3000
4000
5000
6000
7000
8000
9000
10000
11000
200
6 Q
3
200
6 Q
4
200
7 Q
1
200
7 Q
2
200
7 Q
3
200
7 Q
4
200
8 Q
120
08
Q2
200
8 Q
3
200
8 Q
4
200
9 Q
1
200
9 Q
2
200
9 Q
3
200
9 Q
420
10
Q1
201
0 Q
2
201
0 Q
3
201
0 Q
4
201
1 Q
1
201
1 Q
2
201
1 Q
320
11
Q4
201
2 Q
1
PLN
/ m
2
10 cities offer SM 7 cities offer SM Warsaw offer SM
10 cities transaction SM 7 cities transaction SM Warsaw transaction SM
Source: NBP. Source: NBP.
10 Average weighted with the share of particular city’s housing stock in the analysed sample of cities.
16
Figure 16 Weighted average10 price of 1 square meter of housing versus hedonic price – secondary market, transactions
2000
3000
4000
5000
6000
7000
8000
9000
10000
200
6 Q
32
006
Q4
200
7 Q
12
007
Q2
200
7 Q
32
007
Q4
200
8 Q
12
008
Q2
200
8 Q
32
008
Q4
200
9 Q
12
009
Q2
200
9 Q
32
009
Q4
201
0 Q
12
010
Q2
201
0 Q
32
010
Q4
201
1 Q
12
011
Q2
201
1 Q
32
011
Q4
201
2 Q
1
PLN
/m2
Warszaw 7 cities 10 cities
Warszawa hed. 7 cities hed. 10 cities hed.
Source: NBP.
Figure 17. Price of 1 square meter of housing and hedonic-adjusted price – secondary market, part a
4 000
6 000
8 000
10 000
200
6 Q
32
006
Q4
200
7 Q
12
007
Q2
200
7 Q
32
007
Q4
200
8 Q
12
008
Q2
200
8 Q
32
008
Q4
200
9 Q
12
009
Q2
200
9 Q
32
009
Q4
201
0 Q
12
010
Q2
201
0 Q
32
010
Q4
201
1 Q
12
011
Q2
201
1 Q
32
011
Q4
201
2 Q
1
PLN
/ m
2
Kraków hed. Warszawa hed. Wrocław hed.
Kraków Warszawa Wrocław
Figure 18. Price of 1 square meter of housing and hedonic-adjusted price – secondary market, part b
1 000
3 000
5 000
7 0002
006
Q3
200
6 Q
42
007
Q1
200
7 Q
22
007
Q3
200
7 Q
42
008
Q1
200
8 Q
22
008
Q3
200
8 Q
42
009
Q1
200
9 Q
22
009
Q3
200
9 Q
42
010
Q1
201
0 Q
22
010
Q3
201
0 Q
42
011
Q1
201
1 Q
22
011
Q3
201
1 Q
42
012
Q1
PLN
/ m
2
Poznań hed. Łódź hed. Gdańsk hed.Poznań Łódź Gdańsk
Source: NBP. Source: NBP.
Figure 19 Index of average weighted 10price of 1 square meter of housing and real price to CPI (2006 Q3 =100) - primary market, transactions
90100110120130140150160170180
200
6 Q
32
006
Q4
200
7 Q
12
007
Q2
200
7 Q
32
007
Q4
200
8 Q
12
008
Q2
200
8 Q
32
008
Q4
200
9 Q
12
009
Q2
200
9 Q
32
009
Q4
201
0 Q
12
010
Q2
201
0 Q
32
010
Q4
201
1 Q
12
011
Q2
201
1 Q
32
011
Q4
201
2 Q
1
10 cities BaRN 10 cities BaRN defl. by CPI7 cities BaRN 7 cities BaRN defl. by CPIWarsaw BaRN Warsaw BaRN defl. by CPI
Figure 20 Index of average weighted 10price of 1 square meter of housing and real price to CPI (2006 Q3 =100) - secondary market, transactions
90
110
130
150
170
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
10 cities BaRN 10 cities BaRN defl. by CPI7 cities BaRN 7 cities BaRN defl. by CPIWarsaw BaRN Warsaw BaRN defl. by CPI
Source: NBP. Source: NBP.
Nominal prices in the primary and secondary market in Warsaw and in the seven cities
continued a downward trend in 2011 (see Figure 19 and Figure 20). In the case of the ten
17
cities, nominal prices were almost stable, which may be attributed to a very small number of transactions. Real prices, especially in the case of Warsaw and the seven cities, are indicative of prices returning to the level observed before the boom.
Price falls mitigate tensions in the housing market, fuelled by excessively high prices being the result of the lending boom. This is indicated by the housing price-to-income ratio (P/I), which is the most measurable and applicable index for the Polish housing market (see Figure 21). In 2011 this ratio went down considerably, yet failed to reach the pre-boom level which might be considered as the long-term equilibrium level11.
Rental payments resulting from sublease were subject to considerable fluctuations in particular quarters, however their general trend levelled off (see Figure 22). Their level was considerably affected by the shift in the financing structure of purchased housing. The departure from the dominant share of foreign currency denominated loans to basically only PLN loans at the end of 2011, dramatically reduced the value of housing12 as compared to the alternative rental option (see Figure 23 and Figure 24). The considerably cheaper rental option as compared to home purchases, mitigated the pressure to purchase new real estate developer housing and pushed up the demand for rental housing. This might have added to further falls in the prices of owner-occupied housing purchased from real estate developers and growth in market rents. These trends continued in 2011.
Figure 21. Ratio of price of 1 square meter of housing to income (P/I) (in years)
2
3
4
5
6
7
8
9
10
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Warsaw Kraków GdanskWrocław Poznań Łódź
Figure 22. Price of 1 square meter of rented housing (average of lease transaction prices)
15,0
20,0
25,0
30,0
35,0
40,0
45,0
50,0
55,0
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
PL
N /
m2
Warsaw Kraków Gdańsk Wrocław
Poznań Łódź średnia
Source: GUS, NBP. Source: NBP.
11 For example, for the Warsaw market, the P/I index in 2011 is approx. 6 years as compared to 4 years in 2002. 12 These values are calculated as rents discounted with interest rate on housing loans.
18
Figure 23. Relation of interest costs per 1 square meter of housing (PLN loan) to rents
0,20,40,60,81,01,21,41,61,82,02,2
200
5 Q
3
200
5 Q
4
200
6 Q
1
200
6 Q
2
200
6 Q
3
200
6 Q
4
200
7 Q
1
200
7 Q
2
200
7 Q
3
200
7 Q
4
200
8 Q
1
200
8 Q
2
200
8 Q
3
200
8 Q
4
200
9 Q
1
200
9 Q
2
200
9 Q
3
200
9 Q
4
Warsaw Kraków GdańskWrocław Poznań Łódź
Figure 24 Relation of interest costs per 1 square meter of housing (CHF loan) to rents
0,20,40,60,81,01,21,41,61,82,02,2
200
5 Q
3
200
5 Q
4
200
6 Q
1
200
6 Q
2
200
6 Q
3
200
6 Q
4
200
7 Q
1
200
7 Q
2
200
7 Q
3
200
7 Q
4
200
8 Q
1
200
8 Q
2
200
8 Q
3
200
8 Q
4
200
9 Q
1
200
9 Q
2
200
9 Q
3
200
9 Q
4
Warsaw Kraków GdańskWrocław Poznań Łódź
Source: GUS, NBP. Source: GUS, NBP.
Availability of an average mortgage-financing housing13, purchased by an average household, following the 2007-2008 slump, rebounded steadily. After 2008, the fall in home prices accelerated and, despite foreign currency denominated lending almost coming to an end, at the end of 2011 the availability of the mortgage-financed housing #14 (see Figure 25-Figure 28) continued on a slight upwards trend, ensuring a rather sustainable demand. Yet, these market adjustments led to fluctuations in home prices.
Figure 25 Availability of the mortgage-financed housing in square meters (PLN loans)
20
40
60
80
100
120
140
160
180
200
4 Q
12
004
Q2
200
4 Q
32
004
Q4
200
5 Q
12
005
Q2
200
5 Q
32
005
Q4
200
6 Q
12
006
Q2
200
6 Q
32
006
Q4
200
7 Q
12
007
Q2
200
7 Q
32
007
Q4
200
8 Q
12
008
Q2
200
8 Q
32
008
Q4
200
9 Q
12
009
Q2
200
9 Q
32
009
Q4
201
0 Q
12
010
Q2
201
0 Q
32
010
Q4
201
1 Q
12
011
Q2
201
1 Q
32
011
Q4
201
2 Q
1
Warsaw Kraków GdańskWrocław Poznań Łódź
Figure 26 Availability of the mortgage-financed housing in square meters (CHF loans)
20
40
60
80
100
120
140
160
180
200
4 Q
12
004
Q2
200
4 Q
32
004
Q4
200
5 Q
12
005
Q2
200
5 Q
32
005
Q4
200
6 Q
12
006
Q2
200
6 Q
32
006
Q4
200
7 Q
12
007
Q2
200
7 Q
32
007
Q4
200
8 Q
12
008
Q2
200
8 Q
32
008
Q4
200
9 Q
12
009
Q2
200
9 Q
32
009
Q4
201
0 Q
12
010
Q2
201
0 Q
32
010
Q4
201
1 Q
12
011
Q2
201
1 Q
32
011
Q4
201
2 Q
1
Warsaw Kraków GdańskWrocław Poznań Łódź
Source: GUS, PONT Info. Source: GUS, PONT Info.
13 Average housing availability for a particular city is the quotient of an average wage in the enterprise sector in this city and a price per 1 square meter of an average housing on sale in this city. 14 Given the low quality of information on real housing demand, due, in the first place, to the absence of reliable statistical data on the number of purchased homes, the availability of loan-financed housing may serve as its approximated size.
19
Figure 27 Availability of the mortgage-financed housing in square meters (EUR loans)
20
40
60
80
100
120
140
160
180
20
04
Q1
20
04
Q2
20
04
Q3
20
04
Q4
20
05
Q1
20
05
Q2
20
05
Q3
20
05
Q4
20
06
Q1
20
06
Q2
20
06
Q3
20
06
Q4
20
07
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
08
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
09
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
10
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
11
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
12
Q1
Warsaw Kraków GdańskWrocław Poznań Łódź
Figure 28 Availability of mortgage-financed housing in square meters (weighted loans)
20
40
60
80
100
120
140
160
180
200
4 Q
12
004
Q2
200
4 Q
32
004
Q4
200
5 Q
12
005
Q2
200
5 Q
32
005
Q4
200
6 Q
12
006
Q2
200
6 Q
32
006
Q4
200
7 Q
12
007
Q2
200
7 Q
32
007
Q4
200
8 Q
12
008
Q2
200
8 Q
32
008
Q4
200
9 Q
12
009
Q2
200
9 Q
32
009
Q4
201
0 Q
12
010
Q2
201
0 Q
32
010
Q4
201
1 Q
12
011
Q2
201
1 Q
32
011
Q4
201
2 Q
1
Warsaw Kraków GdańskWrocław Poznań Łódź
Source: GUS, PONT Info. Note: Loan weighted with the currency structure of quarterly rise in housing loans for individuals; Source: GUS, PONT Info.
Residential sector environment The development of the market residential sector considerably affects the economy
through investment expenditures, the banking sector and the households. The strongest impact is exerted on the banking sector and the relations with the global financial market. The world experience is strongly negative. Financing of the residential and commercial real estate sector with short-term international market instruments has always led to real estate crises and disintegration of the banking system, no matter how and where the exchange rate risk was present.
In Poland, the rather small vulnerability of the GDP to the real estate development risk results from a relatively minor share of housing construction in economic growth. The main channel through which the sector impacts GDP is the financial sector and household expenditures on housing.
Figure 29 Relations of housing loans in Poland
0%
50%
100%
150%
200%
250%
300%
0%
5%
10%
15%
20%
25%
30%
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
In relation to bank assets In relation to GDP
In relation to housing wealth In relation to banks' equity (R ax)
Figure 30 Structure of the housing loan debt of individuals at current FX rates
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
1996
Q4
1997
Q4
1998
Q4
1999
Q4
2000
Q4
2001
Q4
2002
Q4
2003
Q4
2004
Q4
2005
Q4
2006
Q4
2007
Q4
2008
Q4
2009
Q4
2010
Q4
2011
Q4
PLN
FX
Source: GUS, PONT Info, Sekocenbud and NBP. Source: NBP.
The consequence of expanding credit portfolios is their growing impact on banks’ security. This impact is visible while comparing basic cross-sectoral relations. Although the
20
housing loans to housing stock and GDP ratio has all the time continued below the EU average, this is almost three times the value of banks’ equity (see Figure 29).
The borrower population is concentrated in the largest cities. Since 2006 there has been a steady rise in the share of housing loan repayments in household budgets, reaching almost 20% in 2011 (see Figure 31). At the same time, throughout the whole analysed period, there was a steady rise, in nominal terms, in the income of households repaying housing loans. This means that the amounts of loans granted grew bigger and bigger (see Figure 32). Since 2008 also the share of household expenditure on home use has been growing, exceeding 20% in the seven largest cities 15 and 18% in the cities whose population ranges from 200 to 500 thousand inhabitants# (see Figure 33). Starting from 2005, this expenditure has been on a rise in nominal terms (see Figure 34) and has already reached 40% of household budgets. In the future, this factor may have a significant effect on the quality of mortgage portfolios as such a large share of fixed expenditure makes these households vulnerable to labour market problems.
Figure 31 Share of housing loan repayment in disposable income in Poland #
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
20%
2005 2006 2007 2008 2009 2010 2011
7 cities (500+) cities (200-500)
Figure 32 Disposable income of Polish households # repaying housing loans
0
1 000
2 000
3 000
4 000
5 000
6 000
7 000
8 000
2005 2006 2007 2008 2009 2010 2011
PL
N
7 cities (500+) cities (200-500)
Note: figures 31-34 concern two groups of cities: cities with 200-500 thousand inhabitants and cities with over 500 thousand inhabitants Source: GUS.
Source: GUS.
Figure 33 Home use expenses in disposable income in Poland #
15%
17%
18%
20%
21%
2005 2006 2007 2008 2009 2010 2011
7 cities (500+) cities (200-500)
Figure 34 Home use expenses of households in Poland #
0
100
200
300
400
500
600
700
2005 2006 2007 2008 2009 2010 2011
PLN
7 cities (500+) cities (200-500)
Source: GUS. Source: GUS.
15 Cities with more than 500 inhabitants
21
3. Relations between the financial sector and the residential real estate sector
The credit system, starting from the beginning of the new millennium, has been quite flexible, allowing the transfer of home ownership among economic entities. Still, there is no institutionalized or universal system of subsidized housing, providing assistance to people who have lost their loan repayment capacity and face the risk of losing their homes (mortgage foreclosure). As the legislator shifts these problems onto the lender, it boosts the housing loan risk faced by banks, and if the scale of the problems will increase, this may be an additional element of systemic risk. One of the possible solutions, as the experience of Germany shows, is the establishment of a social housing system. Social dwellings may be a short-term accommodation for economically weaker borrowers.
Another problem, also faced by other countries, is the creation of shocks by the financial sector, likely to accelerate cycles in the real estate sector. This may, in turn, trigger a real estate crisis. Liberalization of capital movements in Poland deregulated the relations between credit price, home price and household income. In 2011, as in the previous years, there was some progress in restoring the right proportions, yet the course of the process is hardly predictable. International experience shows that the market price of housing should not exceed the four-year average of household income and the interest expense should not be higher than rents for comparable housing. Further on, the rate of return in the real estate development sector adjusted for the risk premium should not drastically differ from rates of return in the whole economy.
Growing disparities within the financial sector, threatening the further, sustainable development still remain unsolved. The sector growth relied on large universal banks, the leading players in this market. An attempt to build a system based on the model of a central mortgage bank refinancing other banks via the capital market (like the Federal Home Loan Bank model), failed. Similarly, the dispersed model of specialized institution, fed by the capital market (like the German Hypothekenbank model) also failed. As a result, Poland has in place a deposit-based system of real estate sector financing. With a large share of mortgage portfolios, this raises the risk for portfolio financing. Currently, the bank risk is shifted onto the public sector thorough deposit guarantees. This creates moral hazard both for the banking sector and depositors16. The public sector risk should be limited by prudential regulations and instruments (e.g. mortgage bonds, as in the German system). Yet, reconstruction of the portfolio financing structure would entail higher costs, which taking into consideration low rate of return on PLN loans, must be spread in time. Changes are possible almost only for newly granted loans in the case of which there are no limitations as regards interest rate levels and methods how they are set.
Another problem in the system structure is the high share of foreign currency denominated loans (the effect of the credit boom) driven by interest rate disparity. Under present conditions, foreign currency denominated loans add to the banks’ risk. Therefore, foreign
16 Banks expect that in the case of problems they will be granted assistance by the public sector which should limit the collapse of the financial sector (as in Ireland). On the other hand, person making a deposits places his/her funds with a bank at the highest deposit rate, notwithstanding his/her awareness of running higher risk.
22
currency lending should be drastically limited so that its balance does not build up with the time.
Another problem is gradual fall in the quality of housing loan portfolio, especially as regards the 2007-2008 loans and the 2009-2010 loans17. This is a natural process, yet the scale of portfolio deterioration, especially as the above mentioned loan generations are concerned, raises serious concerns.
Until 2011, the investment system relied on the banking sector to a small extent only. The role of the banking sector was, in fact, served by real estate developers who took households’ funds and invested them in real estate. As a result, home buyers signing a home construction contract assumed the risk themselves, being neither able to assess or manage it. The Act on the protection of home buyer rights which were introduced in 2011 and entered into force in April 2012 might change the situation. Yet, it should be remembered that this act will increase the risk for construction companies.
Liabilities of the Polish banking system are dominated by current deposits (see Figure 35), whereas its assets are dominated by long-term receivables (see Figure 36). In accordance with the data published by the Polish Financial Supervision Authority (UKNF) banks’ assets in excess of 5 years account for 34.5% of the balance sheet total, whereas liabilities in excess of 5 years account for 5.3%. Such a mismatch is a risk factor for the banking sector.
Figure 35 Structure of deposits in the banking system
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
XII
20
02
VI
20
03
XII
20
03
VI
20
04
XII
20
04
VI
20
05
XII
20
05
VI
20
06
XII
20
06
VI
20
07
XII
20
07
VI
20
08
XII
20
08
VI
20
09
XII
20
09
VI
20
10
XII
20
10
VI
20
11
XII
20
11
more than 2years
1 - 2 years
less than 1year
Currentdeposits
Figure 36 Structure of loans in the banking system
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
XII2002
XII2003
XII2004
XII2005
XII2006
XII2007
XII2008
XII2009
XII2010
XII2011
remaining loans fornoncommercialinstitution
remaining loans forenterpreneuers
remaining loans forhouseholds
housing loans fornoncommercialinstitution
housing loansforenterpreneuers
housing loans forhouseholds
Source: NBP. Source: NBP
In 2011 interest rates on PLN housing loans continued on an upward trend, reaching approx. 7% at the end of 2012. Interest rates of foreign currency denominated loans were lower, yet, practically inaccessible for households.
A more detailed analysis of interest rates, especially CPI-deflated or Disposable Income-deflated interest rates, shows that real interest have fallen, and are even negative in the case of foreign currency denominated loans.
17 In accordance with the “Report on banks situation in 2011” of the Polish Financial Supervision Authority (UKNF) (2012).
23
Figure 37 Interest rates in the financial market in Poland
2,0%
3,0%
4,0%
5,0%
6,0%
7,0%
200
5 Q
12
005
Q2
200
5 Q
32
005
Q4
200
6 Q
12
006
Q2
200
6 Q
32
006
Q4
200
7 Q
12
007
Q2
200
7 Q
32
007
Q4
200
8 Q
12
008
Q2
200
8 Q
32
008
Q4
200
9 Q
12
009
Q2
200
9 Q
32
009
Q4
201
0 Q
12
010
Q2
201
0 Q
32
010
Q4
201
1 Q
12
011
Q2
201
1 Q
32
011
Q4
201
2 Q
1
interest rate on deposits in PLN (volume)interest rate on deposits in PLN (new)interest rate on bonds (10 years)POLONIAWIBOR
Figure 38 Real interest rates on household deposits in Poland
-3,0%
-2,0%
-1,0%
0,0%
1,0%
2,0%
3,0%
4,0%
5,0%
6,0%
2005
Q1
2005
Q2
2005
Q3
2005
Q4
2006
Q1
2006
Q2
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
less than 1 m. 3 -6 m. more than 1 m. in general
Source: NBP. Note: CPI-deflated after initial withholding of capital gains tax Source: NBP.
Figure 39 Interest rates on housing loans to households (GD) in Poland
1,0%
2,0%
3,0%
4,0%
5,0%
6,0%
7,0%
8,0%
9,0%
2004
Q1
2004
Q2
2004
Q3
2004
Q4
2005
Q1
2005
Q2
2005
Q3
2005
Q4
2006
Q1
2006
Q2
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
PLN new CHF new EUR new
PLN (stock) CHF (stock)
Figure 40 CPI-deflated interest rates on housing loans or DI#
-4,0%-3,0%-2,0%-1,0%0,0%1,0%2,0%3,0%4,0%5,0%6,0%7,0%8,0%9,0%
2004 2005 2006 2007 2008 2009 2010 2011
%PLN %PLN defl.CPI %PLN defl.DI%CHF %CHF defl.CPI %CHF defl.DI
Source: NBP. Source: NBP.
Disbursement of housing loans and related increase in Poland’s debt was connected with structural development of the market, business cycles and consequences of the global credit boom. The share of foreign currency denominated loans in the portfolio of housing loan receivable due from individuals has remained at a similar level since 2010; at the end of 2012 Q4 it reached 60% (see Figure 41- Figure 42). Due to the negative experience related to the global financial crisis, as well as regulatory measures, the share of foreign currency denominated loans in quarterly increase in newly granted housing loans declined from the record 90% in 2008 Q3 to 20% in 2010 Q4 to almost disappear at the end of 2011. At the beginning of 2012 a decline of foreign currency denominated loans was recorded.
24
Figure 41 Currency structure of growth in housing loan debt of individuals in Poland (under constant foreign exchange rates)
-2 000-1 000
01 0002 0003 0004 0005 0006 0007 0008 0009 000
10 00011 00012 00013 000
200
4 Q
120
04
Q2
200
4 Q
320
04
Q4
200
5 Q
120
05
Q2
200
5 Q
320
05
Q4
200
6 Q
120
06
Q2
200
6 Q
320
06
Q4
200
7 Q
120
07
Q2
200
7 Q
320
07
Q4
200
8 Q
120
08
Q2
200
8 Q
320
08
Q4
200
9 Q
120
09
Q2
200
9 Q
320
09
Q4
201
0 Q
120
10
Q2
201
0 Q
320
10
Q4
201
1 Q
120
11
Q2
201
1 Q
320
11
Q4
201
2 Q
1
bln
PLN
PLN foreign currency with exchange rate adjustment
Figure 42 Currency structure of growth in housing loan debt of individuals in Poland (under constant foreign exchange rates)
90%
91%
55%
-7% 20
% 28%
74%
22% 36
%
62%
8%
74%
50%
46%
-61%
10%
9%
45%
107%
80% 72%
26%
78% 64
%
38%
92%
26%
50%
54%
161%
-30%
-20%
-10%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
199
8
199
9
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
201
2 Q
1
PLN FX
Source: NBP. Source: NBP.
In 2011, the nominal annual rise in receivables from individuals was close to the one seen in 2010. Yet, at constant FX rates, the recorded increase was lower (see Figure 43). 2012 Q1 brought a decline in the loan volume which may be indicative of on-going repayments, yet moderate increase in lending growth at constant FX rates results from Polish zloty depreciation in this period. The value of housing loans in 2011 both according to the Polish Financial Supervision Authority as according to the BIK data was similar to the previous year’s value (see Figure 44). The geographical structure of lending did not change significantly – the bulk of lending was recorded in the Warsaw market (see Figure 45 – Figure 46).
Figure 43 Changes in the value of housing loan debt of individuals in Poland (PLN billion)
-10
0
10
20
30
40
50
60
70
80
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
Q1
PLN
mill
ion
change in nominal terms
adjusted for foreign exchange rates
Figure 44 Value of newly disbursed housing loans, net of interest (broken down into place of loan granting)
0
5
10
15
20
25
30
35
2005 2006 2007 2008 2009 2010 2011 2012 Q1
PLN
bill
ion
6 cities 10 cities rest of Poland
Source: NBP. Source: NBP based on BIK data.
25
Figure 45 Geographical structure of housing loans in Poland’s 16 cities
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2005 2006 2007 2008 2009 2010 2011 I kw.2012
Warszawa 5 cities 10 cities
Figure 46 Geographical structure of housing loans in Poland’s 6 cities
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2005 2006 2007 2008 2009 2010 2011 2012 Q1
Warszawa Kraków Wrocław Poznań Gdańsk Łódź
Source: NBP based on BIK data. Source: NBP based on BIK data.
Housing loan risk faced by banks The analysis of the breakdown of foreign exchange risk premium and credit risk premium
(see Figures 47 – Figure 49) show the bank’s complex attitude to risk.
Figure 47 Risk related to new housing loans # in banks’ assessment
-0,5%
0,0%
0,5%
1,0%
1,5%
2,0%
2,5%
3,0%
3,5%
200
4 Q
120
04
Q2
200
4 Q
320
04
Q4
200
5 Q
120
05
Q2
200
5 Q
320
05
Q4
200
6 Q
120
06
Q2
200
6 Q
320
06
Q4
200
7 Q
120
07
Q2
200
7 Q
320
07
Q4
200
8 Q
120
08
Q2
200
8 Q
320
08
Q4
200
9 Q
120
09
Q2
200
9 Q
320
09
Q4
201
0 Q
120
10
Q2
201
0 Q
320
10
Q4
201
1 Q
120
11
Q2
201
1 Q
320
11
Q4
201
2 Q
1
loan PLN exchange CHF exchange EUR
Figure 48 Credit risk assessment versus its actual level and the share of outstanding loans #
0,0%
1,0%
2,0%
3,0%
4,0%
5,0%
2005 Q
1
2005 Q
2
2005 Q
3
2005 Q
4
2006 Q
1
2006 Q
2
2006 Q
3
2006 Q
4
2007 Q
1
2007 Q
2
2007 Q
3
2007 Q
4
2008 Q
1
2008 Q
2
2008 Q
3
2008 Q
4
2009 Q
1
2009 Q
2
2009 Q
3
2009 Q
4
2010 Q
1
2010 Q
2
2010 Q
3
2010 Q
4
2011 Q
1
2011 Q
2
bonus for credit risk (credit interest rate - bond) bonus for credit risk (credit interest rate - bank rate)
delays in repayment (1-30 days) delays in repayment (31-90 days)
delays in repayment (91-180 days)
Source: NBP. Source: NBP.
We may distinguish two different exchange rate risk management strategies. During the credit boom period, i.e. until 2008, despite growing risk of the Polish zloty depreciation, risk margins continued on a downward trend. Yet, following radical exchange rate adjustments in the years 2009-2010, when this risk materialized, margins became more reality based and this trend continued in 2011. The negative correlation of interest rates on foreign currency denominated loans and foreign exchange rates (see Figure 49 – Figure 50) was beneficial for the sector.
This kept the costs of housing loans servicing at a relatively low level (see Figure 51 – Figure 52). Yet, in the case of certain foreign currency loan generations, their servicing costs considerably exceeded the level which was assumed at the time when the loan were granted
26
(as measured by the current to the first repayment ratio, which served as the basis for loan calculation). Shocks also affected PLN loans whose servicing costs were on a gradual rise in 2011, however, did not exceed the level of the current to the first repayment ratio. Consequently, the situation in the banking sector has not yet deteriorated considerably. Nevertheless, in the case of drastic and long-lived foreign exchange shocks the situation may be expected to deteriorate.
Figure 49 Foreign exchange risk premium on foreign current denominated housing loans and their actual level
0,0
0,1
0,2
0,3
0,4
0,5
0,6
-1,0%
-0,5%
0,0%
0,5%
1,0%
1,5%
I kw
.20
04
I kw
. 2
005
I kw
. 2
006
I kw
. 2
007
I kw
. 2
008
I kw
. 2
009
I kw
. 2
010
I kw
. 2
011
I kw
. 2
012
curr
ency
ten
sion
s
bo
nus
for
curr
ency
ris
k
bonus for credit risk (EUR) bonus for credit risk (CHF)
bonus for credit risk (weighted average) CHF / PLN (right hand axis)
EUR / PLN (right hand axis) weighted currency*/ /PLN (right hand axis)
Figure 50. Foreign exchange rate and the cost of money
0,0%
1,0%
2,0%
3,0%
4,0%
5,0%
2,0
2,5
3,0
3,5
4,0
4,5
5,0
Inte
rba
nk in
tere
st r
ate
s
PLN
/cu
rre
ncy
PLN / CHF PLN / EURLIBOR3MCHF LIBOR3MEUR
Source: NBP. Source: NBP.
Figure 51. The actual repayment to first repayment ratio for housing loans in PLN
75%
80%
85%
90%
95%
100%
105%
110%
115%
120%
125%
130%
135%
200
6 Q
1
200
6 Q
2
200
6 Q
3
200
6 Q
4
200
7 Q
1
200
7 Q
2
200
7 Q
3
200
7 Q
4
200
8 Q
1
200
8 Q
2
200
8 Q
3
200
8 Q
4
200
9 Q
1
200
9 Q
2
200
9 Q
3
200
9 Q
4
201
0 Q
1
201
0 Q
2
201
0 Q
3
201
0 Q
4
201
1 Q
1
201
1 Q
2
201
1 Q
3
201
1 Q
4
201
2 Q
1
loan agreement entered into in 2006 loan agreement entered into in 2007
loan agreement entered into in 2008 loan agreement entered into in 2009
loan agreement entered into in 2010
Figure 52 The actual repayment to first repayment ratio for housing loans in CHF
75%
80%
85%
90%
95%
100%
105%
110%
115%
120%
125%
130%
135%
200
6 Q
1
200
6 Q
2
200
6 Q
3
200
6 Q
4
200
7 Q
1
200
7 Q
2
200
7 Q
3
200
7 Q
4
200
8 Q
1
200
8 Q
2
200
8 Q
3
200
8 Q
4
200
9 Q
1
200
9 Q
2
200
9 Q
3
200
9 Q
4
201
0 Q
1
201
0 Q
2
201
0 Q
3
201
0 Q
4
201
1 Q
1
201
1 Q
2
201
1 Q
3
201
1 Q
4
201
2 Q
1
loan agreement entered into in 2006 loan agreement entered into in 2007
loan agreement entered into in 2008 loan agreement entered into in 2009
loan agreement entered into in 2010
Note: loans in the amount of PLN 200 thousand with fixed margin plus average quarterly WIBOR 3M plus 20-year depreciation are used in calculations; fixed-instalment repayment plan. Source: NBP.
Note: loans in the amount of PLN 200 thousand as translated into CHF, with fixed margin plus average quarterly LIBORCHF 3M plus 20-year depreciation are used in calculations; fixed-instalment repayment plan. Source: NBP.
27
The quality of the housing loan portfolio18 varied across major cities (see Figure 53-54). Local factors (e.g. unemployment, demographic situation, prices, speculation) also made a strong contribution to macroeconomic shocks. The analysis shows that after the the shocks disappeared, the portfolio quality went back to the previous level. Yet, a more detailed analysis conducted by the Polish Financial Supervision Authority shows a long-term trend of portfolio quality deterioration, which is typical for relatively new mortgage portfolios19. These trends continued in 2011. Yet, larger and longer exchange rate shocks are bound to lead to portfolio quality deterioration below standard levels (2-3%).
Figure 53. 91-180 days delinquent housing loans in Poland’s 16 cities
0
5
10
15
20
25
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
Q1
Q2
Q3
Q4
2005 2006 2007 2008 2009 2010 2011
PLN
mill
ion
Zielona Góra
Wrocław
Warszawa
Szczecin
Rzeszów
Poznań
Opole
Olsztyn
Łódź
Lublin
Kraków
Kielce
Katowice
Gdańsk
Bydgoszcz
Białystok
Figure 54 Quality # of housing loan portfolio in Poland’s 6 cities
-2,0%
-1,0%
0,0%
1,0%
2,0%
3,0%
4,0%
5,0%
6,0%
7,0%
8,0%
9,0%
0,0%
1,0%
2,0%
3,0%
4,0%
5,0%
6,0%
7,0%
8,0%
9,0%
2005 2006 2007 2008 2009 2010 2011 2012
aver
age
shar
es
shar
e o
fsub
stan
dard
/no
n-p
erfo
rmin
g lo
ans
Gdańsk Kraków ŁódźPoznań Warszawa Wrocławweighted average*/
Source: NBP based on BIK data. Source: NBP based on BIK data.
In 2011, the risk of losing a part of collateral as a result of zloty depreciation materialized to a larger extent than in 2010 (see Figure 55 and Figure 56). This is suggested by the analysis of the LTV ratios used to estimate subsequent cohorts of PLN and CHF loans in the Warsaw market. This risk materialized also in the six major cities, for which the LTV ratio, with the most doubtful 2007 and 2008 loans generations, was close to or in excess of 100% as presented in Figure 57.
18 As measured by the share of 91-180 delinquent loans 19 Mortgage loan portfolios show a steady deterioration in quality over 3-8 years, and then a quality improvement over approx. 10-15 years, (see J. Łaszek „Ryzyko kredytowe a bankowe wymogi ostrożnościowe” in „Ryzyko kredytowe wierzytelności hipotecznych, modelowanie i zarządzanie”, edited by K. Jajuga and Z. Krysiak; and „Credit risk, credit scoring, and the performance of home mortgages” R.B. Avery et all. (Federal Reserve Bulletin, July 1996)).
28
Figure 55 LTV for particular generations of housing loans, Warsaw, in the case of PLN loans
20%
40%
60%
80%
100%
120%
140%
2004 2005 2006 2007 2008 2009 2010 2011
purchase in 2004 purchase in 2005purchase in 2006 purchase in 2007purchase in 2008 purchase in 2009purchase in 2010
Figure 56 LTV for particular generations of housing loans, Warsaw, in the case of CHF loans
20%
40%
60%
80%
100%
120%
140%
2004 2005 2006 2007 2008 2009 2010 2011
purchase in 2004 purchase in 2005purchase in 2006 purchase in 2007purchase in 2008 purchase in 2009purchase in 2010
Note: 20-year loan taken at the beginning of the beginning of the year, fixed-instalment repayment plan; initial LTV=80%; balance at the end of the period. Source: NBP based on BIK data.
Note: 20-year loan taken at the beginning of the beginning of the year, fixed-instalment repayment plan; initial LTV=80%; balance at the end of the period. Source: NBP based on BIK data.
Figure 57 LTV (at the end of the year) for the most risky loan generations (2007 and 2008), in the 6 cities in the case of CHF loan
60%
70%
80%
90%
100%
110%
120%
2007 2008 2009 2010 2011
Warszawa 2007 Warszawa 2008 Kraków 2007
Kraków 2008 Gdańsk 2007 Gdańsk 2008
Wrocław 2007 Wrocław 2008 Poznań 2007
Poznań 2008 Łódź 2007 Łódź 2008
Note: 20-year loan taken at the beginning of the beginning of the year, fixed-instalment repayment plan; initial LTV=80%; balance at the end of the period. Source: NBP, GUS.
29
The main factor affecting housing loan supply, apart from capital availability and the regulatory framework, is the loan yield. The estimative profitability of20 housing loan portfolio in 2011, similarly to 2010, remained moderate. Yet, the profitability of foreign currency denominated loans exceeded by far that of PLN loans (see Figure 58 and Figure 59). This explains banks’ pressure on foreign currency lending, despite the huge turmoil in the international financial market and considerable foreign exchange risk, which could have translated into a loan repayment risk. Only as a result of warnings of the Polish Financial Supervision Authority and stronger risk perception by banks, FX lending has almost disappeared.
Although the share of the minimum required equity in the case of FX loans was slightly higher than in the case of PLN loans, the estimated profitability (estimated Return over Equity - ROE) of FX loans has been twice as high. Such a considerable difference is the effect of the adjusted interest margin21 on FX loans being higher that in the case of PLN loans. The ROE is calculated as the ratio of adjusted interest margin on housing loans to the minimum required equity22. In the case of foreign currency denominated loans, the effect of closing the foreign currency position23 is an important income generator as it boost considerably the adjusted interest margin. It should be noted that write-offs as a result of bad loans in the case of foreign currency denominated loans are lower than in the case of PLN loans. The currency exchange is an additional source of banks’ income, both during loan granting as well as during loan repayment (see Figure 60).
20 It should be emphasized that the loan yield is estimative and takes into account financial costs only, with all other operating costs incurred by banks left out. 21 The adjusted interest margin is the result of adding all income and subtracting all financial costs. The effective costs of financing was computed on the basis of 3 month WIBOR and LIBOR by adding estimative costs involved in this operation. Loan write-offs are the result of doubtful loans or depreciating loans. 22 The minimum required equity is estimated on the basis of LTV estimates from AMRON data. The equity requirement for housing loans is set by the Polish Financial Supervision Authority as follows: minimum required equity is 8% of the loan granted. Yet, with the LTV below 50% the required equity in the case of PLN loan is multiplied by 75% (and amounts to 6%). With the LTV above 50%, full required equity in mandatory, namely 8% of the loan granted. The total share of the minimum required equity may be calculated on the basis of the balance of PLN loans and foreign currency denominated loans as well as information about loan distribution according to the LTV range (AMRON SARFIN report) below and above 50%. In the case of PLN loans the required equity ranges from 6.5% to 7% of the credit portfolio, whereas in the case of foreign currency denominated loans, the required equity is very close to 7.5% of the loan granted. 23 The effect of closing the currency position and effective interest rate on loans are taken from the study Financial Stability Report. July 2011. NBP (2011)
30
Figure 58 Estimated yield on housing PLN loans for banks in Poland #
Figure 59 Estimated yield on housing FX loans for banks in Poland #
Note: financial income and costs related to housing loan portfolio; Source: NBP based on AMRON, SARFIN data.
Note: financial income and costs related to housing loan portfolio; Source: NBP based on AMRON, SARFIN data.
Figure 60 Estimated yield on granted and repaid FX loans for banks# in Poland
0,00%
0,05%
0,10%
0,15%
0,20%
0,25%
0,30%
0,35%
0
2 000
4 000
6 000
8 000
10 000
12 000
14 000
20
04
Q1
20
04
Q2
20
04
Q3
20
04
Q4
20
05
Q1
20
05
Q2
20
05
Q3
20
05
Q4
20
06
Q1
20
06
Q2
20
06
Q3
20
06
Q4
20
07
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
08
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
09
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
10
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
11
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
12
Q1
bln
. P
LN
income on FX transactions (repayment of foreign currency loans)income on FX transactions (given foreign currency loans)increase in foreign currency loans (left-hand axis)repayment of foreign currency loans (left-hand axis)
Source: NBP.
-40%
-30%
-20%
-10%
0%
10%
20%
30%
40%
-8%-7%-6%-5%-4%-3%-2%-1%0%1%2%3%4%5%6%7%8%
2008 Q
4
2009 Q
1
2009 Q
2
2009 Q
3
2009 Q
4
2010 Q
1
2010 Q
2
2010 Q
3
2010 Q
4
2011 Q
1
2011 Q
2
2011 Q
3
2011 Q
4
2012 Q
1Loan write-offsEffective cost of loan fundingEffective interest rate on loansAdjusted interest margin on loansROE (right-hand axis)
-20%-10%0%10%20%30%40%50%60%
-3%-2%-1%0%1%2%3%4%5%6%7%
20
08
Q4
20
09
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
10
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
11
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
12
Q1
Results on FX operations of loan granting and repaymentLoan write-offsEffective costs of loan fundingEffective interest rate on loansThe effect of closing the foreign currency positionAdjusted interest margin on loansROE (right-hand axis)
31
Factors affecting demand for housing loans The turmoil in the financial markets and related rises in weighted interest rates on
mortgage loans have translated into declines in weighted availability of mortgage loans to households in 2011 (see: Figure 61 – Figure 62). Interest rates on EUR denominated loans got back to their previous level, whereas interest rates on CHF denominated loans are now lower than before the crisis. As a result, the theoretical availability of foreign currency denominated loans is now lower than before the global crisis. The main factor behind the disappearance of foreign currency denominated loans in 2011 was banks’ withdrawal from this type of financing due to its-inherent risk rather than the change in foreign loan availability.
Figure 61 Available mortgage loans in PLN (in PLN thousand)
70
90
110
130
150
170
190
2004
Q1
2004
Q2
2004
Q3
2004
Q4
2005
Q1
2005
Q2
2005
Q3
2005
Q4
2006
Q1
2006
Q2
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
Warszawa Kraków GdańskWrocław Poznań Łódź
Figure 62 Available weighted mortgage loans 24 (in PLN thousand)
70
90
110
130
150
170
190
200
4 Q
120
04
Q2
200
4 Q
320
04
Q4
200
5 Q
120
05
Q2
200
5 Q
320
05
Q4
200
6 Q
120
06
Q2
200
6 Q
320
06
Q4
200
7 Q
120
07
Q2
200
7 Q
320
07
Q4
200
8 Q
120
08
Q2
200
8 Q
320
08
Q4
200
9 Q
120
09
Q2
200
9 Q
320
09
Q4
201
0 Q
120
10
Q2
201
0 Q
320
10
Q4
201
1 Q
120
11
Q2
201
1 Q
320
11
Q4
201
2 Q
1
Warszawa Kraków GdańskWrocław Poznań Łódź
Source: GUS, PONT Info. Source: GUS, PONT Info.
Investors investing in residential construction in 2011 and financing it with PLN or foreign currency denominated loans, suffered losses as in the previous year. This was driven by declines in home prices and the Polish zloty depreciation. As a result, the investment purchases market waned as a too risky one (see Figure 63 – Figure 66).
24 Loan weighted with the currency structure of the quarterly rise in mortgage loans to individuals. Since 2009, PLN loans have dominated the credit market. Foreign currency loans practically disappeared as of 2011 Q4.
32
Figure 63 Housing investor’s real terms yield (CPI) in particular years, in Warsaw, in the case of PLN loans25
-50%
-30%
-10%
10%
30%
2004 2005 2006 2007 2008 2009 2010 2011
changes in the real value real interest rate outcome
Figure 64 Housing investor’s real terms yield (CPI) in particular years, in Warsaw, in the case of CHF denominated loans25
-50%
-40%
-30%
-20%
-10%
0%
10%
20%
30%
40%
2004 2005 2006 2007 2008 2009 2010 2011
real interest rate currency difference
changes in the real value outcome Note: the investor’s yield is calculated on housing purchased in the previous year. Source: NBP, GUS.
Note: the investor’s yield is calculated on housing purchased in the previous year. Source: NBP, GUS.
Figure 65 Housing investor’s yield, for particular housing loan generations, in Warsaw, in the case of PLN loans25
20%
40%
60%
80%
100%
120%
140%
2004 2005 2006 2007 2008 2009 2010 2011
purchase in 2004 purchase in 2005purchase in 2006 purchase in 2007purchase in 2008 purchase in 2009purchase in 2010
Figure 66 Housing investor’s yield, for particular housing loan generations, in Warsaw, in the case of CHF denominated loans 25
20%
40%
60%
80%
100%
120%
140%
2004 2005 2006 2007 2008 2009 2010 2011
purchase in 2004 purchase in 2005purchase in 2006 purchase in 2007purchase in 2008 purchase in 2009purchase in 2010
Source: NBP, GUS. Source: NBP, GUS.
25 Under the assumption of LTV=80%, 20-year loan, fixed-instalment repayment plan.
33
Box: Analysis of particular elements of risk inherent in residential and commercial real estate investment, including the risk run by the lending bank The analysis highlights selected elements of risk inherent in residential and commercial real estate investment. Moreover, in the case of Poland, also the risk for the lending bank was analysed.
The major part of residential real estate purchases are financed through mortgage credit as housing is an expensive and long-term asset, which only few households can afford to buy with cash. Home purchases may be considered as consumption. The buyer changes liquidity, in the form of loan repayment, into a stream of services coming from possessed housing (see Article by Augustyniak, Łaszek and Olszewski in Annex no. 2). Also commercial real estate, which is treated as investment, is also largely financed with external sources, mostly with loans. This allows for a faster diversification of the investment portfolio, cheaper fundraising than on the stock market and shifting some risk to banks. In fact, commercial real estate loans are advantageous for banks as they bear higher interest and each loan granted is of considerable value, which reduces client service costs. Yet, real estate financing may constitute a serious threat to banks26, if they fail to conduct a thorough and prudent analysis of investors’ collateral and their repayment capacity. Thus, it is important to analyse credit risk related to both types of real estate to enable analysts to compare these risks and their scale for the Polish banking system.
Many investors and researchers erroneously consider commercial real estate as safer than residential real estate. According to Wallace (2010) this may result from incorrect measurement of price fluctuations in the market. In-depth analysis of the components of the U.S. Real Estate Investment Trust (REIT) index for the years 1995-2005 showed that fluctuations in office and retail real estate prices were 50% higher than those in residential real estate prices. It should also be added that the experience of one country may not necessarily be the replicated in another country.
What is important is the analysis of the factors which cause that the risk inherent in commercial real estate investment is higher than the risk related to residential real estate investment as this risk translates into bank’s loan risk exposure.
Woods (2007) quotes Nyberg (2005), who argues that risk related to commercial real estate faced by the bank is higher than risk related to residential real estate. The owner of the building repays interest with rental income, yet, should the number of vacancies rise, his income falls considerably and solvency risk grows. Contrary, home owners generally finance their interest repayments with sources not related to the real estate market. To increase their loan repayment capacity, a home owner also has the possibility to let his housing and rent another, smaller housing instead. Moreover, a home owner bears liability with all his property and income which may provide a stronger incentive for debt repayment. Yet, housing is subject to greater political pressure than commercial real estate.
Healthy relations between investors and banks can hinder investors’ perverse incentives discussed by Herring and Wachter (1999). The investor relies largely on external capital to shift the risk, to the largest possible extent, to the lender and increase the rate on return on equity. At the same time, the lender is not able to thoroughly analyse the investment risk. Moreover, the investor assumes that the lending bank will suffer a great loss if forced to sell the real estate in turbulent times. Thus, the investor may reckon with certain tolerance of the bank should they face problems with loan servicing.
It seems that as a result of all the above mentioned factors the share of doubtful loans will at least be slightly bigger in the case of commercial real estate than in the case of residential real estate.
26 Bank’s risk related to commercial real estate, its origin and possibilities to limit this risk are discussed in more detail by Kucharska-Stasiak (2006), Czerkas (2006) and Jajuga (2006).
34
Analyses of credit risk faced by banks in Poland as a result of mortgage loans There are no clear-cut conclusions following from theoretical analysis of the credit risk, as
residential and commercial markets may differ considerably across countries in terms of the credit risk. Additionally, such analysis makes sense in the long term only (business cycles usually last for 8 years and they occur every 20-30 years). Yet, certain information may be provided by the comparison of the impaired loan ratio for residential and commercial real estate, available in the NBP data base, as it gives the picture of short-term, current risk level.
In order to analyse the risk level, the impaired loan ratio27 for residential and commercial real estate should be compared. Yet, we should emphasise the significant difference in the scale of mortgage loans and commercial loans granted to enterprises for real estate purchases (see Figure R1and R2).
The value of mortgage loans at the end of 2012 Q1 amounted to PLN 311 billion, while that of commercial property loans was PLN 42 billion. This phenomenon may be attributed to dominating share of foreign investors which raise funding abroad on much more favourable conditions, through the stock market or debt securities, thus do not have any recourse to bank lending.
Figure R1 Residential real estate loans (in PLN billion)
0%
5%
10%
15%
20%
25%
30%
0
50
100
150
200
250
300
350
20
09
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
10
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
11
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
12
Q1
Housing loans Impaired housing loan ratio
Figure R2 Structure (in PLN billion) and quality of real estate loans for enterprises
0%
5%
10%
15%
20%
25%
30%
0
10
20
30
40
50
200
8 Q
3
200
8 Q
4
200
9 Q
1
200
9 Q
2
200
9 Q
3
200
9 Q
4
201
0 Q
1
201
0 Q
2
201
0 Q
3
201
0 Q
4
201
1 Q
1
201
1 Q
2
201
1 Q
3
201
1 Q
4
201
2 Q
1
Commercial loans (office space) Housing loans
Other real estate loans Impaired housing loan ratio
Impaired office space loan ratio Impaired other real estate loan ratio Source: NBP. Note: exclusive of BGK
Source: NBP. In 2012 Q 1 the impaired loan ratio accounted for 2.5% of mortgage loans to households. This
ratio seems marginal although it is twice as high as at the beginning of 2009. In the case of office space this ratio is approx. 5%, sometimes even fell below 4%. The situation of other commercial real estate looks a bit bleaker with the impaired loan ratio on a steady rise from the initial 5% to over 8% at the moment. Loans granted to residential real estate development companies run the highest risk; since the outbreak of the global crisis in 2008 the impaired loan ratio has risen to 8% to exceed 20% at present.
It follows from the analysis that banks may run higher risk in the case of commercial real
estate financing than in the case of residential real estate financing. Yet, considering the up-to-date small scale of those loans, the risk run by the banking system seems to be limited. According to the NBP’s report (2012) neither loans for office space purchases nor other real estate loans raise any
27 In accordance with the NBP’s definition (2012), it is expressed as the ratio of impaired loans to the value of loans in the portfolio.
35
concerns. As far as loans granted to residential real estate development companies are concerned, only few banks face credit risk 28and those loans account for a small percentage of loans extended by those banks.
So far, in Poland, the situation of commercial real estate and that of banks financing their purchases seems sound. According to Cyburt (2010), banks generally make a thorough analysis of loan repayment ability with the income generated by the financed commercial real estate, examining the tenant structure and quality. Moreover, banks limited their risk exposure, financing local-scale real estate, with a value usually not exceeding EUR 50 million. Cyburt says that larger investment projects used to be financed by international banks, e.g. German ones, or by real estate funds. Yet, it should be borne in mind that Polish banks used to establish consortia in order to finance larger investment projects. Consequently, their risk exposure may sometimes be significant and may be shifted to the whole financial system.
Yet, the scale of commercial real estate loans, mainly other than real estate development loans
seems to be small. This results from the fact that such real estate may be financed through the stock exchange and with debt securities and the fact that international investors dominate this market. Yet, as Polish investors are growing wealthier and mortgage lending is likely to slow down considerably, growth in office real estate loans and other commercial property loans seems probable. This means that financial supervisors will have to analyse in more depth the risk from the same segment to the whole financial sector.
References Cyburt P. (2010) Rynek nieruchomości komercyjnych zagrożeniem dla ożywienia gospodarczego. Press article in
Rzeczpospolita, 30.10.2010. Czerkas K. (2010). Finansowanie Nieruchomości Komercyjnych w Polsce. Czynniki Ryzyka i Modele Transakcji.
Instytut Rynku Hotelarskiego. Herring R. J. and S. Wachter (1999) Real Estate Booms and Banking Busts: An International Perspective. The
Wharton School - Financial Institutions Center Paper Nr 99-27. Jajuga K. (2006) Ryzyko kredytowe i metoda jego pomiaru. In E. Kucharskiej-Stasiak (Edt.): Ryzyka banku w
zakresie określania wartości nieruchomości dla celów kredytowych w Polsce na tle trendów w Unii Europejskiej. Zeszyt Hipoteczny 23.
Kucharska-Stasiak E. (2006) Wycena nieruchomości jako źródło ryzyka bankowego w procesie ustanawiania zabezpieczenia hipotecznego. In E. Kucharskiej-Stasiak (Edt.): Ryzyka banku w zakresie określania wartości nieruchomości dla celów kredytowych w Polsce na tle trendów w Unii Europejskiej. Zeszyt Hipoteczny 23.
NBP (2012) Raport o stabilności systemu finansowego, lipiec 2012. Nyberg L. (2005) House Price Developments and Monetary Policy. Speech at Evli Bank, Stockholm, 19
December 2005, Bank of Sweden. Wallace N. (2011). Real Estate Price Measurement and Stability Crises. IRES Working Paper 2011-012. Woods M. (2007) A Financial Stability Analysis of the Irish Commercial Property Market. In: Financial Stability
Report 2007, Central Bank of Ireland.
28 The Report also says, based on the analysis conducted by the Polish Financial Supervision Authority (UKNF), that in the case of some of those loans, tightened banks’ policy, results in loans being considered as impaired despite regular repayment schedule.
36
4. The real estate development and construction sector
The residential construction sector in Poland consists of the so-called order-based system,
with the key role played by construction companies and real estate development companies, and the self-built construction. The first one concerns multi-family housing construction and its role grows along with the size of the local market. The second one is connected with single-family housing construction, predominant in small towns, and to even greater extent, in the countryside. From the point of view of the constructed space supply and the involvement of the financial system, both systems are of similar size, yet differ in structure and risk exposure. From the point of view of the housing situation of the society and market risk, the order-based system and the largest markets were of key importance.
The current situation of the real estate development sector is a direct consequence of the lending boom and home price surges in the years 2005 – 2007. As a result of high concentration of production in the sector and the pre-paid financing scheme, prices were rather rigid despite declining demand. This reduced the risk faced by the financial sector, yet failed to adjust the output volume. Moreover, the Act on the protection of home buyers’ rights, which came into force in April 2012, boosted the construction contracts supply. Inexperienced with the newly implemented Act and concerned about its legal consequences, real estate developers solicited orders during the vactio legis of the Act. Of considerable importance was also the situation in general construction which faced a high domestic and international competition. Finally, also general construction companies shifted to residential construction. As a result of growing supply pressure, a further decline in home prices may be expected.
Both the analysis of business indicators for real estate development companies as well as simulations of their investment activity show that the housing production is profitable and the situation of an average real estate development company does not raise, as for now, any major concerns. Yet, unless the growth rate of new investment projects slows down, the market may face larger-scale price adjustments and companies’ difficulties.
Figure 67 Completed housing in Poland #
0
20 000
40 000
60 000
80 000
100 000
120 000
140 000
160 000
180 000
2004 2005 2006 2007 2008 2009 2010 2011
hous
ing
units
6 cities 10 cities rest of Poland
Figure 68 Completed housing per 1000 inhabitants in Poland#
2,0
3,0
4,0
5,0
6,0
7,0
8,0
9,0
2002 2003 2004 2005 2006 2007 2008 2009 20102011*
ho
usi
ng /
10
00
inha
bita
nts
6 cities 10 cities villages Poland Source: GUS. Note: Estimates for the countryside in 2011
Source: GUS.
37
The year 2011 was unfavourable for the Polish construction industry. The performance of construction companies and real estate development companies involved in general construction was somewhat poorer than the one observed in the previous years. This was driven by the fact that real estate developers facing competitive pressure tended to signcontracts failing to account for rising construction costs. This was reflected in stock indices of the construction sector which fell faster than the WIG20 (index of the 20 largest companies on the Warsaw stock exchange, see Figure 69). Still, it should be remembered that these indices concern a much larger construction sector area than the residential housing sector only. The bulk of those declines were driven by difficulties experienced by construction companies when performing public contracts for road construction and other UEFA Euro 2012 related projects.
Figure 69 Growth rate of stock indices for real estate developers and real estate development companies
0
20
40
60
80
100
120
200
4 Q
12
004
Q2
200
4 Q
32
004
Q4
200
5 Q
12
005
Q2
200
5 Q
32
005
Q4
200
6 Q
12
006
Q2
200
6 Q
32
006
Q4
200
7 Q
12
007
Q2
200
7 Q
32
007
Q4
200
8 Q
12
008
Q2
200
8 Q
32
008
Q4
200
9 Q
12
009
Q2
200
9 Q
32
009
Q4
201
0 Q
12
010
Q2
201
0 Q
32
010
Q4
201
1 Q
12
011
Q2
201
1 Q
32
011
Q4
201
2 Q
1
200
7 Q
2 =
10
0
WIG20 WIG-development companiesWIG-construction companies
Figure 70 Economic indicator for housing production (commenced housing minus completed housing)
-70 000
-50 000
-30 000
-10 000
10 000
30 000
50 000
70 000
90 000
110 000
130 000
150 000
I k
w. 2
006
II k
w. 2
006
III
kw
. 200
6
IV k
w. 2
006
I k
w. 2
007
II k
w. 2
007
III
kw
. 200
7
IV k
w. 2
007
I k
w. 2
008
II k
w. 2
008
III
kw
. 200
8
IV k
w. 2
008
I k
w. 2
009
II k
w. 2
009
III
kw
. 200
9
IV k
w. 2
009
I k
w. 2
010
II k
w. 2
010
III
kw
. 201
0
IV k
w. 2
010
I k
w. 2
011
II k
w. 2
011
III
kw
. 201
1
IV k
w. 2
011
I k
w. 2
012
mie
szk
an
iaPolska
6 miast
Note: harmonized data, 2007 Q2 = 100. The WIG Developers Index has been published since 2007 Q2.
Source: Warsaw Stock Exchange
Note: the index in a 12-month moving average Source: NBP based on PABB concept.
The situation of real estate development companies involved in housing construction was complicated. Despite a considerable housing stock and large number of home construction contracts, the whole of 2011 saw a rise in housing construction (see: Figure 70). This was related to a considerable, as compared to other activities, share of real estate developers’ profits. Also sector “exit costs” and the lack of alternative investment opportunities were an important factor. An additional aspect accelerating the start of investments, especially at the end of 2011, was the imminent expiry of vacatio legis of the Act on the protection of rights of real estate developer’s client29. As lending halted dramatically and prices did not fall30, contrary to prospective clients’ expectations, those housing units failed to find buyers. The launch of new housing construction contracts onto the market, despite the growing number of
29 The new Act did not apply to any constructions started before April 2012. 30 Among other things, due to high concentration of production – see Figure 71. .
38
completed, unsold housing, resulted in higher costs for real estate developers. These costs were related to both frozen capital in unsold housing, its maintenance costs and repayment of bank loans and liabilities to subcontractors. Another important factor affecting the yield of real estate development companies, were considerable fluctuations in the growth rate of construction and assembly production costs, in particular, construction material costs and labour costs, which might have been incorrectly accounted for by real estate developers at the investment planning stage. Consequently, the actual situation of particular companied varied, depending on the time of signing of the construction contracts31 . This is reflected in the indicator analysis of real estate development companies (See Figure 72).
Figure 71 Concentration of real estate development output32 in Poland’s 6 major markets # in 2011
0%
10%
20%
30%
40%
50%
60%
70%
80%
Łódź Poznań Trójmiasto Wrocław Warszawa Kraków
10 largest developers 5 largest developers
Figure 72 Profitability of real estate development companies
0%2%4%6%8%
10%12%14%16%18%20%22%24%
2005 2006 2007 2008 2009 2010 2011
ROE (large real estate developers)
ROE (small real estate developers)
Operating rate of return (large and small companies - average for 6 cities) (R ax) Source: REAS. Statistical data for small companies are considerably
delayed. Source: NBP.
Home construction costs in all the analysed markets, apart from Warsaw, were similar (Figure 73). 2011 saw a slight drop in the share of real estate developer’s yield in home price (see Figure 74 – Figure 80). The Warsaw, Cracow and Poznań markets continued to generate the highest profits for real estate developers, which is connected, among other things, with high GDP per capita. In Gdańsk and Wrocław these yields levelled off at a steady and similar level, whereas in Łódź they followed the strongest downward trend. The main factor affecting the real estate developer’s profitability, amidst relatively stable home construction costs, were market prices of housing in particular cities.
31 For a more detailed analysis see Annex 3. 32 The share of housing built by five and ten major real estate developers, in the total number of completed housing or housing scheduled for completion in particular market by the end of 2011.
39
Figure 73 Construction cost per 1 square meter of usable area in a residential building (type 1121#)
1 500
1 700
1 900
2 100
2 300
2 500
2 700
2 900
200
4 Q
22
004
Q3
200
4 Q
42
005
Q1
200
5 Q
22
005
Q3
200
5 Q
42
006
Q1
200
6 Q
22
006
Q3
200
6 Q
42
007
Q1
200
7 Q
22
007
Q3
200
7 Q
42
008
Q1
200
8 Q
22
008
Q3
200
8 Q
42
009
Q1
200
9 Q
22
009
Q3
200
9 Q
42
010
Q1
201
0 Q
22
010
Q3
201
0 Q
42
011
Q1
201
1 Q
22
011
Q3
201
1 Q
42
012
Q1
PLN
/ m
2
Warszawa Kraków Poznań
Wrocław Gdańsk Łódź
Figure 74. Changes in the share of the real estate developer’s profits in the market price per 1 square meter of housing (type 1121#)
0%
10%
20%
30%
40%
50%
60%
70%
200
4 Q
4
200
5 Q
4
200
6 Q
4
200
7 Q
4
200
8 Q
2
200
8 Q
3
200
8 Q
4
200
9 Q
1
200
9 Q
2
200
9 Q
3
200
9 Q
4
201
0 Q
1
201
0 Q
2
201
0 Q
3
201
0 Q
4
201
1 Q
1
201
1 Q
2
201
1 Q
3
201
1 Q
4
201
2 Q
1
Warszawa Kraków GdańskWrocław Poznań Łódź
Source: NBP based on Sekocenbud data. Source: NBP based on Sekocenbud data.
Figure 75 Warsaw – price structure of 1 square meter of housing usable area (type 1121#) for consumer
01 0002 0003 0004 0005 0006 0007 0008 0009 000
200
4 Q
4
200
5 Q
4
200
6 Q
4
200
7 Q
4
200
8 Q
4
200
9 Q
4
201
0 Q
4
201
1 Q
4
201
2 Q
1
PLN
/ m
2
labour building materials equipment indirect costs
constructor profit overheads project land
VAT developers profit
Figure 76 Cracow - price structure in 1 square meter of housing usable area (type 1121#) for consumer
01 0002 0003 0004 0005 0006 0007 0008 0009 000
200
4 Q
4
200
5 Q
4
200
6 Q
4
200
7 Q
4
200
8 Q
4
200
9 Q
4
201
0 Q
4
201
1 Q
4
201
2 Q
1
PLN
/ m
2
labour building materials equipment indirect costs
constructor profit overheads project land
VAT developers profit Source: NBP based on Sekocenbud, REAS data. Source: NBP based on Sekocenbud, REAS data.
40
Figure 77 Gdańsk - price structure of 1 square meter of housing usable area (type 1121#) for consumer
01 0002 0003 0004 0005 0006 0007 0008 0009 000
2004
Q4
2005
Q4
2006
Q4
2007
Q4
2008
Q4
2009
Q4
2010
Q4
2011
Q4
2012
Q1
PLN
/ m
2
labour building materials equipment indirect costs
constructor profit overheads project land
VAT developers profit
Figure 78 Poznań - price structure of 1 square meter of housing usable area (type 1121#) for consumer
01 0002 0003 0004 0005 0006 0007 0008 0009 000
200
4 Q
4
200
5 Q
4
200
6 Q
4
200
7 Q
4
200
8 Q
4
200
9 Q
4
201
0 Q
4
201
1 Q
4
201
2 Q
1
PLN
/ m
2
labour building materials equipment indirect costs
constructor profit overheads project land
VAT developers profit
Source: NBP based on Sekocenbud, REAS data. Source: NBP based on Sekocenbud, REAS data.
Figure 79 Wrocław - price structure of 1 square meter of housing usable area (type 1121#) for consumer
01 0002 0003 0004 0005 0006 0007 0008 0009 000
200
4 Q
4
200
5 Q
4
200
6 Q
4
200
7 Q
4
200
8 Q
4
200
9 Q
4
201
0 Q
4
201
1 Q
4
201
2 Q
1
PLN
/ m
2
labour building materials equipment indirect costs
constructor profit overheads project land
VAT developers profit
Figure 80 Łódź - price structure of 1 square meter of housing usable area (type 1121#) for consumer
01 0002 0003 0004 0005 0006 0007 0008 0009 000
200
4 Q
4
200
5 Q
4
200
6 Q
4
200
7 Q
4
200
8 Q
4
200
9 Q
4
201
0 Q
4
201
1 Q
4
201
2 Q
1
PLN
/ m
2
labour building materials equipment indirect costs
constructor profit overheads project land
VAT developers profit
Source: NBP based on Sekocenbud, REAS data. Source: NBP based on Sekocenbud, REAS data..
41
The analysis of the situation of real estate development companies should take into account a number of important aspects. A real estate developer incurs a series of costs which are not shown in the costs estimates released by Sekocenbud33, which, when used by the NBP analysts, are subject to uncertainty.
Real estate development companies, as suggested by the index-based analysis of investment project business plans, perform better than indicated in the Central Statistical Office (F01 and F02)34 reports. According to CSO data, the year 2011 saw return on equity ratios fall for both large and smaller real estate development companies (see Figure 81-Figure 82). Yet, a ROE at the level of approx. 12%, in times of a worsening general economic situation, was not a bad result. As the decline in the ROE ratio was faster than the decline in the share of developer’s margin in the price per one square meter of housing, the performance deterioration should be attributed to another factor. This was the signing of new investment projects and purchases of land made in advance (see Figure 83-Figure 84), with sales that could not keep the pace35 (see Figure 85-Figure 86). Despite noticeable attempts of real estate development companies to cut costs, ROA and ROE ratios were on the decline.
Figure 81 Profitability indicators # of an average large real estate development company
0
5
10
15
20
25
2005 2006 2007 2008 2009 2010 2011
%
ROA ROE Selling Reliability
Figure 82 Profitability indicators # of an average small-sized real estate development company
0
1
2
3
4
5
6
7
8
9
10
2005 2006 2007 2008 2009 2010
%
ROA ROE sales profitability
Source: GUS (F01, F02). Source: GUS (F01, F02).
33 Sekocenbud does not take into account project expenses, taxes, land purchase costs and related cost estimates. 34 CSO data (F01 and F02) concern a small number of entrepreneurs classified as engaged in real estate development activity under a relevant PKD 2007 code. Yet, a large part of real estate development and construction companies operate under another PKD code, thus being left out from the analysis. 35 According to the International Accounting Standard, sales of construction production involve completed housing rather than sold home construction contracts.
42
Figure 83 Broad economic categories of an average large real estate development company
0,0
0,2
0,4
0,6
0,8
1,0
1,2
1,4
0
20
40
60
80
100
120
2005 Q4 2006 Q4 2007 Q4 2008 Q4 2009 Q4 2010 Q4 2011 Q4
pro
duc
tion
ye
ars
PL
N m
illio
n
ongoing real estate development projects completed housing
land stock land stock in production years
Figure 84 Broad economic categories of an average small-sized real estate development company
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
4,5
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
4,0
2005 2006 2007 2008 2009 2010
lata
mln
zł
projekty deweloperskie w toku mieszkania gotowe
bank ziemi bank ziemi w latach produkcji Source: GUS (F01, F02). Source: GUS (F01, F02).
Figure 85 Sales of an average large real estate development company
0
50
100
150
200
250
300
350
0,0
20,0
40,0
60,0
80,0
100,0
120,0
140,0
160,0
180,0
2005 2006 2007 2008 2009 2010 2011
tho
usan
d o
f m2
PLN
mill
ion
sales (in PLN million) sales (in thousands of m2)
Figure 86 Sales of an average small-sized real estate development company
0
2
4
6
8
10
12
14
16
18
0,0
1,0
2,0
3,0
4,0
5,0
6,0
7,0
8,0
2005 2006 2007 2008 2009 2010
thou
sand
s of
m2
PLN
mill
ion
sales (in PLN million)
Source: GUS (F01, F02), NBP. Source: GUS (F01, F02), NBP.
The discussed processes are easily observable in an index-based analysis of large real estate development companies (see: Figure 87). A similar situation was seen in the case of small-sized real estate development companies, yet, statistical data on them are limited to 2010 (see Figure 88).
43
Figure 87 Financial results of an average large real estate development company
Figure 88 Financial results of an average small-sized real estate development company
0
2
4
6
8
10
12
14
16
18
0
20
40
60
80
100
120
140
160
180
2005 2006 2007 2008 2009 2010 2011
%
PLN
mill
ion
total revenues total costsnet result ROA (right-hand axis)ROE (right-hand axis) sales profitability
0
1
2
3
4
5
6
7
8
9
10
0
1
2
3
4
5
6
7
8
9
10
2005 2006 2007 2008 2009 2010
%
PLN
mill
ion
total revenues total costs
net result ROA (right-hand axis)ROE (right-hand axis) sales profitability
Source: GUS (F01, F02). Source: GUS (F01, F02).
The analysis of the financing structure of large real estate development companies indicates a far-reaching conservatism (see Figure 89- Figure 90). Equity had a large share in investment financing (approx. 38% at the end of 2011), whereas other sources of financing were strongly diversified and safe for the real estate developer (client’s pre-payments accounted for 12%, debt securities for 17% 36, and amounts payable to suppliers for approx. 7%). Borrowings accounted for less than 13%. The financing structure of small-sized companies in 2011 (more recent data are not available) was even more conservative, and their equity in 2011 amounted to approx. 48% as compared to 66% recorded in 2009.
Figure 89 Financial structure of an average large real estate development company
52,4 51,1 47,7 44,5 40,434,3 38,1
13,210,8
11,8 18,618,5
11,512,6
7,0 10,2 12,711,5
9,3
16,617,0
10,3 13,9 16,1 10,714,2
10,7
12,0
8,77,5 6,5
6,96,9
5,8
6,9
8,4 6,6 5,2 7,8 10,721,0
13,4
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2005 2006 2007 2008 2009 2010 2011
otherliabilities
tradeliabilities
clientprepayments
indebtedbonds
loans
equity
Figure 90 Financial structure of an average large real estate development company
73,0 69,5
54,444,8
66,4
48,2
11,712,5
6,4
12,5
7,5
14,6
0,00,0
0,00,0
0,0
1,2
6,6 9,8
10,87,3
7,2
7,0
6,4 5,8
25,731,6
11,620,0
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
2005 2006 2007 2008 2009 2010
otherliabilities
tradeliabilities
clientprepayments
indebtedbonds
loans
equity
* Including client pre-paid payments and amounts payable to construction companies; Source: GUS (F01, F02).
Source: GUS (F01, F02).
36 Growing role of debt securities, amidst declining lending volume, suggests that large companies managed to replace bank borrowing with funds raised directly in the financial market.
44
The financing structure of real estate developers is affected by banks’ behaviour. Data on banks’ large exposures37 suggest that in 2011 they were rather more inclined to finance real estate development activity. As compared with 2010, the number of loan-financed enterprises rose, and many companies increased their bank debt. Despite the growing bank exposure to real estate development financing, the scale of this phenomenon was not large (see Figure 91-Figure 92).
Figure 91 Number of real estate development companies presented in B300 report and banks’ exposure to these companies
0
5
10
15
20
25
0
100
200
300
400
500
600
700
800
900
1000
2005 2006 2007 2008 2009 2010 2011
PL
N b
illi
on
number of real estate developers
total banks' capital commitment
Figure 92 Number of companies to which banks’ increased their exposure in a particular year
0
10
20
30
40
50
60
70
80
0
100
200
300
400
500
600
2006 2007 2008 2009 2010 2011
%
number of companies in which banks' capital commitment increased share
Source: NBP. Source: NBP.
The quality of loan portfolio for the construction sector deteriorated slightly in 2011 which might have been driven by the sector’s declining profitability. As compared to 2010, the share of real estate development companies facing bank debt servicing problems increased. Also the share of debt of such enterprises picked up38 (see Figure 93). In half of the enterprises facing debt servicing difficulties, the problems started during the last year (see Figure 94).
Yet, when analysing the data it should be borne in mind that more companies than indicated in the CSO classification are engaged in real estate development, and thus, the analysis of reported data is subject to a certain error.
37 Bank’s exposure to a particular client (e.g. extended loan) exceeding PLN 500 thousand. Real estate development companies were selected mainly on the basis of their PKD 2007 number (as in the case of GUS data). 38 Companies whose debt with any of their banks was classified as doubtful debt.
45
Figure 93 Quality of loan portfolio to developers
0
2
4
6
8
10
12
14
16
18
20
2005 Q4 2006 Q4 2007 Q4 2008 Q4 2009 Q4 2010 Q4 2011 Q4
%
share of number share of debt
Figure 94 Companies which started to face debt servicing problems in a particular year
0,00
1,00
2,00
3,00
4,00
5,00
6,00
7,00
8,00
9,00
0
10
20
30
40
50
60
70
80
2006 2007 2008 2009 2010 2011
number of companies which started to face financial problems in a particular year
share
Note: The green line shows the percentage of all real estate development companies represented in the register of large exposures, facing bank debt servicing problems at the end of the year. The violet line shows the share of those companies in the aggregate debt of all real estate development companies. Source: NBP.
Note: The green line shows the percentage of companies facing financial difficulties where such problems started during a particular year, or the share of companies which faced financial difficulties in a particular year, and which did not experience such problems a year before (provided a particular company was not in the previous year’s base, it was assumed not to have suffered any financial problems) Source: NBP.
5. Commercial real estate in Poland 39
In 2011 the commercial real estate market40 was in its upswing phase of the cycle41, which
started already in 2010. This is the period of rebound, following a downturn, driven by the collapse of the global economy. It should be emphasised that the analysed office, modern retail and warehouse real estate market, located in prime locations, is dominated by foreign investors, thus is largely affected by the euro area developments. The growth in the value of investment transactions to EUR 2.5 billion in 2011 (see Figure 95) and declining
39 The study focuses on modern commercial real estate as it is the object of transactions conducted by large real estate agencies and the scale of the transactions has a strong, direct impact on the domestic economy. Modern real estate enables the adjustment of space to the customers’ needs. The analysis was supported with the knowledge of experts of particular agencies involved in commercial real estate consulting, intermediation or management. It should be emphasized that particular agencies, on whose data the authors relied, may use various definitions and indicators applicable to commercial real estate. This fact should be taken into account when drawing conclusions on the situation in the commercial real estate market in Poland. 40 An introduction to the commercial real estate market can be found in the previous edition of the NBP’s annual report. 41 The commercial real estate market cycle in Poland, and in particular its dynamics, may be divided into four phases: (1) a gradually growing value of transactions in pre-accession years; (2) acceleration in the years close to the EU accession, with the highest level in 2006; (3) a gradual drop in the years 2007–2008 under the intensifying financial market crises faced by many countries, to the lowest level of transactions recorded in Poland in the year 2009; (4) rebound of the upward trend in 2010 and 2011, with approx. EUR 2 and 2.5 billion worth transactions (See Cushman & Wakefield data and Figure 95).
46
capitalization rates on transactions involving prime real estate42 (see Figure 96) point at a recovery in this market. Similarly to the previous years, over 90% of investment transactions, in value terms, were conducted by foreign investors. The share of Polish investors as compared to the share of foreign investors is one of the lowest in Europe (see: data from the Cushman & Wakefield Marketbeat Report, Spring 2012). In 2011, the bulk of investments involved retail real estate transactions (47%) and office real estate transactions (46%), whereas 6% of investment concerned warehouse real estate (see: data from the Cushman & Wakefield Marketbeat Report, Spring 2012).
Figure 95 Value of investment transactions in EUR million
0
1 000
2 000
3 000
4 000
5 000
6 000
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Q1
2012
Offices Retail Warehouses Others
Figure 96 Capitalization rate on investments in real estate in prime locations
5%
6%
7%
8%
9%
10%
11%
12%
13%
14%
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Q1
2012
Offices Retail Warehouses
Source: Cushman & Wakefield Source: DTZ
Office space At the end of 2011, the 7 largest cities offered 5.6 million square metres of modern office
space, of which 3.6 million square meters were in Warsaw (see data by Jones Lang LaSalle). Warsaw continues to be the largest office space market in Central and Eastern Europe (see data by Jones Lang LaSalle). Due to a qualified workforce, relatively low wages as well as its good location and access to international communication networks, companies decide to invest and create jobs in Poland. Many international companies have their Central and Eastern European headquarters in Warsaw and other big cities, which also attracts large inflows of foreign direct investment. An important factors behind the demand for modern office space is the on-going, dynamic expansion of the modern business services sector (SSC/BPO i.e. shared service centres/ business process outsourcing).
The space growth in 2011 was not strong. This was caused by the crisis in the global financial markets in 2009 which slowed down demand for office space in Poland. As a result, the pace of new investment slackened, with investors waiting for a more favourable time to embark on new investments. This fact, due to an average construction period of 18-24 months, affected new space supply in 2011. On the other hand, stabilization of the global economy and sustainable economic growth in Poland in 2010, urged investors to embark on new construction, which will result in a strong rise in office space in the near future (see
42 Prime real estate is well-located and has a good standard. Moreover, real estate with high rate of rented space and with highly valued tenants is considered as prime.
47
Figure 97). Developers’ decisions were also affected by declining vacancy rates# (see Figure 98) and stable or slightly growing rents, which did not differ much across regional cities (see Figure 99). Rents outside central Warsaw were slightly higher, with rents in the Central Business District (CBD) almost twice as high as in other local markets. The share of non-rented space in the total space, which in 2010 differed considerably across markets, was getting close to the 5%-10% level (see Figure 98). Łódź sees continued high vacancy ratio, oscillating around 20%. Due the economic slowdown, strong rise in the supply of new office space observed in the years 2008-2009 has not yet been matched by a considerably growing demand. Since the beginning of 2010, Wrocław observed the lowest vacancy rate of all the regional markets, with is the effect of a dynamic development of business outsourcing centres amidst limited supply of new space. Also Katowice saw a very robust demand for office space, which led to a relatively strong decline in non-rented space.
Figure 97 Aggregate supply of modern office space (in millions of square metres)
Figure 98 Office space vacancy rate in half-year periods
0%
5%
10%
15%
20%
25%
H2
200
6
H1
200
7
H2
200
7
H1
200
8
H2
200
8
H1
200
9
H2
200
9
H1
201
0
H2
201
0
H1
201
1
H2
201
1
H1
201
2
Warszawa Kraków Wrocław Trójmiasto
Poznań Łódź Katowice
Source: DTZ Source: DTZ
The cycle in the market of office real estate investments and the market perception by investors, are reflected in the capitalization rates on prime real estate (see Figure 100). Capitalization rates in most markets in 2011 remained stable, yet differed considerably as far as location is concerned. Warsaw, as the most mature and stable market, enjoys a significant popularity among international investors, continuing to register the lowest and stable capitalization rates.
48
Figure 99 Rents (EUR/square meter/month) for office space in prime locations
5
10
15
20
25
30
35
2005
Q1
2005
Q2
2005
Q3
2005
Q4
2006
Q1
2006
Q2
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
Serie1 Warszawa - non CBD
Kraków Wrocław
Gdańsk Poznań
Łódź Katowice
Figure 100 Capitalization rates on investment in modern office space in prime locations
5%
6%
7%
8%
9%
10%
11%
2005
Q1
2005
Q2
2005
Q3
2005
Q4
2006
Q1
2006
Q2
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
Warszawa - CBD Warszawa - non CBDKraków WrocławGdańsk PoznańŁódź Katowice
Source: Cushman & Wakefield Source: Cushman & Wakefield
The size of the office space stock, a relatively low vacancy rate and high rents sets the Warsaw office space market apart from the rest of the country. On account of the importance and size of the Warsaw market, the Warsaw Research Forum (WRF)43 distinguishes between the Central Business District, the city centre and the rest of the city, divided into zones. Similarly to the previous years, in 2011 the bulk of new office space was built outside the city centre (see Figure 101). This was driven by the relatively high availability of cheaper land and relatively improving urban transport in the remaining part of the city. Provided the tenant does not have to locate its office in the Central Business District, they chose a place outside the city centre where rents are lower. A lot of completed office space was located near the Warsaw airport which provides an easy access to international communication networks. The vacancy rate in most parts of Warsaw stabilized already in 2010 has remained at almost identical, low level since 2011 (see Figure 102).
43 The Warsaw Research Forum is a forum of consulting companies, including CBRE, Colliers, Cushman&Wakefield, DTZ, Jones Lang LaSalle, Knight Frank and Savills.
49
Figure 101 Annual supply of new office space in Warsaw (in square meters)
0
50000
100000
150000
200000
250000
300000200
5
200
6
200
7
200
8
200
9
201
0
201
1
2012
*
CBD City Centre Non Central
Figure 102 Vacancy rate in particular parts of Warsaw
0%2%4%6%8%
10%12%14%16%18%20%
2004
2005
2006
2007
2008
2009
2010
2011
2012
*
CBD City CentreNon Central Total
Source: Jones Lang LaSalle, WRF Note: The 2012 estimates are from Jones Lang LaSalle
Source: Jones Lang LaSalle, WRF
Modern retail space – shopping centres In 2011 the aggregate supply of retail space in shopping centres #44 in Poland
amounted to the total of 8.7 million square meters, of which 5.1 million square meters were located in the largest agglomerations (see Figure 103). As published by the Retail Research Forum of the Polish Council of Shopping Centres, the bulk of the new space was built in medium-sized cities (with 50-200 thousand inhabitants), or even smaller ones. Such investment decisions might have been driven by some level of saturation in large agglomerations, observed by the investors45. Moreover, the absence of a direct competition and low prices of land in smaller cities and towns encouraged investors to take the risk and invest in cities where they may not necessarily reckon with wealthier clients of large cities.
The supply of retail space per one thousand inhabitants46 points to significant differences across markets (see Figure 104). In 2011, similarly to the preceding year, the highest space per one thousand inhabitants ratio was recorded in the Wrocław agglomeration and Poznań agglomeration, whereas the fastest growth was observed in the Szczecin agglomeration. Taking into account the purchasing power of individuals in particular countries, the retail space saturation in Poland, when compared with the situation in the Western European markets, continues to be low and there is a considerable growth potential for the Polish retail market.
44 The Retail Research Forum of the Polish Council of Shopping Centres works on the standardization of market information. 45 DTZ Property Times „Poland – Shopping centres in medium-sized cities. Steady growth of interest.” December 2011. 46 The numbers in the figure are estimative as the number of persons which may potentially visit shopping centres may exceed the number of agglomeration inhabitants given by the CSO.
50
Figure 103 Aggregate supply of modern retail space (in million square metres) in large agglomerations and the rest of Poland
0
1
2
3
4
5
6
7
8
9
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Warszawska Krakowska WrocławskaTrójmiejska Poznańska ŁódzkaKatowicka Szczecińska Rest of Poland
Figure 104 Aggregate supply of modern retail space in large agglomerations in square metres per 1000 inhabitants
Note: Data refer to agglomerations and other small and medium-sized cities (the rest of Poland) Source: Polish Council of Shopping Centres. .
Note: Data refer to agglomerations. Source: Polish Council of Shopping Centres.
Rents in prime locations, which were on the decline in the majority of local markets
since the outbreak of the crisis, have continued on an upward trend since mid-2011 (see Figure 105). Rents in all cities are closer to each other than before the onset of the crisis. The relatively big number of customers visiting large, modern shopping malls in Warsaw boost tenant demand for such retail space. As a result, rents in Warsaw considerably exceed rents in other cities. Generally, shopping centres enjoy high demand from the tenants' side. The vacancy rate in the largest agglomerations was very low47, and ranged between 0.5%-3%. In Warsaw and Szczecin it was below 1%. Moreover, the vacancy rate decreased slightly both in the Silesian agglomeration and Tricity agglomeration.
Capitalization rates falling persistently until the end of 2009 amidst constant or growing rents suggest that prices of commercial real estate were on a steady rise (see Figure 106). Apart from considerable impact of the level of capitalization rates in Western Europe on capitalization rates in Poland, this may be indicative of real estate buyers expecting further rent rises. Yet, in the current economic situation and amidst robust growth in new retail space, rents might not rise.
47 Colliers “Poland – Research & Forecast Report 2012”.
0
100
200
300
400
500
600
700
800
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Warszawska Krakowska WrocławskaTrójmiejska Poznańska ŁódzkaKatowicka Szczecińska
51
Figure 105 Rents (EUR/square meter/month) for retail space in prime locations
25
35
45
55
65
75
85
95
2005
Q1
2005
Q2
2005
Q3
2005
Q4
2006
Q1
2006
Q2
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
Warszawa (in-town) Warszawa (out-of-town)Kraków WrocławTrójmiasto PoznańŁódź Kon. KatowickaSzczecin
Figure 106 Capitalization rates on investments in retail space in prime locations
5%
6%
6%
7%
7%
8%
8%
9%
9%
2005
Q1
2005
Q2
2005
Q3
2005
Q4
2006
Q1
2006
Q2
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
Serie1 Warszawa (out-of-town)Kraków WrocławTrójmiasto PoznańŁódź Kon. Katowicka
Note: Rents refers to fashion and accessories retail space of approx. 100 m square meters, located on the ground flood. Source: Cushman & Wakefield.
Note: Capitalization rates for all the markets except for Warsaw were identical by the end of 2008. Source: Cushman & Wakefield.
Other commercial space Many shops and retail premises, apart from shopping centres, are located in main
shopping streets. Although these premises are considerably smaller than premises in shopping centres and more difficult to arrange to meet customers’ needs, their rents are similar, and sometimes even exceeds those paid for shopping centre premises. This is driven by their prestigious and attractive location and easy access to urban communication. Rents for those premises vary considerably depending mainly on the city and location within the city. Figure 107 shows rent ranges in main shopping streets in 2011, whereas Figure 108 shows the path for the highest rents in main shopping streets in the years 2005-2012. In 2011, rents at main shopping centres in Warsaw and Cracow were stable or followed an upward trend, showing stagnation or even a downward trend in other cities.
Figure 107 Rents (EUR/square meter/month) in main shopping streets in 2011.
Figure 108 The highest rents EUR/square meter/month) in main shopping streets.
-
10
20
30
40
50
60
70
80
90
2005
Q1
2005
Q2
2005
Q3
2005
Q4
2006
Q1
2006
Q2
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
Warszawa Kraków Wrocław Gdańsk
Poznań Łódź Katowice Szczecin Note: The highest and the lowest rents; Source: Cushman & Wakefield.
52
Warehouse space In 2011 almost 400 thousand square metres of new warehouse space were completed.
As a result, the total warehouse stock increased to reach approx. 7 million square meters (see Cushman & Wakefield data). Poland is an important logistic centre in the Central and Eastern Europe, and in 2011 Poland’s total warehouse space supply accounted for 43% of the total warehouse space supply in the CEE (see report published by Jones Lang LaSalle48). As warehouse space needs to be situated in the vicinity of large cities and provide an easy access to good transportation network, the preferred location are the environs of the largest agglomerations and proximity of international communication networks. Similarly to the previous years, the bulk of the warehouse space is located in the Warsaw region, Upper Silesian region, Łódź region, Poznań region and Lower Silesian region. In year-on-year terms, warehouse space growth in 2011 outpaced that in 2010 by 100 thousand square meters. The largest growth in warehouse space was observed in the Upper Silesian region, and the highest growth rate was reported in the Cracow region (see Figure 109). In total, the share of speculative investment declined, and approx. 40% of warehouse space was built to order (see: CBRE data49), which may suggest a more prudential approach of investors. Strong concentration in the warehouse space market increased further, and five largest real estate developers hold the total of 83% of warehouse stock (as compared to 79% in 2010, see Jones Lang LaSalle data). In 2011 rents for warehouse space in prime locations were stable all over Poland (see Figure 110), with persisting considerable differences across regions. Rents in Warsaw remain at the highest level. Offer price for warehouse space in Łódź and Cracow hovered around EUR 4.5 monthly per square meter, whereas they were EUR 1.00 lower in other cities.
Figure 109 Aggregate warehouse space supply in Poland’s regions in millions of square metres
0
1
2
3
4
5
6
7
2005
2006
2007
2008
2009
2010
2011
Warszawski Krakowski WrocławskiTrójmiejski Poznański Central PolandGórnośląski
Figure 110 Rents (EUR/square meter/month) for warehouse space in prime locations
2,00
2,50
3,00
3,50
4,00
4,50
5,00
5,50
6,00
2005
Q1
2005
Q2
2005
Q3
2005
Q4
2006
Q1
2006
Q2
2006
Q3
2006
Q4
2007
Q1
2007
Q2
2007
Q3
2007
Q4
2008
Q1
2008
Q2
2008
Q3
2008
Q4
2009
Q1
2009
Q2
2009
Q3
2009
Q4
2010
Q1
2010
Q2
2010
Q3
2010
Q4
2011
Q1
2011
Q2
2011
Q3
2011
Q4
2012
Q1
Warszawa Kraków Wrocław Gdańsk
Poznań Łódź Katowice Source:Cushman & Wakefield. Source: Cushman & Wakefield.
48 Jones Lang LaSalle, 2011: Warehouse space market in Poland 49 CBRE „Big Box Poland – Warehouse space market in Poland”, 2011 Q4.
53
Glossary of terms and acronyms
AMRON – System for the Analysis and Monitoring of Real Estate Market Transactions.
BaRN – Real Estate Market Database. The database that includes offer and transaction prices of housing in the markets of 16 voivodship capitals of Poland. It also holds data on market rents. The data come from real estate brokers, housing cooperatives and real estate developers who volunteered for the study and partially also from the Registers of Prices and Values of Real Estate kept by particular counties. The data are gathered and verified by the Regional Branches of the NBP.
BIK – Credit Information Bureau.
Shopping centre – retail real estate that has been planned, constructed and managed as a single retail entity, consisting of common parts, with a minimum gross leasable area (GLA) of 5,000 sq. m, and a minimum of 10 shops (definition developed by the Polish Council of Shopping Centres).
D/I – households’ gross disposable income
Loan availability – measure of potential loan available at a specific interest rate, depreciation, lending requirements (social minimum) and average monthly wage in the enterprise sector. It indicates the amount of loan that the borrower is able to obtain for the average monthly wage in the enterprise sector in a particular market (GUS), in view of bank’s lending requirements and loan parameters (interest rate, depreciation period, social minimum understood as the minimum income after the payment of loan instalments). Important information is provided by the rate of changes and regional differentiation rather than the indicator value alone.
Housing availability – measure of potential ability to purchase housing at the offer price for the average monthly wage in the enterprise sector. It indicates the number of square metres of housing with an average offer price in a particular market (PONT Info) that can be purchased for the average wage in the enterprise sector in a particular city (GUS).
Ten cities – Szczecin, Katowice, Bydgoszcz, Opole, Olsztyn, Rzeszów, Kielce, Zielona Góra, Białystok, Lublin.
Financial leverage –ratio of liabilities and provisions for liabilities to equity.
PONT Info Nieruchomości (PONT Info) – database holding data on real estate offer prices. The data are gathered by the company of PONT Info.
Global creditworthiness – measure indicating overall creditworthiness (mortgage loans) of all households in Poland’s cities. It is calculated based on individual household disposable income (household budgets according to GUS) as well as bank lending requirements and loan parameters.
Hedonic housing price index – measure reflecting the ‘pure’ price change, i.e. a change resulting from factors other than home quality differences. The index accounts for
54
changes in housing quality in the study samples in each quarter, which distinguishes it from the growth rate of an average price median50.
Weighted average index – measure reflecting price growth adjusted for one of the most important home quality variables – location. Home price growth is calculated independently for selected locations (districts) and then aggregated in the weighted average formula.
Quality of mortgage loan portfolio – measure of percentage share of mortgage loans overdue for 91-180 days in the total of mortgage loans in a particular city.
Availability of loan-financed housing –measure specifying how many square metres of housing at an average offer price in a particular market (PONT Info) may be purchased for a mortgage loan obtained based on the average monthly wage in the enterprises sector in a particular market (GUS), in view of bank’s lending requirements and loan parameters (interest rate, depreciation period, social minimum understood as the minimum income after payment of loan instalments). Also index growth rate and spreads between particular markets provide important information
LTV (Loan to Value) – ratio of the value of the loan granted to the value of the loan collateral.
Small and large real estate developers – analysed real estate development companies selected on the basis of economic activity classification number PKD2007. They were divided into large and small ones taking into account both the headcount and the value of earnings. Companies employing less than 50 people are considered as small, others are large.
MDR (Mortgage Debt Ratio) – percentage share of mortgage loans repayment in the borrower’s budget.
Cities 200+ – means all cities in Poland with a population of at least 200 thousand.
Financial market surplus liquidity – surplus of financial capital that has not been distributed by financial institutions.
Building 1121 – a residential multi-family five-storey building, which since 2004 has served as the basis for monitoring the average price of construction of one square metre of an average housing unit (see: the Construction Prices Bulletins by Sekocenbud).
P/I (Price to Income) – ratio determining the relationship of the price of an average housing unit in a particular year to the average disposable income.
Sub-rental (or occasional rental) –temporary rental by home owner of the whole or part of his real estate against a specific fee.
PSBD – Polish Association of Home Builders.
Credit rationing – restricted lending resulting from banks’ own assessment of growing risk. In specific situations this may lead to declines in the value of newly granted loans, despite the absence of major changes in the current creditworthiness of the borrower, which may lead to self-fulfilling forecast.
50 More information in the article entitled Hedonic price indexes determination as the method of goods quality change control, M. Widłak (2010), Wiadomości Statystyczne (Statistical News) No 9.
55
Recommendation S – collection of good practices regarding mortgage-secured credit exposures. It was introduced in 2006 by the Commission for Banking Supervision, based on Article 137 clause 5 of the Banking Law Act (Journal of Laws No. 72/2002, item 665, as amended).
Recommendation T – collection of good practices in managing the risk of retail loan exposures. It was introduced in 2010 by the Polish Financial Supervision Authority, based on Article 137 clause 5 of the Banking Law Act (Journal of Laws No. 72/2002, item 665, as amended).
Sales profitability – ratio of net result to sales revenues.
RoA (Return on Assets) –ratio of net result to assets at the end of the period.
RoE (Return on Equity) – ratio of net result to equity at the end of the period.
Rodzina na Swoim (RnS) (Family’s own housing) – government-subsidized programme intended to support housing construction.
SARFIN – Analytical System for the Real Estate Financing Market.
Sekocenbud –publishing house gathering data on costs in the construction sector; the team makes use of the quarterly Construction Prices Bulletins (BCO) – building.
Office real estate standard – office space is classified according to the standard offered. Classification depends on the age of the building, its location, possibility to customize the space, technical specification (e.g. raised floors or suspended ceilings), underground and over ground parking lots and other factors important from the tenant’s point of view.
Capitalisation rate – quotient of net operating income that may be gained on the market and the market price of real estate (in accordance with the General Domestic Valuation Principles).
Six cities – Warsaw, Cracow, Wrocław, Poznań, Gdańsk, Łódź (whenever seven cities are mentioned, Gdynia is included in the group).
TBS (Social Building Society) – company operating under the Act of 26 October 1995 on certain forms of subsidizing housing construction (consolidated text in Journal of Laws No. 98/2000, item 1070, as amended). The object of the company’s operation is housing construction and its rental, provision of management and administration services and conduct of business related to housing construction and accompanying infrastructure. It was planned that TBS offer would be addressed to non-affluent families eligible for loan subsidy from the National Housing Fund (KFM). The tenants pay rent, which is usually higher than in municipal housing (as loan is repaid from the rent) but lower than the market rent.
TDR (Total Debt Ratio) – percentage share of loan repayment in the borrower’s budget.
Vacancy rate – relation of non-rented space to the accumulated (total) supply of commercial space in a particular location, e.g. town or district.
Profitability ratios – ROA (return on assets) – relation of net income to assets at the end of the period, ROE (return on equity) – relation of net income to equity at the end of the period, profitability of net sales – net profit in relation to sales income.
Specialized rental – lease of housing space constructed particularly for that purpose. The owner of the stock for rental may be both a legal person (municipality, local government, real estate fund) and a private individual. In Poland the market is small and decapitalised.
56
Risk assessment: PLN loan risk (PLN mortgage loan rate minus 10-year Treasury bond yield) CHF/PLN exchange rate risk (CHF mortgage loan rate minus LIBOR3MCHF), EUR/PLN exchange rate risk (euro mortgage loan risk minus valuation of PLN loan risk minus LIBOR3MEUR).
ZBP – Polish Bank Association.
57
Part II Analytical research papers
A1. Development trends in the local markets (comparative analysis of 16 cities in Poland)
Baldowska Grażyna 51, Leszczyński Robert 52, Myszkowska Barbara 51 In 2011, the residential real state market saw relative stabilization and growing number of
transactions, both in the primary and the secondary market as compared to 2010. In the BaRN data base, the primary market saw 40% growth in the number of offers and 7% growth in the number of transactions. On the other hand, in the secondary market, the number of offers increased by 16% and the number of transactions rose by 30% (one should take into account continuous search for new real estate broker, which does not necessarily translate into higher number of transactions). In this context, it is difficult to set the breakdown criteria to divide cities into groups with similar behaviour of one particular variable (e.g. housing structure, demographic data or BaRN price data). Cities grouped together at particular stages of the analysis do not form permanent groups. Markets having in common one feature, differ considerably as regards another feature. In the majority of conducted analyses, the largest markets, including, in particular the Warsaw market, diverge from the rest. These are the most liquid markets, with the largest number of observations (offers and transactions). Initially, cities were grouped with the use of tree diagrams (see Figure 111 – Figure 116) which identify, at subsequent stages, similar features. The shorter the distance between particular cities, the more similar they are from the point of view of analysed variables.
Figure 111 Tree diagram of housing situation in voivodship cities (average housing area, usable housing area per person, average number of rooms in a dwelling, average number of persons in a dwelling) in 2011.
Figure 112 Tree diagram of demographic data (demographic growth, migration balance, marriages per 1000 inhabitants) in voivodship cities in 2011.
Source: GUS, NBP. Source: GUS, NBP.
51 Warsaw Regional Branch, National Bank of Poland. 52 Białystok Regional Branch, National Bank of Poland.
58
Figure 113 Tree diagram of population structure (at pre-production, production or post-production age) in voivodship cities in 2010
Figure 114 Tree diagram of economic and demographic factors (unemployment rate and migration per 1000 inhabitants) in voivodship cities in 2010 and 2011
Source: GUS, NBP. Source GUS, NBP.
Figure 115 Tree diagram of the effects of housing construction (completed dwellings per 1000 inhabitants and per 1000 celebrated marriages) in voivodship cities in 2011
Figure 116 Tree diagram of quarter-on-quarter price growth in 2011 (sale transactions in voivodship cities – secondary market)
Source: GUS, NBP. Source: GUS, NBP.
Similarly to the previous years, the optimal solution is the breakdown of regional markets
into seven largest voivodship cities with population exceeding 400 thousand inhabitants (Gdańsk, Cracow, Łódź, Poznań, Szczecin, Warsaw, Wrocław) and into other centres, as in the chapter below. Analysis of the statistical data (including BaRN data) confirms that Poland is constantly divided into western and eastern part.
It should be emphasized that the analysis of demographic and statistical changes gives better results if conduced over a broader time range. Observation of appropriate trends over only two subsequent years does not necessarily lead to appropriate conclusions.
59
Housing situation in 16 voivodship cities In 2011 as compared to 2010, housing conditions in Poland’s voivodship cities improved
slightly (see Figure 117 – Figure 124). Similarly to the previous years, housing saturation and population density indices were better in the seven largest urban markets. This was the effect of several factors, namely, (1) intensified activity of real estate development companies, (2) newcomers, arriving at large academic centres and cities with lower unemployment rate and higher wages, not getting registered at any permanent address, and (3) population decline (e.g. Łódź with negative migration rate and negative population growth).
The years 2012-2013 are expected to see higher number of dwellings in the housing stock. This will be primarily driven by larger scale of real estate development construction and a small-scale process of demolitions and changes in the intended use of premises. Yet, it is likely that improvement in housing indicators (e.g. average usable housing area) slows down. This may be due to the trend observed in the local real estate markets, of supply being adjusted to the market conditions, i.e. real estate development companies reducing the number of investment projects with large usable housing areas to focus on compact dwellings53.
Figure 117 Housing stock per 1000 inhabitants in 7 cities
Figure 118 Housing stock per 1000 inhabitants in 9 cities
Source: GUS. Source: GUS.
53 Compact dwelling – housing unit with a small usable area and a big number of rooms.
300
320
340
360
380
400
420
440
460
480
500
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
mie
szka
nia/
1000
ludn
ości
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
300320340360380400420440460480500
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
mie
szka
nia
/10
00
lud
ności
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
60
Figure 119 Average usable housing area in the housing stock (square metres) in 7 cities
Figure 120 Average usable housing area in the housing stock (square metres) in 9 cities
Source: GUS. Source: GUS.
Figure121 Average usable housing area in the housing stock per 1 person in 7 cities
Figure 122 Average usable housing area in the housing stock per 1 person in 9 cities
Source: GUS. Source: GUS.
Figure 123 Average number of persons per dwelling in 7 cities
Figure 124 Average number of persons per dwelling in 9 cities
Source: GUS. Source: GUS.
50
52
54
56
58
60
62
64
66
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
m k
w/m
iesz
kan
ie
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
50
52
54
56
58
60
62
64
66
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
m k
w/m
iesz
kani
e
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
181920212223242526272829
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
m k
w/n
a os
obę
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
181920212223242526272829
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
m k
w/n
a os
obę
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
2,0
2,2
2,4
2,6
2,8
3,0
3,2
3,4
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
liczb
a os
ób
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
2,0
2,2
2,4
2,6
2,8
3,0
3,2
3,4
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
liczb
a os
ób
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
61
Demographic situation in 16 voivodship cities In 2011 demographic situation in voivodship cities deteriorated slightly as compared to
2010. Gradual decline was observed in fundamental demographic indicators which have had a favourable impact on housing market developments over the past few years. This was driven by the fact that the second post-war baby boom generations were starting to get on their own two feet. Consequently, such indicators as the number of celebrated marriages and demographic growth deteriorated in the majority of voivodship cities (see Figure 125 – Figure 126 and Figure 129 – Figure 130).
In 2011, similarly to the previous years, positive migration was boosting housing demand in Warsaw only, which had to do with Warsaw being a capital city and enjoying more favourable economic conditions (e.g. lower unemployment rate and higher wages). No common trend may be identified for the remaining six major cities and a group of smaller cities.
The latest demographic forecast for voivodship cities is not optimistic. According to GUS data, over the next few years only three voivodship cities, e.g. Cracow, Olsztyn and Warsaw will see their population rise. Population ageing process in likely to intensify in all the cities. This will be driven by continuing negative birth rates, which, combined with extending life expectancy will boost demographic burden in voivodship cities. For a few years now, all voivodship cities have seen growing number of post-production population and declining number of population at pre-production age; yet, the path of those changes has differed (see Figure 131 – Figure 132). In the future, this may result in lower housing demand.
Figure 125 Demographic growth per 1000 inhabitants in 7 cities
Figure 126 Demographic growth per 1000 inhabitants in 9 cities
Source: GUS. Source: GUS.
-7-6-5-4-3-2-101234
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
-7-6-5-4-3-2-101234
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
62
Figure 127 Migration per 1000 inhabitants in 7 cities
Figure 128 Migration per 1000 inhabitants in 9 cities
Source: GUS. Source: GUS.
Figure 129 Marriages per 1000 inhabitants in 7 cities
Figure 130 Marriages per 1000 inhabitants in 9 cities
Source: GUS. Source: GUS.
Figure 131 Ratio of population age change in 2010 (2002=100) in 7 cities;
Figure 132 Ratio of population age change in 2010 (2002=100) in 9 cities
violet – production age; green - pre-production age; pink - post-production age Source: GUS.
Source: GUS.
-7-6-5-4-3-2-10123456
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
-7-6-5-4-3-2-10123456
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
3,5
4,0
4,5
5,0
5,5
6,0
6,5
7,0
7,5
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
3,5
4,0
4,5
5,0
5,5
6,0
6,5
7,0
7,5
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
7580859095
100105110115120125130135
Gdańs
k
Krak
ów
Łódź
Pozn
ań
Szcz
ecin
War
szaw
a
Wro
cław
produkcyjnym przedprodukcyjnympoprodukcyjnym
7580859095
100105110115120125130135
Bia
łyst
ok
Byd
gosz
cz
Kat
owic
e
Kie
lce
Lubl
in
Ols
ztyn
Opo
le
Rze
szów
Ziel
ona
Gór
a
produkcyjnym przedprodukcyjnympoprodukcyjnym
63
Economic factors in 16 voivodship cities In the majority of Poland’s voivodship cities, the impact of economic factors on demand
for real estates was less favourable in 2011 as compared to the preceding year. All cities recorded wage growth in nominal terms, yet, accounting for CPI inflation, wages in real terms were higher than in 2010 in 11 cities only. This slight growth differed across cities and ranged from several to several dozen PLN only. Similarly to 2010, higher wages were observed in the largest regional centres, with the exception of Katowice where the highest wage level on the national scale was generated by wages in mining (see Figure 137 – Figure 138).
In 2011 also the situation in the labour market deteriorated. Higher unemployment rate was seen in almost all major regional market (with the exception of Wrocław) and in the group of smaller cities (with the exception of Rzeszów, Bydgoszcz, Kielce, Opole and Lublin, which in case of last four cities may be connected with demographic migration), (see Figure 133 – Figure 134). An optimistic side was a slight decline of the number of persons up to 34 years of age in the structure of the unemployed in 2011 in comparison to 2010 in both groups of cities. Despite the latest, the problem of more severe unemployment among the young was still actual in the smaller voivodship cities (see Figure 135 – Figure 136).
In 2011, a slight adjustment in home prices with simultanious average wage growth, led to a higher, as compared to 2010, availability of housing in all voivodship cities. Similarly to the previous years, due to relatively lower prices per square meter of housing in smaller voivodship cities, its availability was higher than in the seven largest cities (see Figure 139 – Figure 142). Moreover, as regards housing availability, Katowice could be set apart in the group of smaller cities as a one with a relatively high average wage level and relatively low home prices as compared to other cities.
As compared to 2010, all voivodship cities recorded a moderate decline in loan availability, mainly as a result of higher interest on PLN loans (see Figure 143 – Figure 144). As a result, credit availability of housing slightly declined, mainly in cities recording the lowest fall in home prices and slower average wage growth. In the case of larger cities, a more favourable value of this indicator was the result of stronger declines in home prices and higher average monthly wages in the enterprise sector (see Figure 145 – Figure 146).
In 2012, the level of loan availability will be dependent upon interest rates and bank margins. Should interest rates and bank margins increase, loan availability may be further limited within banks’ restrictive lending policy already put in place. The latest is driven, among other things, by the amendments to the S Recommendation implemented by the Financial Supervision Commission concerning creditworthiness procedures. Of significant importance will also be household income which should increase slightly.
In 2011 all the cities recorded growing demand for mortgage loan granted under the government-subsidized housing scheme called Family on their Own (RNS) which translated into growing demand for housing. Increased interest in the RNS was driven by the announced cuts in the scheme and further plans to withdraw completely from the scheme at the end of 2012. The number and value of those loans grew dynamically during the first three quarters of 2011, to fall in 2011 Q4 as a result of threshold housing prices limitation, (housing price threshold among others is the condition for getting interest subsidy). Despite above mentioned threshold prices limitations at the end of 2011, the number of households benefiting from the government subsidy exceeded the 2010 figure. As a result, the share of RNS loans in the total mortgage loans for households, increased (see Figure 149 – Figure 150). Due to lower level of price per one square meter of housing entitling to benefit from the RNS, in 2012 Q 1 similarly to the end of 2011, popularity of subsidized loans declined. Lower demand is expected to continue until 2012 Q3 and increases slightly at the end of 2012 as a result of final withdrawal from the scheme.
64
Comparison of the median transaction price and actual housing threshold price in the RNS confirms a large mismatch between those two numbers in voivodship cities, especially in the secondary market which is less flexible than the primary market (see Figure 151 – Figure 154).
Figure 133 Unemployment rate in 7 cities
Figure 134 Unemployment rate in 9 cities
Source: GUS. Source: GUS.
Figure 135 Percentage of the unemployed below 34 years of age in 7 cities
Figure 136 Percentage of the unemployed below 34 years of age in 9 cities
Source: GUS. Source: GUS.
0
3
6
9
12
15
18
21
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
%
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław Polska
0
3
6
9
12
15
18
21
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
%
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona GóraPolska
252831343740434649525558
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
%
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław Polska
252831343740434649525558
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
%
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona GóraPolska
65
Figure137 Average monthly wage in the enterprises sector in 7 cities
Figure 138 Average monthly wage in the enterprises sector in 9 cities
Source: GUS. Source: GUS.
Figure 139 Housing availability for an average wage in 7 cities – primary market
Figure 140 Housing availability for an average wage in 9 cities – primary market
Source: GUS. Source: GUS.
Figure 141 Housing availability for an average wage in 7 cities – secondary market
Figure 142 Housing availability for average wage in 9 cities – secondary market
Source: GUS, NBP. Source: GUS, NBP.
1500
2000
2500
3000
3500
4000
4500
5000
5500
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
zł/m
iesię
czni
e
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław Polska
1500
2000
2500
3000
3500
4000
4500
5000
5500
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
zł/m
iesię
czni
e
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona GóraPolska
0,30,40,50,60,70,80,91,01,11,2
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
m k
w/pr
zecię
tne
wyn
agro
dzen
ie
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
0,30,40,50,60,70,80,91,01,11,2
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
m k
w/p
rzec
iętne
wyn
agro
dzen
ie
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
0,2
0,4
0,6
0,8
1,0
1,2
1,4
1,6
1,8
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
m k
w/p
rzec
iętne
wyn
agro
dzen
ie
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
0,2
0,4
0,6
0,8
1,0
1,2
1,4
1,6
1,8
III kw
. 20
06
I kw
. 2
007
III kw
. 20
07
I kw
. 2
008
III kw
. 20
08
I kw
. 2
009
III kw
. 20
09
I kw
. 2
010
III kw
. 20
10
I kw
. 2
011
III kw
. 20
11
I kw
. 2
012
m k
w/p
rzecię
tne
wyn
agro
dze
nie
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
66
Figure 143 Availability of PLN loans in 7 cities
Figure 144 Availability of PLN loans in 9 cities
Source: GUS, NBP. Source: GUS, NBP.
Figure 145 Availability of loan-financed housing (PLN loan) in 7 cities
Figure 146 Availability of loan-financed housing (PLN loan) in 7 cities
Source: GUS, NBP. Source: GUS, NBP.
Figure 147 Current mortgage debt (in PLN million) in 7 cities
Figure 148 Current mortgage debt (in PLN million) in 9 cities
Source: BIK. Source: BIK.
707580859095
100105110115120125
IV k
w. 2
006
II k
w.
200
7
IV k
w. 2
007
II k
w.
200
8
IV k
w. 2
008
II k
w.
200
9
IV k
w. 2
009
II k
w.
201
0
IV k
w. 2
010
II k
w.
201
1
IV k
w. 2
011
tys.
zł/p
rzec
i ętne
wyn
agr
od
zeni
e
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
707580859095
100105110115120125
IV k
w. 2
006
II kw
. 200
7
IV k
w. 2
007
II kw
. 200
8
IV k
w. 2
008
II kw
. 200
9
IV k
w. 2
009
II kw
. 201
0
IV k
w. 2
010
II kw
. 201
1
IV k
w. 2
011
tys.
zł/p
rzec
iętne
wyn
agro
dzen
ie
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
30405060708090
100110120130
IV k
w. 2
006
II kw
. 200
7
IV k
w. 2
007
II kw
. 200
8
IV k
w. 2
008
II kw
. 200
9
IV k
w. 2
009
II kw
. 201
0
IV k
w. 2
010
II kw
. 201
1
IV k
w. 2
011
m k
w m
iesz
kani
a za
kre
dyt P
LN
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
30405060708090
100110120130
IV k
w. 2
006
II k
w. 2
007
IV k
w. 2
007
II k
w. 2
008
IV k
w. 2
008
II k
w. 2
009
IV k
w. 2
009
II k
w. 2
010
IV k
w. 2
010
II k
w. 2
011
IV k
w. 2
011
m k
w m
iesz
kani
a za
kre
dyt
PLN
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
01 0002 000
3 0004 0005 0006 000
7 0008 0009 000
10 000
11 000
2005 2006 2007 2008 2009 2010 2011
mln
zł
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
0
500
1 000
1 500
2 000
2 500
2005 2006 2007 2008 2009 2010 2011
mln
zł
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
67
Figure 149 Share of the government-subsidised (RNS) loans in the value of mortgage loans granted in 7 cities
Figure 150 Share of the government-subsidised (RNS) loans in in the value of mortgage loans granted in 9 cities
Source: BGK, BIK, NBP. Source: BGK, BIK, NBP.
Figure 151 Gap/surplus between RNS threshold prices and median transaction prices in 7 cities ( % of median transaction price) – primary market
Figure 152 Gap/surplus between RNS threshold prices and median transaction prices in 9 cities (% of median transaction price) – primary market
Source: BGK, NBP. Source: BGK, NBP.
Figure 153 Gap/surplus between RNS threshold prices and median transaction prices in 7 cities (% of median transaction price) – secondary market
Figure 154 Gap/surplus between RNS threshold prices and median transaction prices in 9 cities (% of median transaction price) – secondary market
Source: BGK, NBP. Source: BGK, NBP.
0
10
20
30
40
50
60
70
2007 2008 2009 2010 2011
%
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
0
10
20
30
40
50
60
70
2007 2008 2009 2010 2011
%
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
-80%
-60%
-40%
-20%
0%
20%
40%
60%
80%
I k
w.
2007
II k
w.
200
7
III
kw
. 20
07
IV k
w.
200
7
I k
w.
2008
II k
w.
200
8
III
kw
. 20
08
IV k
w.
200
8
I k
w.
2009
II k
w.
200
9
III
kw
. 20
09
IV k
w.
200
9
I k
w.
2010
II k
w.
201
0
III
kw
. 20
10
IV k
w.
201
0
I k
w.
2011
II k
w.
201
1
III
kw
. 20
11
IV k
w.
201
1
I k
w.
2012
Gdańsk Kraków ŁódźPoznań Warszawa WrocławSzczecin
-60%
-40%
-20%
0%
20%
40%
60%
80%
I k
w. 2
007
II k
w. 2
007
III
kw
. 20
07
IV k
w.
200
7
I k
w. 2
008
II k
w. 2
008
III
kw
. 20
08
IV k
w.
200
8
I k
w. 2
009
II k
w. 2
009
III
kw
. 20
09
IV k
w.
200
9
I k
w. 2
010
II k
w. 2
010
III
kw
. 20
10
IV k
w.
201
0
I k
w. 2
011
II k
w. 2
011
III
kw
. 20
11
IV k
w.
201
1
I k
w. 2
012
Białystok Bydgoszcz Katowice
Kielce Lublin Olsztyn
Opole Rzeszów Zielona Góra
-80%-60%-40%-20%
0%20%40%60%80%
100%120%
I k
w. 2
007
II k
w. 2
007
III
kw
. 20
07
IV k
w.
2007
I k
w. 2
008
II k
w. 2
008
III
kw
. 20
08
IV k
w.
2008
I k
w. 2
009
II k
w. 2
009
III
kw
. 20
09
IV k
w.
2009
I k
w. 2
010
II k
w. 2
010
III
kw
. 20
10
IV k
w.
2010
I k
w. 2
011
II k
w. 2
011
III
kw
. 20
11
IV k
w.
2011
I k
w. 2
012
Gdańsk Kraków Łódź
Poznań Szczecin Warszawa
Wrocław
-80%-60%-40%-20%
0%20%40%60%80%
100%120%
I k
w. 2
007
II k
w.
200
7
III
kw
. 20
07
IV k
w.
200
7
I k
w. 2
008
II k
w.
200
8
III
kw
. 20
08
IV k
w.
200
8
I k
w. 2
009
II k
w.
200
9
III
kw
. 20
09
IV k
w.
200
9
I k
w. 2
010
II k
w.
201
0
III
kw
. 20
10
IV k
w.
201
0
I k
w. 2
011
II k
w.
201
1
III
kw
. 20
11
IV k
w.
201
1
I k
w. 2
012
Białystok Bydgoszcz Katowice
Kielce Lublin Olsztyn
Opole Rzeszów Zielona Góra
68
Housing construction in 16 voivodship cities In 2011 similarly to the previous years, the growth rate of housing construction varied
across Poland’s voivodship cities and was strongly correlated with their specific demographic and economic situations. The behaviour of market participants on the supply and demand side was also impacted by changes in legal regulations. In 2011, in the majority of cities increased demand for primary market housing and persistently satisfactory rate of return urged developers to embark on more investments than in 2010. This phenomenon took place despite the existing oversupply of unsold housing units in the market. The developers wanted to put on sale as many housing units as possible before the entry into force of the Act on the protection of home buyers’ rights, whose vacatio legis ended in April 2012. Preliminary analysis of the January – April 2012 GUS data corroborated intensified activity of development companies, observed since 2011, as regards the number of new housing construction contracts in most regional markets.
In the majority of voivodship cities in 2011, as compared to the previous year, the performance of housing construction measured by the number of housing completions were slightly less pronounced and resulted from limited housing investment in 2009 on the back of the financial crisis. Yet, it is difficult to assess the performance of housing construction in particular markets in terms of usable area of completed housing. In the years 2010 – 2011 there were significant variations in the average area of completed housing in both city groups (see Figure 159 – Figure 160). Those differences were the result of the changing share of individual construction in the total completions. The individual construction features larger usable housing area as compared to construction intended for sale or rent.
The years 2012 – 2013, especially in the largest voivodship cities, are expected to see a decline in the average usable area of completed housing, as a result of developers shifting to smaller size housing. In terms of housing supply, the beginning of 2012 was promising in the majority of voivodship cities. Yet, the new legal regulations introduced by the Act on the protection of home buyers’ rights may have a negative impact on the volume of housing supply in the longer term. Moreover, also the structure of enterprises in the construction industry is likely to change. This will be driven by a more established position of large enterprises with stronger capacity to raise external financing as compared to smaller, often family-run companies. Consequently, particular markets may see consolidation processes or growing importance of partnerships in the execution of housing projects. In such a case, new investment may be made by a special purpose vehicle established by two or more companies. In such a special purpose vehicle the stronger partner assumes responsibility for the project itself, its financing and sales organization whereas the weaker partner contributes the land.
69
Figure 155 Number of completions per 1000 inhabitants in 7 cities
Figure 156 Number of completions per 1000 inhabitants in 9 cities
Source: GUS. Source: GUS.
Figure 157 Number of completions per 1000 concluded marriages in 7 cities
Figure 158 Number of completions per 1000 concluded marriages in 9 cities
Source: GUS. Source: GUS.
Figure 159 Average usable area of completed housing in 7 cities
Figure 160 Average usable area of completed housing in 9 cities
Source: GUS. Source: GUS.
0
2
4
6
8
10
12
1420
02
2003
2004
2005
2006
2007
2008
2009
2010
2011m
iesz
kani
a/10
00 lu
dności
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
0
2
4
6
8
10
12
14
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
mie
szka
nia/
1000
ludn
ości
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
0200400600800
1 0001 2001 4001 6001 8002 0002 2002 400
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
mie
szka
nia/
1000
małżeńs
tw
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
0200400600800
1 0001 2001 4001 6001 8002 0002 2002 400
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011m
iesz
kani
a/10
00 m
ał żeńs
tw
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
405060708090
100110120130140150160
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
m k
w/m
iesz
kani
e
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
405060708090
100110120130140150160
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
m k
w/m
iesz
kani
e
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
70
Figure 161 Housing construction in 7 cities
Source: GUS.
Figure 162 Housing construction in 9 cities
Source: GUS.
05 000
10 00015 00020 00025 00030 00035 00040 00045 00050 00055 00060 00065 00070 000
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Mieszkania, na budowę których wydanopozwolenia
Mieszkania, których budowę rozpoczęto
mies
zkan
ia
Gdańsk Kraków Łódź Poznań Szczecin Warszawa Wrocław
01 0002 0003 0004 0005 0006 0007 0008 0009 000
10 00011 00012 00013 00014 00015 000
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
Mieszkania, na budowę których wydanopozwolenia
Mieszkania, których budowę rozpoczęto
mie
szka
nia
Białystok Bydgoszcz Katowice Kielce LublinOlsztyn Opole Rzeszów Zielona Góra
71
Analysis of BaRN data Since the NBP started its monitoring of the housing market, the data base concerning both
asking and transaction prices has been steadily expanding, which translates into higher quality of presented analyses. In 2011, data samples continued to provide representative results for all the 16 analysed voivodship markets. Figure 163 Number of records in the BaRN data base
Source: NBP. The year 2011 saw home prices stabilize both in the primary and the secondary market
(see Figure 168 – Figure 171 and Figure 181 – Figure 184). In 2011, in the case of offers in the primary market, average price growth in y-o-y terms was recorded in 5 regional markets, whereas in the case of transactions price growth was seen in 12 regional markets. It should be emphasised that transaction prices posted a merely several percentage growth. In the secondary market the average asking price y-o-y growth in 2011 was observed in 7 cities, (maximum growth was of almost 4%), whereas the average transaction price growth was observed in 8 cities, (maximum growth reached of 5%, maximum drop reached more than 7%).
Analysis of the correlation between changes in the transaction price in the primary market and the volume of housing stock in a particular city did not confirm any such correlation existed. With all the cities included in the analysis, the correlation coefficient of – 0.28 was obtained. Yet, with Warsaw (the largest market) and Opole and Poznań (change in the data structure) left out of the analysis, the obtained correlation coefficient oscillated around 0 (see Figure 164 – Figure 165).
In the secondary market, transaction prices followed a downward trend - the stronger if the bigger a particular market is. The recorded correlation was higher with Warsaw left out of the analysis (correlation coefficient of 0.63 and – 0.42 respectively; see Figure 166 – Figure 167).
Both in the case of the primary and the secondary market, levels of average transaction prices (measured as an arithmetic mean for all the markets) were in 2011 close to those observed in 2010. All the markets saw transaction prices responding faster to external factors than asking prices.
The highest asking and transaction prices per square meter of usable area in both the primary and the secondary market have continued to be observed in Poland’s largest cities, with the Warsaw market leading the way. As regards the primary market, Warsaw sale prices were by approx. 600 PLN /square meter higher than those in Cracow, the second largest
0
5 000
10 000
15 000
20 000
25 000
30 000
35 000
40 000
45 000
I k
w. 2
006
II k
w. 2
006
III
kw
. 200
6
IV k
w. 2
006
I k
w. 2
007
II k
w. 2
007
III
kw
. 200
7IV
kw
. 200
7
I k
w. 2
008
II k
w. 2
008
III
kw
. 200
8IV
kw
. 200
8
I k
w. 2
009
II k
w. 2
009
III
kw
. 200
9
IV k
w. 2
009
I k
w. 2
010
II k
w. 2
010
III
kw
. 201
0
IV k
w. 2
010
I k
w. 2
011
II k
w. 2
011
III
kw
. 201
1
IV k
w. 2
011
I k
w. 2
012
Mieszkania oferty Najem oferty
Mieszkania transakcja Najem transakcja
72
polish city. In 2011 the difference between the two cities in the secondary market prices went up to 1 500 PLN /square meter. Smaller cities are more homogeneous in nature. Similar observations of the price levels and price growth in the local markets result from the analysis of the median of asking and transaction price in the primary and secondary market. The price analysis should also account for the city size, and, consequently, lesser price fluctuations in smaller cities. A large investment project in a smaller city may considerably affect the price change which is hardly observable in the group of larger cities.
Similarly to 2010, the wage level and the city size were the main factors affecting housing sale prices in the secondary market (see Figure 189 – table of correlations). Respective correlation coefficients were slightly lower than in 2010. The correlation coefficient between the price per one square meter and the city size was 0.82 whereas the correlation coefficient between the price and wage level was approx. 0.48. In the primary market, the corresponding correlation coefficients were at a similar level and stood at 0.89 and 0.52 respectively (see Figure 180 – table of correlations).
The residential real estate market in Poland continues to see tensions driven by demand and supply mismatch, in terms of average usable area of housing. The overwhelming majority of the markets see a downward trend in the size of newly constructed housing in response to clients demand (see Figure 176 – Figure 179). In terms of demand and supply adjustment, the secondary market is a much more rigid one (see Figure 190 – Figure 195). Discussed disproportion in house size is the result of too high prices.
In 2011 the average time on market for secondary housing unit offer equalled 139 days and was similar to the 2010 level. In Poland’s six largest cities (Gdańsk, Cracow, Łódź, Poznań, Warsaw and Wrocław), the time on market was 2 weeks longer then in 2010 and reached 150 days.
Figure 164 Percentage change in transaction price in the primary market in 2011 versus the city’s housing stock
Figure 165 Percentage change in transaction price in the secondary market in 2011 versus the city’s housing stock – exclusive of Warsaw, Opole and Poznań
Source: GUS, NBP. Source: GUS, NBP.
73
Figure 166 Percentage change in transaction price in the secondary market in 2011 versus the city’s housing stock
Figure 167 Percentage change in transaction price in the secondary market in 2011 versus the city’s housing stock – exclusive of Warsaw
Source: GUS, NBP. Source: GUS, NBP. Primary housing market in the BaRN data base
Figure 168 Year-on-year growth in asking prices in 7 cities – primary market
Figure 169 Year-on-year growth in asking prices in 9 cities – primary market
Source: NBP. Source: NBP.
Figure 170 Year-on-year growth in transaction prices in 7 cities – primary market
Figure 171 Year-on-year growth in transaction prices in 9 cities – primary market
Source: NBP. Source: NBP.
60
80
100
120
140
160
180
200
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
60
80
100
120
140
160
180
200II
I kw
. 200
7
I kw
. 200
8
III
kw. 2
008
I kw
. 200
9
III
kw. 2
009
I kw
. 201
0
III
kw. 2
010
I kw
. 201
1
III
kw. 2
011
I kw
. 201
2
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
6080
100120140160180200220
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
6080
100120140160180200220
III kw
. 20
07
I kw
. 200
8
III kw
. 20
08
I kw
. 200
9
III kw
. 20
09
I kw
. 201
0
III kw
. 20
10
I kw
. 201
1
III kw
. 20
11
I kw
. 201
2
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
74
Figure 172 Median offer price in 7 cities – primary market
Figure 173 Median sale price in 7 cities – primary market
Source: NBP. Source: NBP.
Figure 174 Median offer price in 9 cities – primary market
Figure 175 Median sale price in 9 cities – primary market
Source: NBP. Source: NBP.
Figure 176 Housing supply and demand; units with usable area less then 41 square meters – primary market in 7 cities
Figure 177 Housing supply and demand; units with usable area between 40 and 61 square meters – primary market in 7 cities
Source: NBP. Source: NBP.
2 000
4 000
6 000
8 000
10 000III
kw
. 200
6
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
zł/m
kw
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
2 000
4 000
6 000
8 000
10 000
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
zł/m
kw
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
2 000
4 000
6 000
8 000
10 000
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
zł/m
kw
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
2 000
4 000
6 000
8 000
10 000III
kw
. 200
6
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
zł/m
kw
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
-120%-100%-80%-60%-40%-20%
0%20%40%60%80%
100%120%140%
2006 2007 2008 2009 2010 2011 2012
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
-120%-100%-80%-60%-40%-20%
0%20%40%60%80%
100%120%140%
2006 2007 2008 2009 2010 2011 2012
%
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
75
Figure 178 Housing supply and demand; units with usable area between 60 and 81 square meters – primary market in 7 cities
Figure 179 Housing supply and demand; units with usable area bigger then 81 square meters – primary market in 7 cities
Source: NBP. Source: NBP. Figure 180 Correlation between the average transaction price in the primary market, average monthly wage in the enterprise sector and the city’s population (2011)
Source: NBP, GUS.
-150%
-100%
-50%
0%
50%
100%
150%
200%
250%
2006 2007 2008 2009 2010 2011 2012
%
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
-150%
-100%
-50%
0%
50%
100%
150%
200%
250%
2006 2007 2008 2009 2010 2011 2012
%
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
76
Secondary housing market according to BaRN data base
Figure 181 Year-on-year growth in asking prices in 7 cities – secondary market
Figure 182 Year-on-year growth in asking prices in 9 cities – secondary market
Source: NBP. Source: NBP.
Figure 183 Year-on-year growth in sale prices in 7 cities – secondary market
Figure 184 Year-on-year growth in sale prices in 9 cities – secondary market
Source: NBP. Source: NBP.
Figure 185 Median offer price in 7 cities – secondary market
Figure 186 Median sale price in 7 cities – secondary market
Source: NBP. Source: NBP.
80
100
120
140
160
180
200
220
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
80
100
120
140
160
180
200
220
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
80
100
120
140
160
180
200
220
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
80
100
120
140
160
180
200
220III
kw
. 200
7
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
1 000
3 000
5 000
7 000
9 000
11 000
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
zł/m
kw
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
1 000
3 000
5 000
7 000
9 000
11 000
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
zł/m
kw
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
77
Figure 187 Median offer price in 9 cities – secondary market
Figure 188 Median sale price 9 cities – secondary market
Source: NBP. Source: NBP.
Figure 189 Correlation between the average transaction price in the secondary market, average monthly wage in the enterprise sector and the city’s population (2011)
1 000
3 000
5 000
7 000
9 000
11 000III
kw
. 200
6
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
zł/m
kw
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
1 000
3 000
5 000
7 000
9 000
11 000
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
zł/m
kw
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
78
Figure 190 Housing supply and demand; units with usable area less than 40 square meters – secondary market in 7 cities
Figure 191 Housing supply and demand; units with usable area less than 40 square meters – secondary market in 9 cities
Source: NBP. Source: NBP.
Figure 192 Housing supply and demand; units with usable area between 40 and 60 square meters – secondary market in 7 cities
Figure 193 Housing supply and demand; units with usable area between 40 and 60 square meters – primary market in 9 cities
Source: NBP. Source: NBP.
Figure 194 Housing supply and demand; units with usable area between 60 and 80 square meters – secondary market in 7 cities
Figure 195 Housing supply and demand; units with usable area between 60 and 80 square meters – secondary market in 9 cities
Source: NBP. Source: NBP.
-100%
-80%
-60%
-40%
-20%
0%
20%
40%
2006 2007 2008 2009 2010 2011 2012
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
-100%
-80%
-60%
-40%
-20%
0%
20%
40%
2006 2007 2008 2009 2010 2011 2012
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
-100%
-80%
-60%
-40%
-20%
0%
20%
40%
60%
2006 2007 2008 2009 2010 2011 2012
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
-100%
-80%
-60%
-40%
-20%
0%
20%
40%
60%
80%
2006 2007 2008 2009 2010 2011 2012
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
-60%-40%-20%
0%20%40%60%80%
100%120%140%160%180%200%
2006 2007 2008 2009 2010 2011 2012
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
-60%-40%-20%
0%20%40%60%80%
100%120%140%160%180%200%
2006 2007 2008 2009 2010 2011 2012
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
79
Figure 196 Average selling time in 7 cities – secondary market
Figure 197 Average selling time in 9 cities – secondary market
Source: NBP. Source: NBP. Hedonic models of real estate markets – according to the BaRN data base Hedonic models of particular local markets constructed by the NBP Regional Branches
fail to clearly identify stabilization rate among markets54. An important price determinant is location which is not always accounted for in the regional models. There are plans to account for this factor55 more thoroughly by using Geographical Information Systems (GIS) to create more accurate variables describing housing unit location. These new variables should improve the quality of constructed models and limit unexpected outcomes. Characteristic feature of large urban centres is higher statistical importance of attributes related to housing location (mainly good communication). Location variables are less important in the case of smaller cities where it is difficult to distinguish between better and worse locations. In the secondary market, this relationship concerns both offer and sale price models. In smaller cities, attributes related to housing quality are more important in statistical terms. When distinguishing between offer and sale price models it should be emphasized that in the case of offer price models, the number of statistically important attributes averaged 19 (this relates to 8 variables which were different across Branches), whereas in sale price models analogous numbers were 15 and 7. The correlation coefficient between the housing stock in a particular city and the number of variables was at a low, negative level in the offer price models and amounted to zero in the case of sale price models. This shows that the market size dosen’t influence the number of price determining attributes in the hedonic models.
A comparison of models quality and local markets existing housing stock volume does not corroborate the opinion that larger markets are better developed and economic agents take more rational decisions. The Warsaw market is an outlier in the analysies. It is more than 2.5 as big at the second largest markets in terms of the housing stock in Poland - Łódź and Cracow (see Figure 198 – Figure 203).
54 Market stabilization is measured as the adjusted determination coefficient of a particular model. 55 The model variable may consist of several attributes (e.g. “housing quality” variable consists of the following
attributes: “low”, “average”, “high”).
050
100150200250300350400450
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
050
100150200250300350400450
III k
w. 2
006
I kw
. 200
7
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Białystok Bydgoszcz KatowiceKielce Lublin OlsztynOpole Rzeszów Zielona Góra
80
Figure 198 Adjusted coefficient of determination of offer price models and volume of the city’s housing stock (in thousands)
Figure 199 Adjusted coefficient of determination of offer price models and the volume of the city’s housing stock (in thousands) – without Warsaw
Source: NBP, GUS. Source: NBP, GUS. Figure 200 Adjusted coefficient of determination of sale price models and the volume of the city’s housing stock (in thousands)
Figure 201 Adjusted coefficient of determination of sale price models and the volume of the city’s housing stock (in thousands) – exclusive of Warsaw
Source: NBP, GUS. Source: NBP, GUS.
Figure 202 Tree diagram of offer price models (taking into account R2, number of variables and the housing stock)
Figure 203 Tree diagram of sale price models (taking into account R2, number of variables and the housing stock)
Source: NBP, GUS. Source: NBP, GUS.
81
Specification of asking price models is a much more complex task than of transaction price models. First of all, asking prices are often a bidding price, and to a large extent, a method of “testing” the market rather than a real price close to the market sentiment.
When discussing and comparing offer price models for the secondary market, it is important to mention that in the case of 5 cities there were no location variables in the model.
The correlation coefficient between the volume of the housing stock and the average assessment of location attribute was 0.65 (excluding Warsaw– 0.34). With outlying cities left out of the analysis (Gdańsk, Łódź, Warszawa, Wrocław) the correlation rose to 0.70 – see Figure 204 – Figure 206. Figure 204 Housing stock and the average assessment of location - all the cities
Figure 205 Housing stock and the average assessment of location – without Warsaw
Source: NBP, GUS. Źródło: NBP, GUS.
Figure 206 Housing stock and the average assessment of location –without outlying markets
Source: NBP, GUS. Gdańsk and Katowice were the only cities where housing age factor was not included in
the analysis. In all the other cities, this factor is statically important, yet with a varying weight. Nevertheless, in all the cities it is weakly correlated with the city’s housing stock. This coefficient rises somewhat when Warsaw is left out of the analysis. It should be highlighted that the fact that the analysis fails to account for the year of construction factor does not necessarily mean that such a factor has no impact on the price. The information of age may be indirectly represented by construction technology or house location (buildings are usually built within a specific period of time) – see Figure 207 – Figure 208.
82
Figure 207 Housing stock and the average assessment of the year of construction
Figure 208 Housing stock and the average assessment of the year of construction –without Warsaw
Source: NBP, GUS. Source: NBP, GUS. Sale price models in the secondary market are based on a relatively small samples, yet,
their quality is definitely better. Firstly home price is verified by the market, we can observe regional trends and there is less outliers in the sample.
As regards standard of the dwelling, and it’s correlation with housing stock in the city was negative. Again, Warsaw was an outlier but even without the capital, there was not seen significant correlation between average house standard and housing stock (see Figure 209 – Figure 210).
Figure 209 Housing stock and average assessment of the housing standard
Figure 210 Housing stock and average assessment of the housing standard – without Warsaw
Source: NBP, GUS. Source: NBP, GUS. Correlation between location variables and housing stock was 0,73. With Warsaw left out
of the analysis, the correlation dropped to 0,2 %. 5 small cities were not included in the analysis. (see Figure 211 – Figure 212).
83
Figure 211 Housing stock and average assessment of the location
Figure 212 Housing stock and average assessment of the location – without Warsaw
Source: NBP, GUS. Source: NBP, GUS. Information related to the year of construction were analysed in the case of 8 cities.
Cracow and Warsaw were outliers, which have an impact on the average correlation coefficient. Leaving the above two cities out of the analysis considerably increases the coefficient. (see Figure 213 – Figure 215).
Figure 213 Housing stock and average assessment of the year of construction
Figure 214 Housing stock and average assessment of the year of construction – without Warsaw
Source: NBP, GUS. Source: NBP, GUS.
Figure 215 Housing stock and average assessment of the year of construction –without Cracow and Warsaw
Source: NBP, GUS.
84
Housing rental market according to BaRN data base Figure 216 Growth in rental offer prices in 7 cities (y/y)
Figure 217 Year-on-y growth in rental offer prices in 9 cities
Source: NBP, GUS. Source: NBP, GUS.
Figure 218 Year-on-year growth in rental transaction prices in 7 cities
Figure 219 Year-on-year growth in rental transaction prices in 9 cities
Source: NBP, GUS. Source: NBP, GUS. Conclusions There are significant variations in the local housing markets in Poland in terms of
analysed prices. The largest markets are specific and they should be subject to a separate analysis. The analysis shows that division of the cities into 2 groups (group of the seven largest agglomerations and group of other voivodship cities) is reasonable. It should be also emphasized that smaller cities feature greater variations of results in characteristics assessment.
In 2011 (as compared to 2010) the majority of factors affecting the residential real estate market in Poland, deteriorated. Economic and demographic factors have an adverse effect on further assessment of the outlook for growth, and are additionally exacerbated by difficulties and limitations in real estate investment financing. The primary market, with greater price elasticity, has already embarked on the adjustment process. In the secondary market, on the other hand, less sought-after housing should get cheaper in the subsequent periods.
Changing consumer situation in the real estate market in 2011 did not considerably affect the assessment of housing situation in particular voivodship cities (see Figure 110 – Figure
50
70
90
110
130
150
170
190
III k
w.
2007
I kw
. 20
08
III k
w.
2008
I kw
. 20
09
III k
w.
2009
I kw
. 20
10
III k
w.
2010
I kw
. 20
11
III k
w.
2011
I kw
. 20
12
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
50
70
90
110
130
150
170
190
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Katowice Kielce LublinOlsztyn Opole RzeszówZielona Góra
50
70
90
110
130
150
170
190
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Gdańsk Kraków ŁódźPoznań Szczecin WarszawaWrocław
50
70
90
110
130
150
170
190
III k
w. 2
007
I kw
. 200
8
III k
w. 2
008
I kw
. 200
9
III k
w. 2
009
I kw
. 201
0
III k
w. 2
010
I kw
. 201
1
III k
w. 2
011
I kw
. 201
2
Bydgoszcz Katowice KielceLublin Olsztyn OpoleRzeszów Zielona Góra
85
112). The city of Katowice remains an unquestioned leader (high wages combined with relatively low prices) when analysing56 price per one square meter, population size and structure, unemployment rate, wages and housing availability. There are no major changes as far as the ranking leaders are concerned, with key role of the largest cities. Białystok is at the bottom end of the ranking, with growing distance to the other cities. The main factors exerting adverse effect on the Białystok market are high unemployment and low wages. No major changes were observed in the middle of the ranking, with the situation in particular cities getting more and more uniform – the ranking is getting dense.
Figure 220 Consumers’ situation in the residential market in voivodship cities in 2007
Figure 221 Consumers’ situation in the residential market in voivodship cities in 2010
Source: NBP, GUS. Source: NBP, GUS. Figure 222 Consumers’ situation in the residential market in voivodship cities in 2011
Source: NBP, GUS.
56 Analysis involved clustering with the use of multi-feature similarity (ranking establishment) and establishing linear hierarchy in terms of given variables and summing up their unitized values and dividing by the number of variables.
86
A2. Modelling of cycles in the residential real estate markets – interactions between the primary and the secondary market and multiplier effects.
Hanna Augustyniak 57, Jacek Łaszek 57,58 and Krzysztof Olszewski 57, with technical
support from Joanna Waszczuk 57
While analysing the housing market, we focus on the short-term modelling of the housing units market instead of analysing the long-term housing space market. In this context, even a minor change in factors affecting the real estate market has a multiplier effect leading to strong shocks on the demand side, and, consequently, to an excessive reaction of the supply side. Those shocks, depending on the price elasticity of supply and demand, may disappear or explode. This articles presents the modelling of cycles in the residential real estate market. We focus on price changes and the number of housing units in the primary and secondary market. We analyse the adjustments, which are necessary to make the residential market move towards an equilibrium.
1. Introduction The residential real estate market, similarly to other markets, shows cyclical variations
in prices and the number of constructed housing units. These changes are not random but driven by specific factors. One of those factors, which distinguishes the housing market from other markets, is its different behaviour in the short run and long run. In the long run, demand is determined by fundamental factors, and the supply side adjusts to the demand side. Supply adjustments take a long time and result from new construction or the depreciation of the housing stock. Moreover, the size of the housing stock determines the magnitude of the demand shocks. Strong short-term demand increases change the nature of the market, as short-term adjustments start to prevail. Demand shocks generate accelerator effects, because the current supply is only marginal as compared to the housing stock, while the demand shock concerns the whole housing stock. Additionally, the financial system and consumer behaviour, including speculative behaviour, have a pro-cyclical effect. As supply substantially lags the price impulse and the supply elasticity generally exceeds the demand elasticity, there are short-term tendencies to generate lasting cycles. If additionally some factors accumulate, the cycles may turn into a real estate crisis.
Real estate market models are well described in the literature. The best known one is the DiPasquale and Wheaton (1992) model, yet it focuses on housing space rather than on housing units, which makes it impossible to analyse cycles in this market. Cycles in the real estate market and their occurrence are described by, among others, Wheaton (1999), whereas the supply side is analysed by DiPasquale (1999). Further on, Hott and Jokipii (2012) show that housing market bubbles are largely affected by the persistence of low interest rates.
Housing as a capital good generates housing services, which can directly meet the owner’s needs, be the object of commercial activity or speculation. As a tangible fixed asset housing is subject to speculations based on expectations about its price growth in the future. According to Case and Schiller (2003), during the last boom buyers cherished too optimistic and unrealistic expectations about a further price growth. Yet, in historical terms, housing is
57 Economic Institute, National Bank of Poland. 58 Warsaw School of Economics.
87
also a consumer good which satisfies the owner’s housing needs, provides housing security and additionally, ensures a relatively safe, long-term investment of savings. Thus, housing is not only the housing space but also an object, which is determined by the stream of utility that it generates. Space is one of its features priced in the market and affecting the value of housing. It should also be added that housing is a heterogeneous good, not only from the point of view of the abovementioned functions it serves, but also in terms of its features, which are often differently evaluated in each of the analysed functions. Therefore, we adopt Rosen’s (1974) approach, defining housing as a heterogeneous good, whose value is determined by the sum of the assessment of its features.
Already King (1976), basing on Lancaster’s (1966) theory on heterogeneous goods, concluded that housing may be considered as a basket of goods generating a stream of services. In the case of housing, this stream depends mainly on its quality and location, which affects the consumers decision how much money should be spent on this basket. In order to find the analogy to the traditional consumer problem, we assume that the mortgage loan repayment is the price we pay for this stream of services. This is in line with Goodman (1988), who presented an analysis of housing demand, accounting for the hedonic value of housing, which considers housing as a good that generates a stream of services. He also accounted for the relation of rents to housing value in consumer decisions.
In our study we consider housing primarily as an asset, generating a stream of services for the owner. Such an approach enables us to analyse the housing market from a macroeconomic perspective, basing our analysis on microeconomic foundations. Only an analysis which rests on correct, realistic assumptions, makes it possible to interpret the market processes and provide useful guidelines for macroeconomic policy.
Despite housing heterogeneity, we can apply elements of the classical economic analysis that is used to analyse markets of homogenous goods (see Rosen (1974), King (1976) and Goodman (1988)). The assumption that the hedonic function applies to the Polish housing market was confirmed by empirical studies by Tomczyk and Widłak (2009). This allows us to take the market value of each housing unit to the level of an average, one-size housing, characterized by its market value, which is the aggregate sum of the assessment of its features and the expectations of the seller or buyer. Under the implicit markets theory, a home buyer chooses not only between housing and other goods, but also between particular features of housing. Analysing the equilibrium from the microeconomic perspective, we have to deal with multi-dimensional solids, which are then reduced to the two-dimensional space in the macroeconomic analysis.
The value added of our article is a well-established demand model with sound microfoundations. Additionally, we present a simple model of housing market cycles which reflects the observed phenomena. We provide a detailed description of the relations between the primary and the secondary market and discuss how, via the multiplier and accelerator effects, even apparently minor demand shocks may generate strong cycles.
The paper is organized as follows. Chapter 2 discusses the microeconomic foundations for the demand modelling. Chapter 3 is an introduction to the modelling of cycles in the housing market. Chapter 4 analyses the relations between the primary and the secondary markets, whereas Chapter 5 concludes the analysis.
2. Microeconomic foundations of macroeconomic relationships for the modelling of
demand in the housing market Models built on microeconomic foundations (see Heckman, 2011) form the basis for a
demand and supply analysis in the macroeconomic context, which allows economists to draw realistic, precise conclusions, which are an useful guide to monetary, fiscal and regulatory policies. The modelling of housing demand, which accounts for the shift from the
88
microeconomic to the macroeconomic perspective, is presented, inter alia, by Westaway (1992) and Pain and Westaway (1997). Our article focuses on particular segments of the market, which we then put together. There is a debate going on whether structural models attempting to analyse all parts of the economy should be used or whether the economy should be looked at from a bird’s eye view, focusing only on those components which are the object of the analysis (reduced form model). Heckman (2011), summarizing the debate concludes that well developed, partial models should be used, which enable an in-depth analysis of the reaction of a part of the economy to particular shocks. We follow suit and present a simplified economy of the housing market, consisting of housing demand and supply. In the entire article we analyse the number of housing units, because the mismatch between housing units desired and housing units available in the market is the measure and determinant of tensions.
The classical macroeconomic analysis focuses usually on a representative consumer who spends some of its income on a consumer goods basket. When analysing housing consumption we adopt a similar approach, which is as follows. A single household takes a decision to purchase housing, which may be considered as a basket of goods and services (referred to as H) and spends a part of its income59 on it. The home purchase decision can be explained using a decision tree model, as proposed by Kim (2010), where the home buyer’s decision is affected subsequently by the price of housing, its location and other features. Limitations of the human brain’s ability to process simultaneously a large set of information leads to taking of hierarchical decisions (see Kahn, Moore and Glazer 1987)60. It should be further emphasised that the decision to purchase a particular dwelling is influenced by both the social standing of the surrounding dwellings and its quality (Phe and Wakely, 2000).
The next step in our analysis is to move from decisions taken by a single household to the whole population of prospective home buyers and the number of housing units actually sold. We assume that one household can buy a large dwelling, another one a small one, and another one will not decide to purchase housing at all or will buy more than one housing unit. The household’s decision to purchase a particular housing unit, in a particular location may be treated as a discrete decision (see Anas, 1982, 1990). As there are many prospective buyers we can use the law of large numbers to move from the individual purchase probabilities to proportions in the whole population. Each household is assigned a vector of purchase probability of housing at a given price61, by which we get the home purchase probability of the whole population. Finally, multiplying this result by the number of persons in the market, we obtain the number of housing units that are demanded. This way, we move from individual demand to total demand in a particular market, which is measured by the number of housing units62. The supply side analysis works in a similar way, which we present in a simplified form.
59 In practice, this purchase decision is more complex. A household takes into consideration many features
such as the housing location, its look, number of rooms, etc. This issue may be solved with the use of the hedonic model (see Widłak and Tomczyk 2010).
60 Consumers’ hierarchical decisions are widely analysed in the marketing literature (see Kahn, Moore and Glazer 1987 and Fotheringham 1988). Kahn, Moore and Glazer (1987) analyse the problem using the example of a refreshment drink, taking into account various conditions influencing the purchase decision.
61 Based on his empirical analysis, Carter (2011) concluded that the home purchase probability depends on income and various household conditions.
62 Although dwellings differ in size, in the long-term perspective, the value of an average housing sold remains stable (see: Report on the situation in the residential and commercial real estate market in Poland, NBP 2012).
89
2.1 Demand side modelling This chapter presents a selective review of micro-founded models that are used for the
analytical description of the supply side in the housing market. We start with a simple model of optimal income allocation between housing consumption and consumption of other goods. An important and empirically justified assumption is that a household finances the home purchase through a mortgage. We assume that under fixed instalments, the annual cost borne by a household63 is the housing price Hp multiplied by the interest rate. This model calibration to the Warsaw market, which allow to analyse the effects of various shocks is presented in a complementary article (Augustyniak, Łaszek and Olszewski, 2012).
2.1.1 The simplest model – rule-of-thumb consumer
The analysis starts with a household which currently decides what part of its income to spend on housing and what part on other goods. The household utility results from housing consumption and consumption of other goods and additionally, from the excepted wealth growth as measured by the housing appreciation. In our model, the utility function has been selected in such a way as to analyse a realistic problem faced by consumers and moreover, to allow its analytical solution. The utility function takes the form of the CES function, whereas the parameter θ is the weight that a consumer attaches to the consumption of other goods and the parameter µ is used to set the elasticity of housing substitution with other goods.
���,�� � ��� �1 � �� ���� The substitution elasticity is calculated as � � ���. Further on, the parameter γ determines
how strong the future appreciation or depreciation of housing affects the consumer’s decision. The appreciation is calculated as the ratio of the next year's expected price to the current price � ������ . Housing appreciation was included in the utility function by, inter alia, Dunsky and
Follain (1997) and Sommervoll et al. (2010). An expected price increase has a positive impact on the house purchase decision, whereas housing depreciation has an adverse effect. The consumer has to obey the following budget constraint: � � ��� � In order to find the consumer's equilibrium and the optimal choice of housing consumption under a given interest rate, price and preferences, we solve his problem using the Lagrange equation. � � ��� �1 � �� �����+��� � ��� � �� In this way, we obtain the optimal substitution between housing consumption and consumption of other goods. �1 � ������� � � ���� In combination with the budget constraint, we obtain the demand function for housing and other goods under a given income, interest rate and housing price.
63 It can be assumed that home maintenance costs are proportional to the home size and its price. Because
whether they are included in the model or not does not drastically change the conclusions, which follow from the model, they will be ignored for simplicity reasons. Yet, as loans are usually repaid in fixed instalments, the interest payment is the dominant part of repayment in the first few years. Therefore, the model includes, as housing costs, the interest rate part of the loan repayment, which is directly affected by interest rates, and in the case of foreign currency denominated loans, exchange rate fluctuations.
90
�∗ � �1 �� �1 � �� �� 1 ��
���
�∗ � ��� �1 � �� �� �� ���
In addition to the consumer income and substitution effects, the increase or decline in the number of households eligible for a loan, according to prudential regulations, will affect the demand. The income elasticity of demand is approximately 1 (see, inter alia, Lin and Lin 1999), which means that a rise in income, or loan affordability results in a proportional growth in housing demand.
2.1.2 The permanent-income-hypothesis consumer The previous model can be easily extended to account for decisions in the life cycle,
which we do basing on the Aoki, Proudman and Vlieghe (2002) model. As our aim is to provide a simple and analytical description of cycles in the housing market, we only make use of selected parts of their detailed and complex model. Specifically, we focus on the permanent-income-hypothesis consumer. This consumer owns always one dwelling, yet, may increase or decrease his housing expenses in the future. Moreover, this consumer can save now (St) to increase his consumption in the future (Ct+1). His aim is to maximize his utility during the whole life cycle. In order to account for the consumer’s life-cycle decision, we should remember that utility varies over the time horizon, and needs to be discounted with the parameter β<1. The time index t denotes in this model longer periods of time, not necessarily single years. For example, a young couple with a limited budget will first buy a small dwelling. Yet, as children are born and the couple’s income rises, it decides to sell the old dwelling and buys a new, bigger one. The household’s problem can be written as follows:
max ���,�� � ∑ !"!#$ ���% �1 � ���%���
under the intra-temporal budget constraints �! � &! �!�!�! (! and �!)� �1 �!�(! � &!)� �!)��!)��!)� from which follows: �!)� �1 �!���!�&! � �!�!�!� � &!)� � �!)��!)��!)� � 0 As above, this problem may be solved using the Lagrange equation:
� � ∑ !"!#$ ���% �1 � ���%���+���!)� �1 �!���!�&! � �!�!�!� � &!)� ��!)��!)��!)� and we obtain the optimal inter-temporal substitution:
�!�� � �!�!�1 �!��!)��� �!)��!)� �!�� � �1 �!��!)��� and also the optimal intra-temporal substitution: �1 � ���!�� 1�!�! � ��!��
91
�1 � ���!)��� 1�!)��!)� � ��!)���
Solving the household problem, we obtain the optimal inter and intra-temporal choice of the household.
2.1.3 Demographics-based model Housing demand is also affected by such fundamental factors as demography or
migrations. Demographical changes shift the aggregate demand curve, increasing or decreasing the number of households willing to purchase housing. Based on the function developed by Waldron and Zampolli (2010), the augmented housing demand function, accounting for the number of persons in a household N looks as follows:
���, �,+� � ,��- �� �.,� �1 � �� �/,��0�.
As a result of such extension of the utility function, housing demand will grow along with a bigger number of household members.
2.2 Optimal allocation with a kinked budget line
All the models presented so far assumed that a household may take out any loan, provided it meets its budget constraints. Yet, as banks impose certain prudential restrictions on the borrower, the amount of available loan may be considerably reduced. This situation concerns practically all countries, in particular, fast developing emerging economies where the housing stock is rather small with respect to income and there is a strong need for mortgage-financed homeownership. Given prudential regulations, a household may spend only a part of its income on loan repayment: �/ � 1� 2 �, 1 ∈ �0,1�. Thus the budget line is kinked and two cases of consumer decisions on housing expenditure should be considered.
� � 4�∗, ���∗ 2 1�1��� , ���∗ 5 1�
The kinked budget line has also an evident impact on the optimal demand for other goods, which takes the following form: � � 6 �∗, ���∗ 2 1��1 � 1��, ���∗ 5 1�
Provided the optimum point is unavailable due to lending restrictions, the consumer will have to adjust its consumption accordingly - it will consume less housing and more other goods than it would like to. This, in turn, leads to non-linear demand shocks. Should interest rates fall considerably, the mortgage-financed loan availability would rise. As suggested by the data on the Polish market (see NBP (2011) and (2012)) as well as the analyses conducted by Brzoza-Brzezina, Chmielewski and Niedźwiedzińska (2010) for Central and Eastern European countries, households easily substituted domestic loans bearing high interest with foreign currency denominated loans bearing a lower interest rate, however failing to account for the high FX risk.
2.3 The impact of the credit channel on the situation in the real estate market Taking into account the previously discussed demand models, we present how demand
is affected by changes in income, prices, interest rates or prudential regulations. We put together the previously described consumer reactions and present how factors such as the interest rate or prudential regulations affect consumer decisions. The financial sector affects the real economy through the credit channel, where the basic parameter affecting the sector’s
92
demand are interest rates and related real and alternative costs of capital as well as bank’s prudential requirements, including the initial down-payment.
The value of capital, namely housing, is transformed through the interest rate into a monthly stream of payments. It constitutes the financial cost of home ownership in the form of loan repayment or alternative costs of occupier owned housing. Basing on the market price and their own preferences, households choose a basket of goods. If housing is considered as a capital generating services, its price is the monthly ownership cost (interest on the mortgage or alternative cost of capital) as well as maintenance costs (Figure 1).
Figure 1. The effect of the financial sector
A change in interest rates reduces consumption through the substitution and income
effect. The substitution effect includes both a growing demand for substitutes, namely rented housing (left out to make things simple) and changes in the consumption structure towards a higher share of other consumer goods. The income effect, on the other hand, will limit or increase housing consumption.
The home purchase decision is affected by interest rates and the required down-payment. Already Burkham (1972) quotes the findings of the Fed’s analysis, which demonstrated that mortgage supply is one of the most important, if not the key factor affecting home construction. This relationship still holds (Aoki et al., 2002 and Levin and Pryce, 2009). Apart from checking the creditworthiness of the buyer, the bank requires an initial down-payment. The empirical analysis by Jaffe and Stein (1979) and Stein (1995) shows that the amount of down-payment greatly impacts both home prices and subsequent decisions to change housing. According to Ortola-Magne and Rady (2006) the down-payment is of great importance for young households who buy their first housing. Further on, Rubaszek (2012) analysed the restrictive impact of the down-payment requirement on the welfare of households in Poland. He demonstrated that a higher down-payment requirement considerably delays the home purchase, which translates into delayed home changes and reduces the household’s utility during the life cycle. While the above mentioned studies analyse the home purchase in terms of the life cycle, we account for the down-payment in the budget constraint.
Hou
sing
ava
ilabi
lity
H2
Household income Consumption of other goods
H1
93
Figure 2a Loan amount and demand for housing amidst consumer’s free choice and banks’ prudential regulations
Figure 2b. Consumer expansion path amidst banks’ prudential restrictions
Figure 2c Mortgage loans and housing demand amidst growing interest rises and easing of banks’ prudential restrictions
Figure 2d Mortgage loans and housing demand amidst growing home prices and impact of the wealth effect
In practice, housing demand is also affected by other financial parameters, which are
not based on the interest rate, such as prudential regulations and quantitative limits routinely applied by banks as well as by those used additionally in the situation of growing risk. These limits lead to a kinked budget line and shift the equilibrium point, reducing housing consumption even further (Figure 2a). Yet, it should be noted that amidst strong housing preferences (when the utility function is strongly inclined towards housing consumption) and banks’ prudential restrictions preventing consumers from reaching their optimum, housing demand will rise along with loan availability (Figure 2b). With the normal budget line, rising income translates proportionally into housing demand. Yet, with a kinked budget line, the consumer has a suboptimal allocation of housing consumption and a rise in income leads to a surge in housing demand, thus generating demand shocks. The consumer does not only spend this additional income on housing, but moreover can give up some consumption of other goods to spend even more on housing.
This phenomenon accounts for the fact that lending follows aggregate loan availability and mortgage-financed housing availability, a process observed for many year in the Polish market (see Łaszek, Augustyniak and Widłak (2009) and NBP (2012)). A rising demand brings mainly price effects as demand for housing is rigid in the short-term. If along with
H1
H0
H
C
B'
A
B
H
C
H
C
A
B
H
C
Budget constraint
Indifference curve
Kinked budget line
94
rising home prices, banks ease their loan restrictions, housing demand may remain stable or even grow until it reaches the consumer’s equilibrium point (the consumer will choose the allocation C rather than B, Figure 2c). The situation will be similar in the case of rising interest rates, which have the same effect as home prices, boosting housing service costs.
The described relations concern buyers of new housing who will be affected by home price increases through the rise in the amount of cash and lending necessary to finance housing. In the case of home buyers who are already home owners, a further price growth should urge them, through the substitution effect, to attempt to capitalize on growth in value and replace their housing with a smaller, lower-priced housing. Consequently, a growing supply should improve the situation in the market. However, high transactions costs64 in this market and consumer habits are factors curbing this phenomenon. The home owner usually agrees to the change, if the additional profit or utility of new housing significantly exceeds the above mentioned costs.
Yet, even very high price increases do not have to be indicative of a massive home sales phenomenon. A change in the value of housing means also a perceived change in consumption of housing. Under such conditions the consumer will shift its preferences towards housing. Consequently, housing demand will be maintained at the current level (the home owner will not sell the higher-priced housing). His preferences should change in such a way, that the substitution effect of the rising home price (reduction of housing consumption and boosting consumption of other goods) is offset with the income effect (income growth results in consumption of higher-priced housing, Figure 2d). In Poland, the boom period brought a rise in the volume of transactions in the secondary market, thus we may have observed both types of behaviour in the market.
2.4 Moving from individual demand to aggregate demand When aggregating the behaviour of individual consumers, we obtain a function of
demand for housing units which, at a given price, is inversely proportional to the interest rate. Multiplying the housing demand of an individual household Ht
* by the number of households in the economy N, we obtain the aggregate demand for the housing stock: Dt=H t
*Nt. In the long-term equilibrium, the number of housing units sought-for will equal the long-term size of housing stock, called St.
2.5 The supply side and price adjustments After having examined the demand side, we analyse the level of housing stock as well
as the supply and price reactions to demand changes. The stock St of housing units occupied and available in the market consists of the depreciated stock from the previous period (d is the depreciation rate), which is restored through new housing construction It (see: Sommervoll et al., 2010). (! � (!���1 � 7� 8!
64 A home change entails costs – both direct and indirect financial and moral costs. Financial costs may
account for up to 15% of the housing price and they include transaction costs resulting from various fees. Moreover, upon moving, the cost may be increased due to rents that are paid during the home refurbishment period. Indirect costs include home refurbishment costs of the sold house, which may be lost as the new owner may have totally different preferences and may want to renovate the dwelling again. We should also add non-financial costs resulting from being accustomed to neighbours and the residence area.
95
In the equilibrium, the production of new housings equals its depreciation, thus the housing stock remains constant. Moreover, the stock St equals demand for housing Dt. If, on the other hand, for reasons mentioned above, demand for housing increases to exceed housing supply, prices start to rise. The price adjustment, which results from the demand and supply mismatch, can be simply described by using the following formula (see Tse, Ho and Ganesan 1999): ∆:! � ;�<! � (!�, where ; determines the price response elasticity to the mismatch (it may be asymmetrical downwards and upwards). As a result of the price growth, real estate developers increase the housing production. A very important fact for the price adjustment is that the demand shock concerns the whole housing stock, while new housing production concerns its marginal part only. The relationship between new housing construction and the housing stock may be described as k=I t/St, a parameter which usually has a value of several percentage points. This multiplier, which we call the fundamental multiplier, causes that even a minor change in housing stock demand generates a shock to the demand for new housing production. This results in a huge jump in prices and urges developers to increase production. Real estate developers often extrapolate the historical price increase assuming that if prices are on the rise this year they will also increase in the future. Their production function depends on the previously observed rises in prices and surges in production costs. The real estate developer usually puts a pre-sale65 contract on sale when the housing has been under construction for two years and its completion is scheduled in approximately two years. At times of very high demand and strong price increases, even contracts for newly commenced investment projects, the so-called holes in the ground get sold. We modify the housing production function proposed by Tse, Ho and Ganesan (1999), adjusting it to empirical observations. Housing production consists of real estate development production, including its autonomous part66 and the part that depends on the lagged price change (∆�!� and lagged construction costs�!. 8! � =$ =�∆�!�> =>∆�!�� �∝@ �!�> Substituting the number of housing units newly built by developers into the previously discussed housing stock equation, we obtain the motion of the housing stock: (! � ($ =�∆�!�> =>∆�!�� �∝@ �!�> This simple model explains the occurrence of cycles in the housing market. Let us assume that interest rates are on the decline, which along with a kinked budget line will result in a leap jump in housing demand. In this way, falling interest rates will boost capital flows from bank accounts to the owner occupied housing sector (OOH, hereinafter). These phenomena have been shown empirically by Levin and Pryce (2009). Under short-term fixed supply prices start to grow quickly. Real estate developers monitor the price increases, calculate potential profits and embark on new constructions. After approximately two years, they put pre-sale construction contracts on sale. As they are unable to precisely estimate future demand and do not know exactly the volume of competitive construction, at some point, they start to build too many housing units. When supply is considerably in excess of demand, prices start to fall and the downward phase of the housing cycle starts. Real estate developers notice that home sale slows down and additionally, amidst falling prices and
65 Pre-sales are not very usual in Western Europe, however they are common in Poland and are used for a long time in Asia (see Chang and Ward 1993). 66 With high, fixed costs, real estate developers build some housing as a reserve, failing to account for current prices and volatile costs. Such a production is called autonomous production.
96
generally rising costs, their profit margin falls. Developers limit new housing production waiting for prices to embark on an upward trend and for a new housing cycle to begin. 3. Introduction to the modelling of housing market cycles Basing on the previously presented, microeconomic foundation of the demand and supply side behaviour, we now present the interactions at the macroeconomic scale. Basic models of real estate markets usually deal with real estate for rental. They base on the model where the real estate stock generates services, and related rents are valued in the capital market which determines real estate prices (Figure 3). In such a model, demand shocks in the short run (a shift from p1 to p2) lead only to price growth through the imputed rent channel67 (a shift from c1 to c2), as the stock, and consequently, supply are constant. The high prices and extraordinary profits, not illustrated in the model, translate, in the long run, into output growth, and further into capital inflows to the sector generating extraordinary profits and supply growth. Growing supply leads to the decline in prices from p2 to p1, fall in rates of return and finally the return to the long-run equilibrium. Therefore, it is possible to assume in the model that supply is rigid in the short run and perfectly flexible in the long run as construction effects accumulate over the years. Supply depends on the market price and construction costs68.
In the case of residential real estate the creation of supply is generally very limited and any changes in demand translate into construction demand. It should be mentioned that when discussing supply adjustments, namely adjustments of the size of the housing stock, given relatively small annual stock increases (1-3%), we mean a perspective of several years or even decades and a similar length of supply cycles. As demand is cyclical and volatile, supply does not match demand. Yet, there have been cases when, especially with the government’s intervention, long-term economic growth has been accompanied by a large, long-run supply of new housing investment projects. Therefore, in the case of residential real estate, the supply adjustment model is generally a very long run equilibrium model. The supply adjustment presented in Figure 3 with a dotted line, being a response to the demand shock, shows the supply after its nearly 20 year rapid growth, which reveals the scale of the discussed phenomenon. Taking the considerably high volatility of demand into consideration, it may be concluded that the market will only seek to reach the equilibrium, usually failing to achieve it. Effects similar to demand growth can be caused by falling interest rates. Those boost prices to the p3 level, at a given rent level, through the falling discount rate which is presented in Figure 3 by moving the slope of line i1 to position i2. Downward adjustments are much more difficult as they result from stock depreciation, which is usually inferior to the size of new construction. In the case of major structural mismatches, the adjustment may take up to several decades. In this model, factors affecting demand are fundamental factors such as interest rate, income, demographic factors, migration and other factors impacting preferences, whereas the supply is determined by long-term production costs (the intersection of marginal costs and average costs).
67 Imputed rents are rents that the home owner would have to pay for a similar housing if he wanted to rent it as a consumer. Therefore, the value of the housing may be expressed as follows: ABCDE � �EF%1 � �EF%�1 ��> ⋯ �EF%�1 ��! 68 The price elasticity of supply and its estimated values for various countries are presented by Phang, Kim and Wachter (2010). Moreover, the investment in and construction of new residential real estate were presented in theory and estimated by Topel and Rosen (1988).
97
Figure 3. Model of the housing space market and the capital market, short and long term
The interactions in the real estate market can be extended by the real estate
development sector which provides new objects and also the depreciation of real properties and their replacement. Such interactions are described by the well-known DiPasquale and Wheaton (1992) model (DPW, hereafter). In this model, the key point is the relationship between the rent level of a real estate and its price, and, consequently, the volume of production or housing stock depreciation and the equilibrium between demand for housing space and its supply.
The housing space market in such a model is in a long-term equilibrium when the demand for housing space translates into a rent which ensures that the housing price, understood as discounted rents, will stand at the level of long-term production costs and will guarantee a production level offsetting depreciation. The real estate development part of the model is able explain the excessive optimism of developers, being one of the cycle driving forces. In the case of housing, similarly to the previous model, it is rather a long-run equilibrium model, where fundamental variables play the main role for this market.
It should be noted that the home owner as a consumer will be in equilibrium if imputed rents equal interest rate payments. This means that the ownership cost equals gross rent, and consequently, there is no arbitrage and the rental housing stock and owned housing stock will be in equilibrium. If ownership costs exceed the imputed rent, the owner will not be in equilibrium as a real estate developer, and, in the opposite case, as a consumer.
In the DPW model (see Figure 4) price bubbles and speculative behaviour can be explained with the discrepancy in housing valuation based on the discounted rent and its current price. The relations between flexible demand and rigid supply of housing space are shown in the first quarter of the figure. The second quarter of the system of coordinates shows the capital market, where the point S is the actual price. Tensions arising in the housing space market, namely the difference between the market rent or imputed rent, and the cost of using the owner occupied housing, denoted as point T, are the consequence of a price bubble. The real estate development market, namely the housing space production in response to price changes, is presented in the third quarter. Finally, the fourth quarter shows the relationship between the production and depreciation of the housing space.
Housing stock
Price of sq. m
rent
Property Market: Rent Determination
C2
p1
p1
p2
p2
p3
Long-term supply = Long-term cost
i2
i1
Capital market
C1
98
Figure 4. The DiPasquale-Wheaton model
However, this model does not present a simple explanation of the formation, absorption and bursting of price bubbles in the OOH market in the short run. This results from the fact that the discussed DPW model and its adaptations constitute fundamental equilibrium models in the housing space market rather than short-term speculation and imbalance models in the market for real estate housing units. The so interpreted classical model or the DPW model will be very close to the OOH market model, however it lacks the short-term price development mechanism, and as a result, price bubble formation. This model does not describe the owner-occupied housing market69 in the short and medium term well enough as it does not analyse housing units and related speculation. The stock adjustment in the housing market may take decades. In order to analyse the disequilibrium in the market, we propose a housing model, which is focused on owner occupied housing units and a short period of time.
69 In the case of commercial real estate for rental, the DiPasquale and Wheaton model explains very well the
current, short- and medium-term deviations from the equilibrium. Those are driven by short-term changes of demand for space and, consequently, changes in rents and prices, interest rate changes and their translation into real estate development.
Ren
t
sq. m Price / sq. m
T
Sup
ply
S
Asset Market: Construction
Property Market: Stock Adjustment
I II
III IV
Property Market: Rent DeterminationAsset Market:
Valuation
sq. m
99
3.1 The owner occupied housing model The DPW model can be relatively easily augmented from the rental model to the owner
occupied housing model (OOH). It is enough to apply the imputed rent instead of the usual rent. Although the DPW model did not account for the credit channel, it can be easily included with the use of the relationship between the housing market and the capital market. In this perspective, the capital market provides capital that is transformed into housing, and further on, through the interest rate, it transforms the housing stock into a stream of payments borne by the home owner.
The starting point for the new model is the fact that the housing market is in disequilibrium and the equilibrium state is more an exception rather than a rule. This is the result of a rigid short-term housing supply, which becomes flexible with a time lag, the volatile demand, its relationship with the financial market and finally, speculations. The housing market model based on the above assumptions is presented in Figure 5.
Figure 5. The model of the OOH market
The initial OOH model is a local market model and, as an equilibrium model, it focuses on the long-term perspective. Its four parts can be illustrated with a system of coordinates. The first quarter is the housing market, represented by the housing stock that is used to generate a stream of utility. The second quarter is the market of financial capital which flows to the housing sector if rate of return is sufficient high. The third quarter is the real estate development and construction market which transforms financial capital into real capital, i.e. housing. The fourth quarter represents the stock depreciation and reconstruction, finally affecting the stock level in the first quarter.
The real capital market, i.e. the housing units market, is in its long-term equilibrium when the current, commercial and available supply intersects its alternative uses, setting the price per unit of capital, its rental cost and the number of vacancies at such a level that the related real estate development production offsets the stock depreciation. In this situation, enterprises
Price / Return on investment
Supply of financial capital Costs of
entry
sq. m
restitution
Long-term reproduction cost
C B
A
Asset Market Valuation
Property Market: Rent Determination
Asset Market: Construction
Property Market: Stock Adjustment
I II
III IV
Nr of housing units
100
are no longer motivated to enter the real estate development sector. A demand shock boosts prices as supply is almost rigid. The price growth leads, through the financial market, to supply growth.
3.2 Housing stock supply elasticity in relation to time
The housing market works in the following way. In the long run, the housing stock supply gets flexible with time, provided that new housing construction offsets depreciation, namely if prices offset long-run production costs. The longer the period, the larger the aggregate supply of the new housing stock and the higher its elasticity (see Figure 6). In the short term, the supply consists of housing units from the existing housing stock that are put on sale and rigid construction that was planned in the previous period. In the medium term, the supply of housing units from the existing stock will remain constant, but firms may plan in advance larger production volumes; consequently, the supply will be determined by the size of the existing, marketed housing stock and marginal costs of the real estate development sector.
In the long term, new capital may flow into the sector, boosting its production capacity and setting costs at the average cost level. The longer the time horizon, the more elastic becomes the supply, thus the supply curve is getting flatter and flatter.
Figure 6. Short- and long-term equilibrium in the market for housing units
Basing on our own observations, we assume that in the long term the market will trigger mechanisms that will offset the supply and demand mismatch in the local markets through construction and stock depreciation. As supply changes are calculated as insignificant percentage points of the housing stock whereas demand changes are considerably larger, these adjustments may take decades and are generally unlikely to result in an equilibrium. To a certain extent, we also have to do with adjustments through the competition of markets attempting to solicit investors and attract demand. As a result, local markets in terms of stock and current changes in demand will always be somewhat unbalanced.
The so defined long-term equilibrium or rather deviations therefrom, may be measured indirectly by monitoring housing situation indicators, especially the number of dwellings per 1000 inhabitants in relation to income of population inhabiting a particular region or its GDP.
In order to analyse the functioning of the housing units market in the short term, the analysed model and related assumptions need to be further modified. The assumption of a free housing stock allocation to alternative uses (e.g. rental, change into commercial use) is quite problematic. Further on, we should highlight the differences in the long- and short-term
Offers
Pric
e
Nr of housing units
101
behaviour of the market. The housing stock is highly diversified both within a particular country as well as across countries. Moreover, as a result of changes in the pursued housing policy they get further diversified with time. In most countries the owner occupied housing dominates the housing stock; yet, in many countries also rental housing, including municipal housing or mixed ownership housing have a large share in the housing stock and strongly affect the whole sector.
The rental housing market, including also the professional one, is subject to tenant protection rights, providing different degrees of protection and limiting the possibility of changing the contract during the rental period. In the majority of regulated housing stock, including the private housing stock, a change of use is not possible at all. What is more, shocks driven by changes in housing policy (e.g. privatization of the municipal housing stock) is often an important driver of the cycle.
The supply in the short-term model is different than the one in the long-term model, both as regards new construction and the existing stock where in lieu of housing stock there are dwellings put on sale. In the short- and medium-term period, supply may be defined in a classical way, i.e. as the number of housing units that households are ready to put on sale at a given place, time and price and also those that have been put on the market by real estate developers at a given place, time and price, taking into account their price in time t-1, thus pt-1. The so understood supply will consist of two components: housing put on sale from the existing housing stock and new real estate development construction70.
3.2.1 Supply in the secondary market In the short term, the supply in the housing units market is rather rigid as it cannot be
increased by new construction. However, supply may be increased from the existing stock as a result of growing real estate prices. Growing housing prices should urge households to change their existing dwelling into a smaller one or hasten their decision to sell the dwelling, should the substitution effect outweigh the income effect. Yet, transaction costs or the fact of housing being considered a consumer good will be strong enough to finally put an end to this trend, as empirical evidence shows. The minimum production costs do not necessarily have to be the bottom end of the supply curve as historically incurred costs are treated as sunk costs. Supply in the short- and long-term period will probably be not very flexible; it will get flexible in the long run only through changes in the use of housing units, large-scale migration as well as owners’ deaths.
3.2.2 Supply in the primary market The supply of real estate developer housing can be illustrated by two curves. In the short
term it will be rigid on the back of the period of 1-4 years necessary to launch and commercialize the construction project. Yet, developers do have certain price expectations
70 In practice, the supply curve not only lacks the character of a continuous line but may turn out to consist
of several lines indicating stock available for alternative use in various time horizons or even several sub-markets separate from one another, presenting a different scale and form of interaction. In order to solve this problem, the model should be limited to the stock available in a given period for alternative commercial use. This stock will generally consist of market rental housing and the OOH market as in the first market contracts may be concluded on a rather fairly basis, especially without excessive time limitations whereas the other housing stock may be converted into rental housing or non-housing stock. In the case of restrictions in the rental market and generally related shortage of this kind of housing, the best idea is the assumption that only the OOH stock is analysed.
102
and are usually prepared to wait for a client that will pay the expected price. Frequently, when real estate developers finance housing investment with bank loans, they just cannot lower the price to suit market conditions. Although the number of new constructions is constant, their real supply will be inclined and there will always be some housing stock unsold. Supply in the medium term will be flexible as real estate developers will start selling pre-sale contracts for housing provided this is permitted by the law and accepted by the market. Generally two years after commencing construction, developers put pre-sale construction contracts on sale. During a price boom, this period is shortened to one year, and the buyer purchases a so-called hole in the ground. In the medium term (i.e. after 3-5 years) new completed housing units are delivered, which may not affect the market if this is a contract market rather than the market of objects. The new supply curve will always start from the minimum profitable production (minimum marginal costs) which will rise together with production growth (growing costs of materials and workmanship in the short term).
What needs to be explained is the relationship between housing production costs and the supply curves in the real estate development sector. In the medium term, the real estate development sector is capable of providing more dwellings at higher costs, thus the cost curve will be close to aggregate marginal costs in the real estate development sector. The medium term supply curve in the real estate development sector (developers plan future investments based on current prices) may, as shown by empirical observations, may differ considerably from the analysed cost curve. Real estate developers generally underestimate the rise in the costs of production factors driven by rising demand and only react to nominal effects. Moreover, developers usually operate amidst a high financial leverage which changes the profitability ratio as the growth of production financed in such a way offsets the rising unit costs. In some countries, it is possible for the developer to finance construction costs with the buyer's pre-payments, usually without paying interests, which increases the investment profitability. As a result, the supply of real estate development pre-sale contracts may be, in the short term, considerably more elastic in terms of price growth than marginal costs.
In the medium term, capital will flow into the sector, pushing costs towards the minimum average cost level (long-term costs) and, consequently, the supply curve will get even more elastic. In practice both processes will be observed, namely new firms will enter the market and incumbent firms will increase their production. If the level of supply growth exceeds a certain value of the existing stock, average costs will rise on the back of tensions arising in the economy (transportation, materials, land, infrastructure, etc.). In the very long period (decades) analysed in the previous model, housing supply - here the already existing housing stock - will get flexible through the aggregation of annual construction effects. The entire economy will undergo structural adjustments aimed to match housing supply with the sector’s needs71.
Thus, the total market supply is the sum of supply of housing or pre-sale development contracts and supply from the secondary market. Supply in the short term (t), medium term
(t+1) and long term (t+2) is shown in Figure 7.
71 It should be added that economy may face erroneous, socially expensive and excessive adjustments to suit
construction. This was last observed, inter alia, in Spain. Considerable productive capital and human capital was used for housing purposes which has been translated into enormous costs generated by vacant buildings and mismatches in the labour market.
103
Figure 7. Supply of housing in the short term t, medium term t+1 and long term t+2 in the OOH market.
3.3 A simple model of the cycle
Basing on the previously described behaviour in the residential real estate market, we analyse a demand shock driven cycle. In the literature it is generally assumed that business cycles are driven by exogenous shocks. Due to specific character of the residential real estate market its cycles are inevitable.
The major housing cycles generators are multipliers which cause that even minor changes in certain factors result in strong fluctuations at the macroeconomic scale. Interest rate changes strongly affect the demand through the credit multiplier, being the product of the interest rate before and after its change. For example, an interest rate reduction by 1 percentage point from 4% to 3% means a 25% fall in actual home purchase costs. Thus, a minor change in interest rates translates, as previously discussed, into a strong change in demand for the entire housing stock. Remember that the growth in demand concerns the entire housing stock whereas the primary market supply is a mere fraction of the whole stock. Thus, the credit multiplier effect is additionally strengthened by the demand multiplier. Under the assumption of a rigid short-term supply, this multiplier is defined as the percentage change of housing demand, and consequently, prices. Each year, the primary market generally accounts for 2% of the whole housing stock. Thus, a 4% rise in demand for the housing stock driven, for example, by a credit shock means a 100% rise in demand for construction. Amidst short-term rigid supply, this translates into a very strong price growth. The price multiplier effect results from both the credit multiplier and the fundamental multiplier. It generates strong shocks in the market and, consequently, excessive supply.
Once the market is put out of equilibrium, it replicates, and often deepens its cycles through a short-term rigid supply and flexible and lagged in time reaction of demand (Figure 8). An additional significantly destabilizing factor is the pro-cyclical behaviour of market participants (speculations), and often also public factors (economic and supervisory policy). Yet, the real estate market can generate cycles without the last factor, as the first two assumption are completely sufficient. In theory, housing cycles could be avoided should companies conduct market research and were able, in reliance thereon, to determine the equilibrium supply and synchronize their supply reaction. Yet, basing on practical knowledge on the real estate market this task may be considered as unfeasible. The basic difficulty is a 2-5 year-long time lag between the investment start and its effects, in which the equilibrium conditions change. Another problem is the fact that it is practically impossible to coordinate the production in a free and competitive market; what is more, such actions could be viewed as cartel practices.
St+2
St St+1
Nr of housing units
Pric
e
Sum of the primary and secondary market
St+2 St, St+1
Nr of housing units
Pric
e
Secondary market
St+2
St St+1
Nr of housing units
Pri
ce
Primary market
104
The cycle mechanism is determined by the shape of the demand and supply curves, in particular, the angles between them, which can be called an instability indicator. It makes fluctuations more expansive or gradually dampened. Observations of the market, especially of the correlations between newly commenced projects, demand and prices shows that demand growth is sufficient to trigger production growth. The price growth is observed with a certain time lag as the disequilibrium starts building up. This factor additionally increases the supply elasticity.
As a result, a change in demand results in a delayed and significant (due to the acceleration effects) reaction of supply, which, in turn, drives prices down. The ensuing disequilibrium, due to a rigid demand and price-elastic supply, is not offset by the classical mechanism of convergence towards the market equilibrium. Also home prices may become temporarily rigid, resulting in housing oversupply in the market. In the subsequent period, declines may be abrupt and construction may collapse. The supply elasticity may also change, modifying the size of construction in response to the price shock. Additional disturbances may be seen in the form of speculative behaviour as well as the impact of regulatory factors. Those previously discussed additional factors affect the shape of the cycle, providing it with a stochastic character actually observed in the housing market.
Figure 8. A simple model of the cycle
4. The relationship between demand in the primary and secondary market and the
acceleration effect In the short- and medium-term analysis, being the basis of assessment of the situation
in the housing market, the stock depreciation and amortization may be viewed as a long-term problem and be left out as such. However, we should address in more detail the emerging problem of the non-linearity between rising demand for housing and the price and supply
S(t2 )
S(t1 )
Pric
e
S, D
S(t2 )
S(t1 )
Pric
e
S, D
S(t2 )
S(t1 )
Pric
e
S, D
Pric
e
S, D
S(t)
105
reactions. Its consequence is the mechanism of the direct allocation of capital to the housing sector and the impact of the financial sector and the related allocation of capital through the financial sector.
Due to the non-arbitrage condition between the primary and secondary housing market, dwellings of similar quality and technical condition should be priced similarly. Yet, the non-arbitrage condition is usually disturbed by fiscal policy (taxes, subsidies) and regulations. In addition, the two housing stocks are not generally comparable in terms of dwelling characteristics and ownership status. Also, developers are more price flexible than sellers in the secondary market and can often encourage homebuyers to purchase homes above their market value. This way, in case of oversupply, developers are able to sell dwellings below secondary market prices.
As a result of the described relations, the secondary market is not, as would be in the case of the non-arbitrage condition and rigid supply determined by the housing stock, a simple shift of the equilibrium in the primary market by a constant. In fact, the primary market is affected by the response of both the primary and secondary markets, determining such parameters as price, flexibility and the number of transactions. Those parameters are determined, partly in the whole aggregate market, and partly in individual markets. This means that as in case of the exchange rate and gold price model, these parameters will develop independently, but should certain thresholds be exceeded, the mechanism of demand flow between both housing stocks will be triggered and the parameters start to converge. Such a mechanism could be called an imperfect non-arbitrage condition. The development of basic economic parameters in these markets amidst an imperfect non-arbitrage condition is shown in Figure 9. Due to such a condition (see NBP 2012), the primary market price is usually slightly higher than the equilibrium price (Ep), and the secondary market price is a little bit lower. This results from a more flexible short-term supply in the secondary market, as well as real estate developer’s marketing opportunities to convince the client of the higher value of a particular dwelling.
Figure 9. Arbitrage in the secondary and primary market
International comparisons (see Hypostat) show that transactions in the secondary market generally account for 2-5% of the housing stock, whereas the number of transactions changes over the cycle. The size of housing construction, namely the primary market size, usually accounts for 1-2% of the housing stock. These figures show that the actual size of the market and the actual supply account for just a few percentage points of the housing stock. An important but in the literature insufficiently well described consequence is the accelerating
106
impact of demand growth on prices. Changes of factors as income, interest rates and loan availability as well as demographics, generally affect the demand for the whole housing stock in a particular local market (and in some cases the national aggregates of these markets) whereas supply factors affect only a marginal part of the housing stock, namely what is currently on sale.
In the real estate development and construction sector the accelerator mechanism exerts a strong influence. Yet, this is not sufficiently highlighted in the literature, but reflected in the cyclical nature of the industry. We illustrate it with the following example. Let us suppose that the standard rate of stock replacement is about 1% which means quite conventionally, that an average housing unit lasts for about 100 years. Additionally, the restitution demand is generally nonlinear and accelerates during a construction boom. This means that the demand shock persists, contrary to the classical model, where the effect fades away after the first shock occurred. When the housing sector is in equilibrium, namely when only the depreciated housing stock is replaced, the annual demand for housing construction accounts for 1% of the housing stock. This works under the assumption that there are no demand shocks driven by growing income, migration and capital markets. However, let us now suppose that the economy is accelerating. As shown by numerous studies, with lower GDP and, consequently, lower level of housing needs satisfaction, the income elasticity of demand for housing may approach 1 (see Lin and Lin, 1999). With a 4-5% income growth, which corresponds to a 5-6% GDP growth, the aggregate demand for housing stock is likely to increase by an additional 4-5% in year-on-year terms. This means a four-, five-fold increase in demand that can be satisfied in the long run only. Consequently, real estate developers embark on long-term investment projects, consumers strive for a better place in the waiting list for housing and pre-sale construction contracts and rights thereto are traded. To speed up the contract realization, developers start purchasing ready-made projects from competitors, thus triggering a boom in the sector.
In our example, the acceleration of economic growth drove demand up to 5% of the replacement size in the real estate market and led to a boom in the real estate development market. Yet, market prices did not post such a significant growth as the additional increase in demand got dispersed in both the secondary housing market (previously 4%) and the new housing market (previously 1%) causing a combined effect of 100% price increase, thus prices doubled. Our theoretical considerations rely on empirical observations, presented in the analytical part of the NBP annual report (NBP 2012).
These phenomena show that in countries with a low level of development and strong housing preferences, loan availability and availability of mortgage-financed housing can be a good measure of demand. They also explain the credit boom phenomenon in the situation of changes in loan availability leading to significant changes in demand in the housing market and considerable price changes. Moreover, they also explain why the price elasticity of housing demand amidst price increases is low, regardless of speculation and well-known arguments about the fundamental nature of the needs for housing services.
The discussed behaviour does not account for speculation72 and collective behaviour enhancing these effects. In the case of the housing market the relationship between changes in
72 Various speculative motives should be distinguished. We may have to do with classical speculation
involving an attempt to generate windfall profits. The investor buys housing today to sell it at a higher price tomorrow. Yet, speculation may also have a fundamental base and be rational. A prospective buyer may expect prices to increase in the future, thus not being able to buy it with available loan. In this situation they decide to purchase housing as soon as possible, thus exerting an upward pressure on prices.
107
interest rates and simply understood loan availability is also non-linear, neglecting the constraints resulting from bank’s prudential policy. The long depreciation period combined with fixed instalments, especially amidst low interest rates causes that minor interest rate changes, in absolute terms, may trigger significant changes in demand. This factor may be multiplied with the previously described effect in the housing market, or even earlier in the real estate development market. This effect has not yet been accounted for in the central bank's monetary policy, yet its impact may be significant. The simplest method to explain it is to continue the example of the accelerating impact of demand growth on the real estate development market.
To achieve the objective of stimulating economic growth, the central bank cut its interest rates by 2 percentage points, i.e. from 4% to 2%. This effect translates not only into the aggregate demand in the economy, but also causes a twofold increase in the availability of mortgage loans, which, amidst given income also doubled the demand for the housing stock. Consequently, home prices doubled and speculative price bubbles emerged in the market. The discussed example may be extended to include GDP growth-induced migration, or an additional shock caused by the young marriage boom. We may also change the assumptions, generating bubbles at a different scale. It is worth noting that such a bubble is the result of fundamental factors, which effects are further amplified by speculative behaviour.
A similar mechanism of nonlinear interactions, yet this time negatively affecting business conditions in the sector, will be observed amidst downward price rigidity that is commonly observed in this market. However, the accelerating effect is considerably lower than in the previously discussed case. Should prices stiffen at a level ensuring that real estate developers generate decent financial results, they will embark on new investments and build housing on stock, waiting for better times to come. Due to a large margin they will be able to reduce the price and sell the supply surplus at a profitable price. Considerable possibilities of financing the unsold housing stock at high margins constitute factors favouring such practices. For example, if the rate of return on equity stands at around 20%, which is not an extraordinary result in this industry, real estate developers may finance sales a three-year stock of unsold housing and even more unfinished housing units (pre-sale construction contracts) with current housing. However, in reality, amidst a relatively low price elasticity of demand at high prices, the possibility of price reductions and profitable sale of the housing surplus are limited. The cumulating unsold housing stock adds to the developer's risk73.
The analysis shows that the short- and medium-term market model may be reduced to a market consisting of housing units only (newly completed and those from the existing stock)
73 These considerations may be illustrated on the example of the Warsaw market, which is based on
empirical observations, described in Part I of the Annual Report of the NBP (2012) and its previous editions. Let us assume that the annual absorption of the primary market at a price of PLN 7 000 per square meter is 13 000 housing units, similar to the secondary market capacity. Price elasticity of demand is -0.5, and the costs of production, including land costs are PLN 4 000 per square meter. This brings down production costs by about 40%. Thus, real estate developers may expect a 20% increase in demand i.e. a growth of 2600 dwellings a year, as the remaining 2 600 will be sold in the secondary market. Even assuming greater rigidity of prices in the secondary market and demand being shifted to this market, this means demand growth of 5200 dwellings. If the market sees its annual supply in real estate development housing double to reach 26 000 housing units, the price would have to fall by at least half, or below the cost of construction; with the threefold growth in annual production the level will be approx. PLN 2 300 per square meter. Rigid prices will result in tension accumulation and sharp price declines in the subsequent period, when additional, negative factors come to play. The NBP’s studies on the Warsaw market demand show that it is quite rigid and may get flexible only at the level of PLN 2000-3000 per square meters, which is below the cost of production.
108
and housing development sector (possibility of future production). In this perspective, the factors affecting the demand at time t concern the housing unit market, while the inflow of new capital (housing) takes place through the real estate development sector at time t+1. Similarly, the inflow of capital providing financing to final home buyers is captured in the demand curve.
This model shows that even relatively minor changes in fundamental factors trigger demand shocks. Those, in turn, first generate oversupply and then trigger downward adjustments, which consequently leads to strong cycles.
5. Conclusions Cycles are a very complex feature of the residential real estate market. Although they
are inevitable, a well-matched demand steering policy can have a smoothing effect. The investigation of housing market cycles must be based on the analysis of the number of housing units, as it is the mismatch between the number of desired and affordable housing units in the short term that boosts prices and, consequently triggers cycles. Our study has shown that the consideration of housing space is important for the long-run analysis. However, the assumption of an relatively easy adjustment of housing space does not provide an explanation as regards the formation of cycles in the short term.
In the housing market a multiplicative nature of price changes may be observed. It triggers severe shocks on the demand side, and consequently leads to an excessive supply. Those shocks, depending on the price elasticity of supply and demand, can either fade away or explode.
The results of the analysis of the impact of interest rates largely depend on the assumptions made, which shows that the macroeconomic housing model must rely on sound microeconomic foundations. Minor changes in interest rates, which at the macroeconomic level appear insignificant, on the account of their strong influence at the microeconomic level, translate into housing market shocks being felt throughout the entire economy.
To sum up, we can say that the main difference between the analysis in the short and in the long term is the importance of housing construction and the percentage of housing stock on sale, participating in the adjustment process. In the short term, the share of the housing stock on sale is small, and production growth is determined by marginal costs. In the long term, housing stock adjustments are becoming increasingly important and the supply of new housing stock is determined by long-term average costs. References Anas, A. (1982). Residential Location Markets and Urban Transportation. New York:
Academic Press. Anas, A. (1990). Taste Heterogeneity and Urban Spatial Structure: The Logit Model and
Monocentric Theory Reconciled. Journal of Urban Economics. Vol. 28(3), 318-335. Augustyniak H., Łaszek J. and Olszewski K. (2012). Housing market cycles – a
disequilibrium model and its calibration to the Warsaw housing market. Paper presented at the Warsaw International Economic Meeting 2012.
Aoki K., Proudman J. and Vlieghe G. (2002). House prices, consumption, and monetary policy: a financial accelerator approach. Bank of England Working Paper nr 169.
Bernanke B., Gertler M. and Gilchrist S. (1999). The financial accelerator in a quantitative business cycle framework. Handbook of macroeconomics (North Holland).
Bracke P. (2011) How Long Do Housing Cycles Last? A Duration Analysis for 19 OECD Countries. IMF Working Paper WP/11/231.
Brus W. (1970). Koszt społeczny usługi mieszkaniowej. IGM. Warszawa.
109
Brzoza-Brzezina M., Chmielewski T. and Niedźwiedzińska J. (2010). Substitution between domestic and foreign currency loans in Central Europe: do central banks matter? ECB Working Paper nr 1187.
Burnham J. B. (1972). Private Financial Institutions and the Residential Mortgage Cycle, with Particular Reference to the Savings and Loan Industry. In: Board of Governors of the Federal Reserve System, Ways To Moderate Fluctuations in Housing Construction.
Carter S. (2011). Housing Tenure Choice and the Dual Income Household. Journal of Housing Economics. Vol. 20(3), 159-170.
Case K., Karl E. and Shiller R. (2003). Is There a Bubble in the Housing Market?. Brookings Papers on Economic Activity, 299-342.
Chang C-O and Ward C. W. R. (1993). Forward pricing and the housing market: the pre-sales housing system in Taiwan. Journal of Property Research. Vol. 10, 217-227.
DiPasquale D. and Wheaton W. C. (1992). The markets for Real Estate Assets and Space: A Conceptual Framework. Journal of the American Real Estate and Urban Economics Association. Vol. 20(1), 181-197.
DiPasquale D. (1999). Why Don't We Know More About Housing Supply?. The Journal of Real Estate Finance and Economics. Vol. 18. Nr 1, 9-23.
Dunsky R.M. and Follain J. R. (1997). The demand for mortgage debt and the income tax. Journal of Housing Research. Vol. 8, 155-199.
Fotheringham A. S. (1988). Consumer Store Choice and Choice Set Definition. Marketing Science. Vol. 7. Nr 3, 299-310.
Goodman A. C. (1988). An Econometric Model of Housing Price, Permanent Income, Tenure Choice and Housing Demand. Journal of Urban Economics 23, 327-353.
Heckman J. (2010). Building Bridges between Structural and Program Evaluation Approaches to Evaluating Policy. Journal of Economic Literature. 48(2). 356–98.
Hott C. and Jokipii T. (2012). Housing Bubbles and Interest Rates. Working papers. The Swiss National Bank.
Khan B., W.L. Moore and R. Glazer (1987). Experiments in Constrained Choice. Journal of Consumer Research. Vol. 14 Nr. 1, 96-113.
Kim M.-J. (2010). Residential Location Decisions: Heterogeneity and the Trade-off between Location and Housing Quality. PhD dissertation at The Ohio State University.
King A. T. (1976). The Demand for Housing: Integrating the Roles of Journey-to-Work, Neighborhood Quality, and Prices. Online: www.nber.org/chapters/c3968
Lancaster K. J. (1966). A new approach to Consumer Theory. Journal of Political Economy. Vol. 74. No. 2.
Lambertini L., Mendicino C. and Punzi M. T. (2012). Expectations-Driven Cycles in the Housing Market. Bank of Finland Research Discussion Paper 2-2012.
Levin E. J. and Pryce G. (2009). What Determines the Responsiveness of Housing Supply? The Role of Real Interest Rates and Cyclical Asymmetries. Centre for Public Policy for Regions Discussion Paper Nr 20.
Lin C. and Lin S. (1999). An Estimation of Elasticities of Consumption Demand and Investment Demand for Owner-Occupied Housing in Taiwan: A Two-Period Model. International Real Estate Review. Vol. 2(1), 110-125.
Łaszek J. (2006) Rynek nieruchomości mieszkaniowych i jego specyfika. in E. Kucharska-Stasiak. (Edt.), Ryzyka banku w zakresie określania wartości nieruchomości dla celów kredytowych w Polsce na tle trendów w Unii Europejskiej. Zeszyt Hipoteczny 23.
Łaszek J., H. Augustyniak and M. Widłak (2009). Euro a ryzyko bąbli na rynku nieruchomości mieszkaniowych. Narodowy Bank Polski. Materiały i Studia. Nr 218.
NBP (2011). Report on the residential and commercial real estate market in Poland in 2010.
110
NBP (2012). Report on the residential and commercial real estate market in Poland in 2011. Pain N. and Westaway P. F. (1997). Modelling structural change in the UK housing market:
A comparison of alternative house price models. Economic Modelling 14, 587-610. Phang S. Y., Kim K. H. and Wachter S. (2010). Supply Elasticity of Housing. International
Encyclopedia of Housing and Home. Elsevier and Science Direct. Phe, H.H., and Wakely, P. (2000). Status, Quality and the Other Trade-off: Towards a New
Theory of Urban Residential Location. Urban Studies, 37(1), 7-35. Sanchez A. C. and Johansson A. (2011). The Price Responsiveness of Housing Supply in
OECD Countries. OECD Economics Department Working Papers. Nr 837. Sommervoll D. E., Borgensen T.-A. and Wennemo T. (2010). Endogenous housing market
cycles. Journal of Banking & Finnce Vol. 34, 557-567. Rubaszek M. and Serwa D. (2011). Determinants of credit to households in a life-cycle
model. NBP Working Paper Nr. 92. Rubaszek M. (2012). Mortgage down-payment and welfare in a life-cycle model. Paper
presented at the NBP workshop: “Current Trends in Macroeconomic and Finance Research”.
Topel R. and Rosen S. (1988). Housing Investment in the United States. Journal of Political Economy. Vol. 96 Nr 4, 718-740.
Tse R.Y.C., Ho C.W. and Ganesan S. (1999). Matching housing supply and demand: an empirical study of Hong Kong's market. Construction Managment and Economicss Vol. 17, 625-633.
Waldron M. and Zampolli F. (2010). Household debt, house prices and consumption in the United Kingdom: a quantitative theoretical analysis. Bank of England Working Paper Nr 379.
Westaway P.F. (1992). A simulation model of consumer spending and housing demand. NIESR Discussion Paper Nr 15.
Wheaton W. C. (1999). Real Estate “Cycles”: Some Fundamentals. Real Estate Economics, Vol. 27 Nr 2, 209-230.
111
A3 Real estate development enterprises in the Polish market and issues related to its analysis
Hanna Augustyniak74, Krzysztof Gajewski 74, Jacek Łaszek 74,75, (with advice from Grzegorza Żochowskiego76)
Introduction The real estate development sector in Poland took a large scale in the years 1994-1995,
upon the disappearance of the effects of loan agreements signed before 1990 with housing cooperatives and related government-subsidized housing. In the subsequent years, this sector gained a dominant position as a supplier of multi-family housing, especially in the largest cities. The commercial functioning of this sector, i.e. expansion of offers addressed to a wide range of clients, was first observed after 2000, when the high inflation came to an end and mortgage loans become the primary form of financing for this type of construction.
During its expansion, the real estate development sector experienced two cycles. The first one during the 1999-2001 boom and the second one in the years 2005-2008. The first of these cycles had a domestic character and was driven by classical real estate market mechanisms, which unfortunately coincided with changes in fiscal regulations. The second one, however, was triggered by the global boom in real estate markets, that was largely driven by the financial system. The consequences of the first cycle were more severe for the sector as it was poorly capitalized, whereas during the second cycle, companies were much wealthier and had already gained a certain experience. Additionally, the sudden burst of the bubble in the U.S. curbed the growth of a bubble in Poland. Highly elastic prices during a demand growth and more rigid ones during declining demand indicate a low level of competition in the sector. In fact, only the past two years have seen a growing competition in the sector.
The analysis of the real estate development sector aims to find answers to the following two questions:
1. What is the condition of the majority of developers, especially in terms of the involvement of the banking sector?
2. Is the current level of home prices stable or is it likely to decline sharply, and how does it affect the housing production sector?
Currently, the share of real estate development production in Poland accounts for approx. 1% of GDP, and the share of real estate development loans in the banking sector assets is even smaller. Housing is classified, in general, as an asset, a capital good generating housing services and, finally, as a consumer good. The problem of the central bank's response to large increases in home prices and the question whether to include home prices into the basket of goods determining the inflation target is still under discussion. Yet, today most economists believe that the central bank should monitor the real estate sector, especially its prices. This is due to the fact that prices and the resulting economic condition of the sector and, to some extent, the impact of the sector on the entire economy (multiplier effects of construction, wealth effects, speculation and price bubbles) are closely related to the level of interest rates
74 Economic Institute, National Bank of Poland. 75 Warsaw School of Economics. 76 REAS.
112
and the mechanism of liquidity supply to the banking sector, the tools used by the central bank. In the recent years, we witnessed numerous negative examples driven by the lack of common sense in this respect - the most spectacular crises in the last eight-year period were observed in the USA, Ireland, Spain. Yet, the sector’s negative experience has a rich history, spanning over the period of several hundred years.
Selected issues in the real estate development sector analysis A real estate developer is a company operating in the residential or commercial real estate
sector, involved in building, renting or selling real estate on its own account, with a view to making a profit. In developed economies, real estate developers usually deal with residential and commercial real estate, while the latter also include residential properties for rent. They operate also in the land development market, converting agricultural land to non-agricultural one, providing the land for construction with appropriate infrastructure and utilities and selling it.
In the case of residential real estate, the developer usually finances construction with his own funds. The cost traditionally includes the acquisition of the land which has to be owned by the developer or be under a long-term lease, expenses related to the execution of architectural design and obtaining a building permit. Developers also finance their investment through investment funds, the so-called mezzanine finance, which are treated as own capital contribution, but are much more expensive than a bank loan. Financing through the capital markets and the issue of debt securities is less costly and troublesome, but usually available to the largest companies with a good reputation only.
In the case of higher risk, the bank may require a larger share of developer’s own funds. Additional sources of financing may also include down payments made by future owners and untimely paying of construction companies’ bills. To execute the investment project, the real estate developer employs construction companies but sometimes the developer performs the whole or part of the investment through their own construction company. Yet, such a system is considered a second-best choice due to the greater difficulty in the enforcement of commitments. The bank financing the investment usually requires the real estate developer to establish a special purpose vehicle (SPV), which owns the land and building under construction as well as to open an account used for all the financial flows associated with the investment. The construction loan with a 3-5 years maturity is disbursed in installments in accordance with the approved loan disbursement schedule as defined in the loan agreement, after the actual advancement of the project has been confirmed and invoices for the previous period verified. Yet, the bank tries to avoid financing of the developer’s profit, which should be financed at the end of investment. Likewise, also the real estate developer provides financing to construction companies only after particular stages of works have been completed and accepted.
When completed, housing is sold to buyers for cash or with a mortgage, and the mortgage at the bank funding the investment is converted into mortgage at the bank providing financing to the home buyer.
In the case of commercial real estate, not only the property itself, but also the company that owns the property is the object of transaction, and the construction loan is converted into a 10-15 year financing.
The time necessary to execute the real estate development project may vary depending on the specific character of a particular investment project, and includes the time associated with investment arrangements and execution of the construction process itself. The former one is largely impacted by formal requirements regarding real estate construction. The latter one depends upon the technical aspect of the project.
The discussed characteristics of real estate development reveal the main problems which arise during the economic analysis of this sector. The most important issues include the
113
discontinuous and unitary nature of production and lengthy production process, usually spanned over several years. Consequently, due to the unitary nature of production, it is difficult to apply the conventional index-based analysis, since the inputs and outputs are in each case unique. Thus, there arises the problem of how to account the construction effects in traditionally accepted periods, as the final result will be available with a time lag.
A solution to the unitary nature of the production is to analyze the whole sector or a basket of selected companies. As a result of the discussed specificity of the sector, larger developers usually operate as holding companies, being owners of investment projects.
With a sufficiently large number of such entities in the sample, the business continuity may be assumed as there are always some investments under construction, some are being embarked on and some are being completed. The problem is that costs and expenses are calculated in a different place; what is more, the expenses include costs incurred at different stages of the construction process. As a result, with a constant sample and stable economy, the calculation results are acceptable indicators, yet with larger shocks leading to changes in costs and the volume of production, the results will be subject to an error. For example, a growing number of new projects will push up costs and bring down profitability indicators, whereas a decline will trigger the opposite effect. This is a significant drawback, since the sector is, by definition, subject to cycles. Moreover, during such changes we need even more reliable information. Nevertheless, these shortcomings do not disqualify the method itself, however, require some caution in the interpretation of the results, knowledge of real processes in the sector and reliance on additional, independent sources of data. Thus, it is pragmatic to say that poorer information, yet verified through a variety of sources and often corroborated with expertise is better than none.
While subject to certain reservations, the indicator-based analysis of the sector may be applied, the analysis of individual companies requires the adoption of additional assumptions, especially regarding future sales; moreover, it is burdened with a significantly higher error.
In Poland, by the end of 2004, the problem was the lack of standards to qualify the building as sold. It was either the time of the sale of the contract or the time when the building was actually put into use. The problem was solved by the large-scale adoption of the International Accounting Standards (IAS).
In the real estate sector, as in other sectors, there is a tendency to hide profits in costs and finance certain expenses in this way. This problem usually occurs amidst higher corporate tax rates and a less developed capital market, which does not discipline players, especially public companies. Another form of the same process is the capital outflow from one company to another one, for example, from a real estate development company to a construction company. This practice may well go hand in hand with loan withdrawal from the project. In both cases, the result will be excessive costs and poor financial performance of the sector and companies, which, in fact, does not have anything to do with the reality. In the real estate sector, this problem may be partially solved if we use, apart from measures based on corporate reporting, also measures based on market prices, cost estimates and valuation of investment projects in reliance on these data. In fact, the financial results of real estate developers, due to the business-related risk and ensuing losses, will correspond to such measures, only in exceptional circumstances. Yet, their level, compared with the level of measures based on financial reporting will indicate a range within which the real estate sector indicators are contained.
With the above reservations in mind, the analysis of real estate developers may rely on most indicators used in the analysis of a usual company, i.e. indicators based on the vertical and horizontal analysis of the balance sheet and the income statement, as well as measures of profitability, costs, liquidity, debt collection effectiveness etc. The object of the study should be the key determinant of the indicator selection. However, while the income statement makes
114
sense when analyzing a large holding company, in small businesses and projects only the balance sheet is available, as there are numerous expenses, growing assets and income appears only at the end of the process.
Assumptions and basic indicators used in the analysis In the analysis of the real estate sector, the National Bank of Poland uses two basic
approaches. In the first approach, based on F-01/I-01 and F-02 forms of the Central Statistical Office (CSO), using a relatively constant sample of enterprises, the index-based analysis is applied which, as previously discussed, treats the sector as a company. The object of the analysis are small and large businesses broken down according to the CSO classification. Measures of the condition of enterprises are adapted to the sector’s specific character and take into account the previously outlined aim of the analysis. These are total revenues, total expenses, net income, ROE and ROA, which, when analyzed from a historical perspective, show the financial condition of the entire sector and of an average company. For the purpose of the study, these indicators are defined and calculated as follows:
• Total revenue – total sum of net sales, other operating and financial income, • Total expenses - operating expenses plus other operating costs and financial expenses
minus change in inventory minus the cost of goods produced for own consumption, • Net income - gross earnings minus income tax, • ROA - net income at end of period / assets at end of period (in %). • ROE - net income at end of period / equity at end of period (in %)
Measures depicting the situation from the real estate assets perspective are development projects under construction, completed housing, land bank and land bank exposed as the average, annual production in the recent years. These include:
• Real estate development projects under construction - stocks of semi-finished products and work in progress,
• Completed housing - stocks of finished products, • Land bank - stocks of goods, • Land bank in the years of production - stocks of goods / 0.15 * income from sales.
As real estate developers, by definition, leverage their activities (use financial leverage), and may have recourse to various sources of financing, with different availability and at different costs, the change of the structure of these sources is of considerable importance. It provides information on potential problems that companies may face (for example, growing share of liabilities, falling share of loans) and financial costs that companies incur. The main problem is usually the availability of data. Based on CSO reporting, the following sources of financing may be identified:
• Equity - fixed assets plus current assets minus liabilities and provisions for liabilities • Loans – long-term liabilities resulting from loans and borrowings plus shor-term
liabilities resulting from loans and borrowings, • Debt securities – long-term liabilities resulting from issuing of debt securities plus
short-term liabilities arising from issuing of debt securities • Client down-payments and - short-term liabilities resulting from advances for
deliveries plus accruals, • Trade liabilities – liabilities resulting from goods delivered and services rendered to
other entities, • Other liabilities.
115
Equally important as the analysis of the sources of financing, is the analysis of costs. As the real estate developer usually employs subcontractors, the basic cost item are third party services, or costs of housing construction. Other costs are usually associated with the activity of the real estate development company that arranges investment projects, exercises supervision, keeps the accounting and sells completed housing. In theory, the real estate development company can operate virtually at no cost, as these costs can be, for the most part, accounted for as third party services. On the other hand, some smaller companies are contractors themselves, and consequently the materials and services consumed do not necessarily have to mean pure development costs, but costs associated with housing production. In the NBP analysis, costs are broken down as follows:
• Third party services, • Payroll, • Materials and energy consumed.
The conclusions resulting from the analysis of real estate companies may be verified by analyzing the real economy, that is the actual costs of housing construction, market prices and profitability of housing projects, which should translate into the condition of real estate developers and provide an answer to the question, whether companies are going to increase housing production. The simplest, and, therefore, fairly reliable measure is the structure of market prices of housings in the local market. The starting point are current housing prices, costs of housing production and sale, and costs of construction sites. The NBP uses for this purpose the transaction price at the local primary market, taken from the NBP BaRN database that collects housing prices. To estimate the construction cost per one square meter of usable area of housing, assumed to be an average cost, we rely on the information from the Bulletin of real estate construction prices published by Secocenbud. The type 1121 multi-family building was adopted as an average building 77.
In each quarter, prices of particular stages of the construction process (e.g, zero state, raw state of the building, external and internal finishing stage, installations) are quoted, and the percentage share of individual components of the cost estimate in the the total price of the object is given. Costs of labor, materials and equipment, indirect costs and profits of a construction company are quoted separately. To those costs the cost of the project (approx. 3%), the cost of land (according to the company’s own base of land prices) and other costs associated with the launch of the investment (approx. 5%) and VAT need to be added The approximate amount of the developer’s cumulative gross profit (for the entire investment period, without taking into account the provisions for investment risk) is obtained residually, after deducting the price of the other elements.
Yet, when interpreting the size of the developer’s profit it should be borne in mind that this does not mean the actual share of the profit in sales, nor the rate of return on equity. The real estate development project takes many years. Thus, to calculate such indicators, the result should be adjusted accordingly. On the other hand, real estate developers do not commit all
77 This is residential, multi-family, five-storey building with an underground garage and retail space on the ground floor; traditional design (masonry over ground part made of ceramic bricks); for simplicity, it was assumed that the cost of construction of a square meter of garage and utility and service areas is similar to the cost of housing construction in the shell unit standard; the actual, based on construction costs, price of one square meter of housing depends on the share of external space, which is different for different buildings; when calculating one square meter of consumer’s usable housing, a 20% share of external surface area in relation to the size of housing was adopted; this value was used to revise upwards the price of a square meter of housing.
116
their capital at the beginning of the project. They also use external sources of financing, including clients’ down-payment. These factors largely influence the expected investment performance indicators.
Based on the observations of the Polish market, it may be assumed that margins in the range of 20-30% ensure a good profitability. Below this range, we may reckon with a decline in the volume of production. This also depends on the developer’s capacity of alternative capital allocation. Consequently, in order to calculate the annual housing production effectiveness, a simplified business plan of the housing project should be developed on the basis of performance indicators. For the purpose of the analysis, the NBP adopts, in this case, the following assumptions.
The entire construction process, which takes approximately 18 quarters, can be broken down into the following three stages:
A. Preparations taking approximately 7 quarters (purchase of construction land, preparation of the project and obtaining of all necessary permits),
B. Execution and construction taking approximately 7 quarters, C. Sale, which should be completed within 4 quarters.
The description of the construction process model is presented in Table 1, whereas developers’ profits calculated on the basis of this model are presented in Chapter 4 of the annual NBP report on the real estate market. Conclusions
Economic indicators calculated on the basis of this model are theoretical indicators, as they fail to take into account the real economic and financial conditions of the company and the specific character of the project. The actual business conditions, include, in particular, the financial situation78 and sources of investment financing of the company. The actual conditions of the project also include additional costs of technical infrastructure, and hard to predict geological difficulties. The real estate development activity involves a significant risk, starting from the cyclical nature of the market to the general economic, political or legal risks, not to mention natural disasters or construction failures.
Consequently, those profits which take into account the risk premium, are generally higher than in other industries. However, real estate developers are highly flexible in adjusting to economic conditions. This is shown by their experience before 2005, when they operated at low profit margins. On the other hand, amidst an economic downturn that is a typical problem for the sector, the possibilities of capital withdrawal and its allocation to other industries become limited.
The experience of foreign markets shows that real estate developers, like construction companies, build housing provided this activity brings even a rate of return of few per cent, rightly hoping the cycle to reverse. Bankruptcies are usually the consequence of excessively imprudent behaviour during the boom (too many deluxe projects, land purchases at high prices, materials, etc.).
78 For example, purchases of land and construction contracts signed during the boom and high prices in projects dominated by luxury apartments, sale in a period of stagnation, or the need to change the project. The opposite scenario may take place, too.
117
Tab
le 1
Mod
el o
f dev
elop
men
t con
stru
ctio
n pr
oces
s (r
esid
entia
l bui
ldin
g, 1
00 d
wel
lings
of 5
0 sq
uare
met
ers
each
)
Pe
rio
d
Pro
ject
sta
ge
Acti
vit
y
Co
sts
Eq
uit
y-b
ase
d
fin
an
cin
g
Q1
23
45
67
89
10
11
12
13
14
kw.
15
16
17
18
sha
re0
%1
5%
20
%2
0%
20
%2
0%
20
%3
0%
30
%3
0%
30
%3
0%
30
%3
0%
sha
re3
0%
30
%3
0%
0%
Q1
23
45
67
89
10
11
12
13
14
16
17
18
sha
re0
%0
%0
%0
%0
%0
%0
%7
%1
5%
25
%2
6%
50
%6
3%
70
%7
0%
35
%0
%
Ave
rag
e q
ua
rte
rly
com
mit
me
nt
of
loa
n in
the
inve
stm
en
t p
roje
ct
Ave
rag
e q
ua
rte
rly
com
mit
me
nt
of
clie
nt
do
wn
pa
ym
en
ts in
the
inve
stm
en
t p
roje
ct
Ave
rag
e q
ua
rte
rly
com
mit
me
nt
of
ow
n
cap
ita
l in
th
e
inve
stm
en
t p
roje
ct
Lan
d (
15
%),
do
cum
en
tati
on
+ p
erm
its
(5%
),
up
to
20
% o
f in
vest
me
nt
cost
s
Lan
d p
urc
ha
se,
do
cum
en
tati
on
an
d p
erm
its
Sta
rt
30
% o
f to
tal i
nve
stm
en
t co
sts
(20
% f
rom
th
e p
revi
ou
s
pe
rio
d p
lus
10
% a
s a
loa
n c
on
dit
ion
)
30
%
Loa
n t
ake
n t
o f
ina
nce
70
% o
f to
tal i
nve
stm
en
t co
sts
acc
ord
ing
to
th
e f
ollo
win
g s
che
du
le:
16
%
Lan
d (
10
0%
), d
ocu
me
nta
tio
n (
10
0%
),
pe
rmit
s (1
00
%)
26
%
51
%
36
%
34
%
0%
0%
Lo
an
fin
an
cin
g
23
%
qu
art
er
15
-18
Sale
an
d c
om
ple
tio
n o
f th
e in
vest
me
nt
Sa
le o
f 4
0%
of
co
mp
lete
d d
we
llin
gs,
loa
n r
ep
aym
en
t, c
om
ple
tio
n o
f
inve
stm
en
t, e
sta
blis
hm
en
t o
f h
ou
sin
g
com
mu
nit
y
Sale
co
sts
(3.5
% o
f re
turn
on
inve
stm
en
t)
40
% o
f d
we
llin
gs
fin
an
ced
wit
h e
qu
ity
cap
ita
l an
d p
rofi
t
(so
ld in
qu
art
ers
15
-18
)
qu
art
er
0-7
0%
15
70
%
qu
art
er
8-1
4
Co
nst
ruct
ion
+ f
ina
l in
spe
ctio
n a
nd
acc
ep
tan
ce+
sa
le
of
som
e d
we
llin
gs
Co
nst
ruct
ion
of
bu
ildin
gs
an
d in
fra
stru
ctu
re,
fin
al
tech
nic
al i
nsp
ect
ion
an
d a
cce
pta
nce
, sa
le o
f 6
0%
of
dw
ellin
gs
Up
to
10
0%
of
tota
l in
vest
me
nt
cost
s
118
Part III. Monographs of 16 cities in Poland
We translated only the description of the Warsaw market, which is the biggest market in Poland. The translation of the graphs, can be used to read the graphs concerning the remaining cities in the Polish version of the report.
1.1 Main developments in the Warsaw housing market
In the Warsaw market, which is the capital city of Poland, in 2011 the supply side tried to adjust itself to the current market conditions. The reduction of price limits in the government subsidized housing program “Family’s own housing” (RNS, hereafter), which were even stronger in the secondary market, shifted demand more towards the primary market. Moreover, the supply of attractive housing from the secondary market was reduced as many of them were purchased with a mortgage in foreign currency. As the Polish currency depreciated and housing prices decreased, in some cases the foreign denominated mortgage was higher than the current value of the dwelling. In such a situation only those owners sold, who really need to do so.
The majority of home brokers decided not to take any fees from buyers in order to improve their competitiveness and increase sales. Moreover, they focused on dwellings which had an attractive price in relation to its location, as this guarantees a fast transaction. However, some brokers had to reduce the number of their offices or even went out of business, while others expanded their services.
Developers continued to start new investment projects, despite a growing stock of unsold housing and a potentially deteriorating demand as a consequence of fundamental factors. One of the reasons for such actions might be their will to reduce the risk associated with the introduction of the Act on the protection of home buyers’ rights in April 2012, as there was no experience with its functioning. Further on, the decline in housing units delivered to the market was the result of the reduction of developer’s investment projects in 2009, which was caused by the global financial crisis.
One of the supply side drivers was the government subsidized housing programme RNS. Even though the price limit was reduced, more families made use of it in 2011 than one year ago. This increase was caused by the fact that also singles could participate in it. Moreover, there was the fear that the limits will be reduced further and it was also announced that the programme will be shut down at the end of 2012. Further on, the Financial Supervision Authority announced, that as of 1 January 2012 the Recommendation S will be renewed. Thus banks are forced to apply more stringent procedures as concerns the analysis of clients credit affordability.
The Warsaw market saw, as already happened in 2010, a strong pressure on a price reduction from the buyers side, which are still considered as too high in comparison to household’s income. The persistently high level of primary market offers increased the competition in the market. Developers were more elastic in negotiations with clients for ready built dwellings. Moreover, they adjusted the dwellings put in pre-sale contracts on the market to meet the demand as concerns the housing size and number of rooms. This way, most developers were able to realize a similar number of transactions as one year ago. However, developers, which to a large extent offered housing that did not meet the limits of the RNS programme or which were started under a quite different market demand, had problems to
119
meet their own sale targets. Thus, some of them, which had financial problems changed their projects under construction, stopped the realization of the investment or sold construction land with building permits in place. Other developers, which had already finished housing with a large usable area on stock, offered special discounts, up to several thousands of PLN, but this reduced their oversupply just slightly.
Due to a slight price decrease and income growth in Warsaw, the affordability of housing increased, to a larger extent in the primary market. The expectations of sellers in the secondary market, similar as in the previous year were rather exaggerated in comparison to prices that the market was willing to accept. As a consequence, the disparity between offer and transaction prices in the secondary market was higher than in the primary market.
The demand side in Warsaw focused, similarly as in 2010, on small dwellings with a relatively high number of rooms. The most demanded dwellings in the secondary market were pre-fabricated dwellings, which had a lower price, good location and well-developed surrounding infrastructure.
The Warsaw housing market in 2012 will depend strongly on the general economic condition of Poland. As some economic indicators worsen (i.e. growing interest rates, inflation and declining credit affordability), households will be more careful when making purchase decisions. There will be a continued pressure to lower prices, caused by a lowered effective demand and a growing competition among developers, due to the high level of demand. Demographic factors, such as a declining number of new marriages and new-born children as well as a large share of young people (up to 34 years) among the unemployed, which should constitute the main group of house purchasers, have a negative effect on the market. On the other hand, the growing net migration towards the capital city – due to a low unemployment rate and high mean salary - will have a positive effect on the market.
120
1.2 Information about the housing market in Warsaw Table 1. Basic data concerning the housing market situation in Warsaw.
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Housing stock:
- the number of Housing units 717784 729889 739324 753182 766223 780911 799661 818874 831014 840370 - housing units per 1000 inhabitants 425,2 432 436,7 443,7 450,2 457,6 467,7 477,6 483,0 491,9 Usable floorage:
- total (sq. m.) 39849890 40738203 41394522 42376069 43370640 44506242 45906295 47312305 48198253 48941119
- mean (sq. m.) 55,5 55,8 56 56,3 56,6 57 57,4 57,8 58,0 58,2 - per person (sq. m.) 23,6 24,1 24,5 25 25,5 26,1 26,8 27,6 28,0 28,6 Average number of rooms in housing unit 3,15 3,15 3,15 3,14 3,14 3,13 3,13 3,12 3,12 3,12 Average number of occupants in housing unit 2,35 2,31 2,29 2,25 2,22 2,19 2,14 2,09 2,07 2,03 The number of households in 2002 757578 The number of householda per one, stock housing unit in 2002 1,06
*Notice: Estimations of Regional Branch in Warsaw; Source: CSO in Warsaw.
1.2.1 Demographic situation in Warsaw Table 2. Demographic situation in Warsaw
Population
Birth rate
Migration balance
Changes in
population size
Marriages Birth rate
Migration balance
Marriages
absolute numbers per 1000 inhabitants
2002 1688194 -5295 6413 1118 7636 -3,2 3,8 4,6
2003 1689559 -5085 7272 2187 8104 -3,0 4,3 4,8
2004 1692854 -3474 6841 3367 8231 -2,1 4,1 4,9
2005 1697596 -2477 8282 5805 8379 -1,5 4,9 5,0
2006 1702139 -960 8179 7219 9023 -0,6 4,8 5,3
2007 1706624 -117 5785 5668 9511 -0,1 3,4 5,6
2008 1709781 728 3897 4625 9865 0,4 2,3 5,8
2009 1714446 636 3837 4473 9800 0,4 2,3 5,7
2010 1720398 2012 3940 5952 8961 1,2 2,3 5,2
2011 1708491 1277 7102 8379 8217 0,7 4,2 4,8 Source: CSO in Warsaw.
121
Figure223.Population of Warsaw and projections
Figure 224. Demographic situation in Warsaw
Source: CSO
Table 3. Households in Warsaw
by NSP, 2002 Households Share (in %)
Total 757578 100,0 One-family households 439062 58,0 -one-family households (2 members) 189267 -one-family households (3 members) 139058 -one-family households (4 members) 110737 Two-family households 12696 1,7 Three or more-family households 358 0,0 Non family households 305462 40,3 - non-family households (1 members) 289790 - non-family households (2 members) 13637 - non-family households (3 members) 1510 -non-family households (4 members). 525
Source: CSO in Warsaw
Table 4. Population ratio at working, pre-working and post working ages in Warsaw
Share of population (in %) at age:
working pre-working post-working 2002 64,9 15,7 19,5 2005 65,6 14,7 19,8 2010 63,9 15,0 21,1
Source: CSO in Warsaw.
1260
1360
1460
1560
1660
1760
1860
196020
02
2003
2004
2005
2006
2007
2008
2009
2010
2011
2015
2020
2025
2030
2035
in th
ousa
nds
-4,0-3,0-2,0-1,00,01,02,03,04,05,06,07,0
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
pe
r 1
000
inha
bita
nts
marriages Migration balance Birth-rate
122
1.2.2 Economic factors in Warsaw
Figure 225. Unemployment rate and selected structure of the unemployed in Warsaw (in %)
Figure 226. Average monthly payment in enterprise sector in Warsaw
Source: CSO Source: CSO
Figure 227. Housing availability per average salary in Warsaw
Figure 228. Loan availability (in thousands PLN) per average salary in Warsaw
Source: NBP, CSO Source: NBP, CSO
Figure 229. Availability of loan-financed housing (for average salary in thousands PLN)
Figure 230. Current housing-loan debt in Warsaw
Source: NBP,CSO. Source: Estimations on the basis of BIK and BGK data.
15
20
25
30
35
40
45
1
3
5
7
9
11
13
15
17
19
21
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1 the
sha
re o
f pe
op
le a
ged
34
(%
)
Un
empl
oym
ent
rate
(%
)
WarsawPolandthe share of people aged 34 among the unemployed
2 2002 4002 6002 8003 0003 2003 4003 6003 8004 0004 2004 4004 6004 800
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
zl/
mon
th
Warsaw Poland
0,35
0,40
0,45
0,50
0,55
0,60
0,65
0,70
0,75
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
sq.
m.
primary market secondary market
70
80
90
100
110
120
130
Q4
20
06Q
1 2
007
Q2
20
07Q
3 2
007
Q4
20
07Q
1 2
008
Q2
20
08Q
3 2
008
Q4
20
08Q
1 2
009
Q2
20
09Q
3 2
009
Q4
20
09Q
1 2
010
Q2
20
10Q
3 2
010
Q4
20
10Q
1 2
011
Q2
20
11Q
3 2
011
Q4
20
11Q
1 2
012
in t
hous
ands
40,0
45,0
50,0
55,0
60,0
65,0
70,0
75,0
Q4
200
6Q
1 20
07
Q2
200
7Q
3 20
07
Q4
200
7Q
1 20
08
Q2
200
8Q
3 20
08
Q4
200
8Q
1 20
09
Q2
200
9Q
3 20
09
Q4
200
9Q
1 20
10
Q2
201
0Q
3 20
10
Q4
201
0Q
1 20
11
Q2
201
1Q
3 20
11
Q4
201
1Q
1 20
12
sq. m
.
0
1 000
2 000
3 000
4 000
5 000
6 000
2005
2006
2007
2008
2009
2010
2011
PLN
mill
ion
Housing loans RNS loans
123
1.2.3 The housing construction sector in Warsaw Table 5. Housing construction sector in Warsaw.
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Permits granted for dwellings construction Total: 10914 6849 12233 13036 18110 30047 20911 14527 10636 15663 - Private 1025 998 1188 1049 1295 1887 1390 974 1051 980 - For sale or rent 7463 3922 8934 10958 16426 27772 19040 13047 9424 13927
Dwellings in which construction has begun Total: 8616 6116 11635 16042 17670 24170 21330 11273 13855 14737 - Private 831 836 925 1043 1326 1610 1343 834 825 705 - For sale or rent 5546 3807 8203 12549 14301 20594 18622 9306 12144 13605
Dwellings and new residential buildings completed Total: 13070 12335 10300 14436 13686 15729 19049 19482 12462 9356 - Private 1360 1582 990 1439 1408 1252 1551 1127 930 1175 - Cooperative 4668 2679 2554 1931 2472 1495 1876 1230 831 622 - For sale or rent 6775 7862 6085 10569 9176 12478 15371 16576 10555 7297 - Public building society 206 127 518 395 159 413 0 0 0 0 - Municipal 0 2 4 102 471 91 251 549 146 262
Source: US
Table 6. Housing construction in Warsaw. 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Dwellings and new residential buildings completed - Total (in units) 13070 12335 10300 14436 13686 15729 19049 19482 12462 9356 - Total (in sq. m.) 905751 900455 734390 1046610 1059296 1215852 1434390 1439720 919696 742866 - per 1000 people 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 0,0 - per 1000 marriages 1712 1522 1251 1723 1517 1654 1931 1988 1391 1139 Marriages minus dwellings and new residential buildings completed - Total 0 0 0 0 0 0 0 0 0 0 - per 1000 people 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 0,00 Average useful floor area in m2 69,3 73,0 71,3 72,5 77,4 77,3 75,3 73,9 73,8 79,4 Number of rooms 40356 37833 28353 41735 40353 45824 57667 56635 36800 28602 Average number of rooms incomleted buildings 3,1 3,1 2,8 2,9 2,9 2,9 3,0 2,9 3,0 3,1
Average floor area of 1 room (in sq. m.) 22,4 23,8 25,9 25,1 26,3 26,5 24,9 25,4 25,0 b.d.
Source: US
124
1.2.4 Housing stock in Warsaw Figure231. Housing stock depending on age in Warsaw in 2002.
Figure 232. Housing stock depending on floorage in Warsaw in 2002.
Source: CSO Source: CSO
Figure 233. Housing stock depending on number of rooms in Warsaw in 2002.
Figure 234.Housing stock depending on type of ownership in Warsaw.
Source: CSO. Source: CSO in Warsaw.
Figure 235 Households and housing stock in Warsaw.
Legend to Figure13:
Households A – one-person B – two-person C – three-person D – four-person and more
Stock depending on number of rooms
B – one- and two-room
C – three-room D – four-room and more
Stock depending on floorage
A –less than39 sq. m. B – 40 -59 sq. m. C – 60 - 79 sq. m. D – more than 80 sq. m.
Source: CSO.
till 194413%
1945-197041%
1971-198833%
1989-2002 total sum with
under constraction
housings13%
0-39 sq. m.32%
40-59 sq. m.41%
60-79 sq. m.16%
more that 80 sq. m.11%
1 room8%
2 rooms22%
3 rooms33%
4 and more rooms37%
0% 20% 40% 60% 80% 100%
2002
2003
2004
2005
2006
2007
2009
Municipalities Housing cooperativesWorking places Private personsSocial Building Societies Other entities
38 32
27
30
41
19
33
16
16
37
11
0
10
20
30
40
50
60
70
80
90
100
Households Stock depending onnumber of rooms
Stock depending onfloorage
%
A B C D
125
1.3 The housing market in Warsaw Table 7. Entities operating on the real estate market
Nr 2009 2010 2011 1 Developers1 144 274 266
2 Real estate agents2 1967 2213 2494
3 Notary Offices 3 289 302 305
4 Certified property valuers4 482 517 580
5 Housing cooperatives5 972 953 940
6 Homeowner associations5 7155 7352 7583
Source: estimations of NBP Regional Branch in Warsaw on the basis of data from the Ministry of Transport, Construction and Maritime Economy, The Ministry of Justice, CSO and the Internet.
Table 8. Housing transactions on the primary market in Warsaw. Primary market 2008 2009 2010 2011
Number of contracts 31389 26259 28085 18125
Value in PLN thousands 16988679,76 12452843,94 13390867 8117719,7
Source: estimations of NBP Regional Branch in Warsaw on the basis of the CSO data.
Figure 236. Średni czas sprzedaży mieszkania na rynku wtórnym w Warszawie
Source: NBP Regional Branch in Warsaw.
0
50
100
150
200
250
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
num
be
r o
f da
ys
126
1.3.1 Prices in the housing markets in Warsaw (on the basis of NBP’s database) Table 9. Average housing prices in Warsaw (per 1 sq. m.)
Primary market Secondary market
Asking prices Selling prices Asking
prices Selling prices
Hedonic prices (selling)
Q3 2006 5873 5605 7179 6232 8618
Q4 2006 6095 6186 8751 7143 8670
Q1 2007 7509 7302 9316 7730 8667
Q2 2007 8000 7523 9740 8696 8540
Q3 2007 8740 7879 10078 9137 8638
Q4 2007 9561 8571 9952 9034 8716
Q1 2008 9427 8535 9850 8921 8732
Q2 2008 9300 8611 9783 8546 8428
Q3 2008 9235 8325 9679 8528 8196
Q4 2008 9821 8149 10196 9046 8950
Q1 2009 9880 7543 9626 8406 8626
Q2 2009 9859 7461 10133 8406 8706
Q3 2009 9993 7585 9705 7949 8375
Q4 2009 9915 7680 9671 8497 8620
Q1 2010 9119 8173 9901 8620 8719
Q2 2010 8551 7974 9982 8933 8860
Q3 2010 8295 7940 9788 8493 8698
Q4 2010 8133 7441 9767 8024 8513
Q1 2011 8211 8037 9706 7915 8529
Q2 2011 8396 7527 9472 7920 8475
Q3 2011 8025 7423 9397 7920 8496
Q4 2011 7826 7226 9363 7889 8650
Q1 2012 7879 6967 9111 7601 8418 Source: NBP Regional Branch in Warsaw
Table 10. Index of average price of 1 square meter of housing.
Dynamics Q/Q Dynamics Y/Y
Primary market Secondary market Primary market Secondary market
Asking prices
Selling prices
Asking prices
Selling prices
Hedonic prices
(selling)
Asking prices
Selling prices
Asking prices
Selling prices
Hedonic prices
(selling)
Q3 2006 . . . . . . . . . .
Q4 2006 103,8 110,4 121,9 114,6 100,6 . . . . .
Q1 2007 123,2 118,0 106,5 108,2 100,0 . . . . .
Q2 2007 106,5 103,0 104,6 112,5 98,5 . . . . .
Q3 2007 109,3 104,7 103,5 105,1 101,1 148,8 140,6 140,4 146,6 100,2
Q4 2007 109,4 108,8 98,7 98,9 100,9 156,9 138,6 113,7 126,5 100,5
Q1 2008 98,6 99,6 99,0 98,7 100,2 125,5 116,9 105,7 115,4 100,8
Q2 2008 98,7 100,9 99,3 95,8 96,5 116,3 114,5 100,4 98,3 98,7
Q3 2008 99,3 96,7 98,9 99,8 97,2 105,7 105,7 96,0 93,3 94,9
127
Q4 2008 106,3 97,9 105,3 106,1 109,2 102,7 95,1 102,5 100,1 102,7
Q1 2009 100,6 92,6 94,4 92,9 96,4 104,8 88,4 97,7 94,2 98,8
Q2 2009 99,8 98,9 105,3 100,0 100,9 106,0 86,6 103,6 98,4 103,3
Q3 2009 101,4 101,7 95,8 94,6 96,2 108,2 91,1 100,3 93,2 102,2
Q4 2009 99,2 101,2 99,7 106,9 102,9 101,0 94,2 94,9 93,9 96,3
Q1 2010 92,0 106,4 102,4 101,4 101,1 92,3 108,4 102,9 102,5 101,1
Q2 2010 93,8 97,6 100,8 103,6 101,6 86,7 106,9 98,5 106,3 101,8
Q3 2010 97,0 99,6 98,1 95,1 98,2 83,0 104,7 100,8 106,8 103,9
Q4 2010 98,1 93,7 99,8 94,5 97,9 82,0 96,9 101,0 94,4 98,8
Q1 2011 101,0 108,0 99,4 98,6 100,2 90,0 98,3 98,0 91,8 97,8
Q2 2011 102,3 93,7 97,6 100,1 99,4 98,2 94,4 94,9 88,7 95,7
Q3 2011 95,6 98,6 99,2 100,0 100,2 96,7 93,5 96,0 93,3 97,7
Q4 2011 97,5 97,3 99,6 99,6 101,8 96,2 97,1 95,9 98,3 101,6
Q1 2012 100,7 96,4 97,3 96,4 97,3 96,0 86,7 93,9 96,0 98,7 Source: NBP Regional Branch in Warsaw
Table 11. The average rent price of 1 sq. m. of housing units in Warsaw.
Secondary market
Asking prices
Selling prices Mean
Q3 2006 39,0 36,0 37,5
Q4 2006 45,0 36,0 40,5
Q1 2007 45,0 38,0 41,5
Q2 2007 46,0 37,0 41,5
Q3 2007 49,0 41,0 45,0
Q4 2007 52,0 41,0 46,5
Q1 2008 52,0 43,0 47,5
Q2 2008 52,0 45,0 48,5
Q3 2008 54,0 45,0 49,5
Q4 2008 53,0 47,0 50,0
Q1 2009 51,0 48,0 49,5
Q2 2009 51,0 44,0 47,5
Q3 2009 46,7 41,9 44,3
Q4 2009 47,9 44,8 46,3
Q1 2010 49,4 41,2 45,3
Q2 2010 47,9 41,0 44,4
Q3 2010 48,7 42,2 45,5
Q4 2010 48,9 43,6 46,2
Q1 2011 48,3 41,6 44,9
Q2 2011 47,3 42,1 44,7
Q3 2011 47,9 40,6 44,3
Q4 2011 48,1 41,5 44,8
Q1 2012 48,4 42,7 45,6 Source: NBP Regional Branch in Warsaw
128
1.3.2 Housing prices and characteristics. 1.3.3 The primary market in Warsaw.
Figure 237. Asking prices - primary market
Figure 238. Selling prices – primary market
Notice: maximum price is presented on right-hand axis; Source: NBP Regional Branch in Warsaw.
Source: NBP Regional Branch in Warsaw.
Figure 239. The structure of housing units depending on usable floorage – primary market, asking price
Figure 240. The structure of housing units depending on usable floorage – primary market, selling price
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
Figure 241The structure of housing units depending on the number of rooms – primary market, asking price
Figure 242. The structure of housing units depending on the number of rooms – primary market selling price
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
-30 000
-20 000
-10 000
0
10 000
20 000
30 000
40 000
50 000
60 000
02 0004 0006 0008 000
10 00012 00014 00016 00018 00020 00022 000
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
PLN
/ sq
. m
.
Minimum asking price Average asking price
Median asking price Asking prices limited by RnS
Maximum asking price
0
2 000
4 000
6 000
8 000
10 000
12 000
14 000
16 000
18 000
20 000
22 000
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
PLN
/ s
q.
m.
Minimum selling price Maximum selling price
Average selling price Median selling price
Selling prices limited by RnS
4% 5%
2% 4% 4%
2% 3%
2%
2% 7% 8% 9%
10
%1
3%
8%
6% 6% 6% 6% 6% 6%
4%
4%
35
%3
2%
16
%3
8%
34
%2
6%
30% 35% 36% 4
0%
37% 35
%35
% 38%
38%
38
%3
6%
35
%3
0%2
6%
27%
33
%32
%
31
%3
4%
25
%2
5%2
9%32
%3
1% 33%
30
% 27%
28%
23
%2
3% 21%
25%
26
%2
7%2
8%
27
%2
9%
28% 28
%2
9%
30% 30
%5
7%
33
%3
3%
40%
36%
29%
33% 27
%2
8%
32
%3
2% 28%
29%
30
%31
%3
2%36
%39
%39
%3
5%35
%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
<=40 (40; 60] (60; 80] >80
1% 3%
2%
1% 4% 5%
3%
3% 6% 8%
4%
4% 9% 8% 18
%1
9%
12
%9
% 14
%1
0%
12% 14
%5
%
46
%4
4%
24
%2
6%
46
%52
%4
8%
39
%3
6% 41%
44%
49% 4
8%4
7% 42
%4
5%
51
%45
% 47
%4
6%41
%4
2%
40
%
26
%2
2%
43%
21
%2
0% 2
4%
28
%3
2% 36
% 32% 34% 33
%26
%2
9%
21
%1
6%
20%
27% 21
%2
8%
23
% 23
%26
%
27
%3
1%3
1%
52
% 30
% 19
%2
1%2
6%
23%
19% 17
%1
5%
18%
16
%1
9%
20%
17%
19
%1
8% 17%
25
%2
0%
28%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%Q
3 2
00
6Q
4 2
00
6Q
1 2
00
7Q
2 2
00
7Q
3 2
00
7Q
4 2
00
7Q
1 2
00
8Q
2 2
00
8Q
3 2
00
8Q
4 2
00
8Q
1 2
00
9Q
2 2
00
9Q
3 2
00
9Q
4 2
00
9Q
1 2
01
0Q
2 2
01
0Q
3 2
01
0Q
4 2
01
0Q
1 2
01
1Q
2 2
01
1Q
3 2
01
1Q
4 2
01
1Q
1 2
01
2
<=40 (40; 60] (60; 80] >80
2% 2% 8%
10
%1
% 2% 9%
7%
5%
3%
2% 10
%8
%5
%3
% 7% 12
%8
%8
%7% 6%
4% 4%
3% 4%
21% 3
3% 4
0%
41
%4
0%
38
% 39
%4
2%
38%
36
%4
0% 38
%4
0%
36
%30
% 42
% 40
%38
%3
5%
32
%3
2%3
6%
34
%3
6%
38
%
46
% 40
% 36
%3
3%
40
%40
% 36
%3
4%
40
%37
%3
9% 37
%3
4%
42
%4
0%
36
%3
3%
36
%3
7%
39%
38
%3
9%
41%
40
%39
%
31
%2
5% 16
%1
6%
19
%2
0%
16
%17
%1
7%2
4%
20%
15
%1
8%
18
%2
7% 15
%1
5%
18
%20
%2
2%
24% 21
%2
1%
21
%1
9%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
1 room 2 rooms 3 rooms 4 and more rooms
1% 3% 5% 1
7%
1% 3% 4% 1
7%
1% 4%
4% 10
%1
1%
4% 4% 6% 9% 9% 15
%1
0%
13
%7
%3
%
30
%5
1%
43
%4
4%
27
% 42
% 50
%4
7%
49
%4
6%
37
%4
9%
52
%4
7%
52
%4
5%
47
%4
4% 46
%4
4%
39
%4
9%
45
%
56
%3
6%
43
% 29
%4
7%
39
% 37
% 27
%3
7%
34
%4
5% 2
7% 29
%3
6% 35
%3
6%
34
%3
5% 29
%3
6%
37
%3
1%
35
%
13
%1
0%
10
%1
0%
24
% 16
% 9%
9%
13
%1
6%
15
%1
5% 7%
14
% 9%
12
%1
0%
12
%1
0%
10
%1
1%
13
%1
7%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
1 room 2 rooms 3 rooms 4 rooms
129
Figure 243. Relations of supply and demand, housing units with floorage <= 50 sq.m. and > 50 sq.m., primary market
Source: NBP Regional Branch in Warsaw.
1.3.4 The secondary market
Figure 244. Asking price of sq. m. – secondary market.
Figure 245. Selling price of sq. m. – secondary market.
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
Figure 246. The structure of housing units depending on usable floorage – secondary market, asking price
Figure 247. The structure of housing units depending on usable floorage – secondary market, selling price
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
-70%-60%-50%-40%-30%-20%-10%
0%10%20%30%40%50%
2006
2007
2008
2009
2010
2011
<=50 sq. m. >50 sq. m.
0
5 000
10 000
15 000
20 000
25 000
30 000
35 000
40 000
45 000
Q3
200
6Q
4 20
06
Q1
200
7Q
2 20
07
Q3
200
7Q
4 20
07
Q1
200
8Q
2 20
08
Q3
200
8Q
4 20
08
Q1
200
9Q
2 20
09
Q3
200
9Q
4 20
09
Q1
201
0Q
2 20
10
Q3
201
0Q
4 20
10
Q1
201
1Q
2 20
11
Q3
201
1Q
4 20
11
Q1
201
2
PLN
/ s
q. m
.
minimum maximum mean
median limited by RnS
0
5 000
10 000
15 000
20 000
25 000Q
3 20
06
Q4
2006
Q1
2007
Q2
2007
Q3
2007
Q4
2007
Q1
2008
Q2
2008
Q3
2008
Q4
2008
Q1
2009
Q2
2009
Q3
2009
Q4
2009
Q1
2010
Q2
2010
Q3
2010
Q4
2010
Q1
2011
Q2
2011
Q3
2011
Q4
2011
PLN
/ sq
. m.
minimum maximum mean median hedonic
32% 44%
28
%33
%3
8%3
4% 37% 42
%3
0%
37%
24% 30
%3
4%32
%2
6% 33%
29%
31% 41
%29
%3
1% 31%
35% 37
%
41% 30
%35
%3
1% 28%
33%
27% 28%
39% 40
%43
%3
3% 34
%31
%45
%31
%39
%3
7%3
6%4
4% 43%
42% 41% 40%
16%
12%
15% 16% 14
%1
5%1
7% 17%
20% 1
3%15
%2
2% 21%
26
%1
2%13
% 14% 17% 1
5%17
%18
%1
5% 16
%1
6%
11%
14%
21% 20
%2
0% 18%
19% 14
%1
1% 10%
18% 16
%10
%1
0%1
7%
23%
18% 14% 8
%10
%7
%1
2% 9% 7%
0%
10%
20%30%
40%
50%
60%
70%
80%
90%
100%
stoc
k in
20
02Q
3 20
06Q
4 20
06Q
1 20
07Q
2 20
07Q
3 20
07Q
4 20
07Q
1 20
08Q
2 20
08Q
3 20
08Q
4 20
08Q
1 20
09Q
2 20
09Q
3 20
09Q
4 20
09Q
1 20
10Q
2 20
10Q
3 20
10Q
4 20
10Q
1 20
11Q
2 20
11Q
3 20
11Q
4 20
11Q
1 20
12
<=40 (40; 60] (60; 80] >80
32%
18%
16%
18% 23% 24
%2
0% 21%
22%
24%
20% 23%
14% 18
%1
9%17
%15
%16
%16
%1
7% 18%
19%
19%
19%
41%
33%
35% 38
% 36%
35%
35% 35
%35
%36
%31
% 35%
29% 31
%32
%31
%32
%33
%34
%34
%34
%35
%34
%36
%
16%
24%
22% 24
%2
3% 24%
25%
24% 23% 22
%2
0% 21%
25% 2
5% 24%
25%
26%
25%
24% 24
%24
%24
%24
%2
4%
11%
26%
27% 20
%1
8% 17%
20% 20% 19
%1
9%2
9% 22%
33%
26% 25
%27
%26
%2
6% 26% 26
%23
%2
3%2
3% 22%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
stoc
k in
200
2Q
3 20
06Q
4 20
06Q
1 20
07Q
2 20
07Q
3 20
07Q
4 20
07Q
1 20
08Q
2 20
08Q
3 20
08Q
4 20
08Q
1 20
09Q
2 20
09Q
3 20
09Q
4 20
09Q
1 20
10Q
2 20
10Q
3 20
10Q
4 20
10Q
1 20
11Q
2 20
11Q
3 20
11Q
4 20
11Q
1 20
12
<=40 (40; 60] (60; 80] >80
130
Figure 248. The structure of housing units depending on the number of rooms – secondary market, asking price
Figure 249. The structure of housing units depending on the number of rooms – secondary market, selling price
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
Figure 250. Asking prices depending on type of ownership in Warsaw – secondary market.
Figure 251. Selling prices depending on type of ownership in Warsaw – secondary market.
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
Figure 252. Asking prices depending on housing standards– secondary market.
Figure 253. Selling prices depending on housing standards– secondary market.
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
8%
11
%1
1%
13
%1
5%
15
%1
2%
13
%1
4%
15
%1
4%
14
%9
%1
2%
13
%1
0%
10
%1
1%
11
%1
1%
12
%1
2%
12
%1
1%
21
% 34
%3
5% 38
%3
8%
39
%3
9%
39
%3
9%
39
%3
4%
39
%3
5%
36
%3
7%
37
%3
7%
37
%3
7%
36
%3
6%
37
%3
6%
36
%
33
%3
5%
35
% 34
%3
4%
34
%3
4%
33
%3
3%
33
%3
4% 32
%3
4%
34
%3
4%
35
%3
6%
36
%3
6%
36
%3
5%
35
%3
6%
36
%
38
% 19
%1
9%
15
%1
3%
12
%1
4%
15
%1
5%
13
%1
8%
15
%2
2%
18
%1
7%
18
%1
6%
16
%1
6%
17
%1
7%
16
%1
7%
17
%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
sto
ck in
20
02Q
3 2
00
6Q
4 2
00
6Q
1 2
00
7Q
2 2
00
7Q
3 2
00
7Q
4 2
00
7Q
1 2
00
8Q
2 2
00
8Q
3 2
00
8Q
4 2
00
8Q
1 2
00
9Q
2 2
00
9Q
3 2
00
9Q
4 2
00
9Q
1 2
01
0Q
2 2
01
0Q
3 2
01
0Q
4 2
01
0Q
1 2
01
1Q
2 2
01
1Q
3 2
01
1Q
4 2
01
1Q
1 2
01
2
1 room 2 rooms 3 rooms 4 and more rooms
8%
33
%1
7%
23%
26
%2
4%
25%
25
%20
%2
1%
18
%1
4% 23
%20
%1
3% 23
%2
0%
22
% 31%
19%
19%
17%
21%
22%
21
%3
9%
42
%3
6%
36
%3
6%
37%
40
%4
3%
49%
44
%4
1% 43
%4
4%5
6%
34
%4
2% 40
% 40%
45
%4
7%
47%
45
%4
4%
33%
16%
27
%2
5%
27
%30
%28
%2
6%
27% 23
%25
%3
3% 25
%2
7% 17
%2
6% 26
%2
5% 23%
27
%2
8%
25
%2
8%
24
%2
5%2
9%
38%
12
%1
4%
16
%1
1% 9%
11
%9
%9
% 8%1
2%
12
%9
%1
0%
13
%1
6%
12%
12
% 7%
9% 6%
11% 6% 9%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
stoc
k in
20
02Q
3 2
006
Q4
20
06Q
1 2
007
Q2
20
07Q
3 2
007
Q4
20
07Q
1 2
008
Q2
20
08Q
3 2
008
Q4
20
08Q
1 2
009
Q2
20
09Q
3 2
009
Q4
20
09Q
1 2
010
Q2
20
10Q
3 2
010
Q4
20
10Q
1 2
011
Q2
20
11Q
3 2
011
Q4
20
11Q
1 2
012
Q2
20
12Q
3 2
012
1 room 2 rooms 3 rooms 4 and more rooms
5 000
6 000
7 000
8 000
9 000
10 000
11 000
Q3
2006
Q4
2006
Q1
2007
Q2
2007
Q3
2007
Q4
2007
Q1
2008
Q2
2008
Q3
2008
Q4
2008
Q1
2009
Q2
2009
Q3
2009
Q4
2009
Q1
2010
Q2
2010
Q3
2010
Q4
2010
Q1
2011
Q2
2011
Q3
2011
Q4
2011
Q1
2012
PLN
/ sq
. m.
Cooperative ownership Ownership
5 000
6 000
7 000
8 000
9 000
10 000
11 000
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
PLN
/ sq
. m
.
Cooperative ownership Ownership
5 500
6 500
7 500
8 500
9 500
10 500
11 500
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
PLN
/ sq
. m
.
low standard high
5 500
6 500
7 500
8 500
9 500
10 500
11 500
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
PLN
/ s
q.
m.
low standard high
131
Figure 254.Asking prices depending on the floor level– secondary market.
Figure 255. Selling prices depending on the floor level– secondary market.
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
Figure 256.Asking prices depending on the construction type – secondary market
Figure 257. Selling prices depending on the construction type – secondary market
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
Figure 258. Asking prices depending on the year of the construction – secondary market.
Figure 259. Selling prices depending on the year of the construction – secondary market.
Source: NBP Regional Branch in Warsaw. Source: NBP Regional Branch in Warsaw.
6 000
6 500
7 000
7 500
8 000
8 500
9 000
9 500
10 000
10 500
11 000Q
3 2
00
6Q
4 2
00
6Q
1 2
00
7Q
2 2
00
7Q
3 2
00
7Q
4 2
00
7Q
1 2
00
8Q
2 2
00
8Q
3 2
00
8Q
4 2
00
8Q
1 2
00
9Q
2 2
00
9Q
3 2
00
9Q
4 2
00
9Q
1 2
01
0Q
2 2
01
0Q
3 2
01
0Q
4 2
01
0Q
1 2
01
1Q
2 2
01
1Q
3 2
01
1Q
4 2
01
1Q
1 2
01
2
PLN
/ s
q.
m.
<= 5 6 - 12
6 000
6 500
7 000
7 500
8 000
8 500
9 000
9 500
10 000
10 500
11 000
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
PLN
/ sq
. m
.
<= 5 6 - 12
5 000
6 000
7 000
8 000
9 000
10 000
11 000
12 000
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
PLN
/ sq
. m
.
brick mixed prefabricated
5 000
6 000
7 000
8 000
9 000
10 000
11 000
12 000
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
PLN
/ s
q.
m.
brick mixed prefabricated
5 000
6 000
7 000
8 000
9 000
10 000
11 000
Q3
20
06
Q4
20
06
Q1
20
07
Q2
20
07
Q3
20
07
Q4
20
07
Q1
20
08
Q2
20
08
Q3
20
08
Q4
20
08
Q1
20
09
Q2
20
09
Q3
20
09
Q4
20
09
Q1
20
10
Q2
20
10
Q3
20
10
Q4
20
10
Q1
20
11
Q2
20
11
Q3
20
11
Q4
20
11
Q1
20
12
PLN
/ sq
. m
.
1945-1970 1971-1978 1979-1988
1989-2002 after 2002
5 000
6 000
7 000
8 000
9 000
10 000
11 000
Q3
20
06Q
4 2
006
Q1
20
07Q
2 2
007
Q3
20
07Q
4 2
007
Q1
20
08Q
2 2
008
Q3
20
08Q
4 2
008
Q1
20
09Q
2 2
009
Q3
20
09Q
4 2
009
Q1
20
10Q
2 2
010
Q3
20
10Q
4 2
010
Q1
20
11Q
2 2
011
Q3
20
11Q
4 2
011
Q1
20
12
PLN
/ sq
. m
.
1945-1970 1971-1978 1979-19881989-2002 after 2002
