View
17
Download
0
Category
Preview:
Citation preview
Introduction to Macroeconomics
Vivaldo M. Mendes
a ISCTE�IUL � Department of Economics
September 2009
(Vivaldo M. Mendes ) Macroeconomics September 2009 1 / 60
Useful information
(Vivaldo M. Mendes ) Macroeconomics September 2009 2 / 60
Useful information
Useful information
Lecturers:
Vivaldo Mendes (vivaldo.mendes@iscte.pt)António Machado Lopes (machado.lopes@ine.pt)
O¢ ce: Room 237 (Building II)
Note: Until the mid November, VM may be more easily found at"ISCTE President�s O¢ ce" (Building I)
Phone numbers: 773271 for room 327; 217903046/47 for theISCTE President�s O¢ ce; 218426100 for INE
Classes: Tuesdays, 18.00h � C407 or on a Computer Lab (to beannounced)
Course homepage: there will be one operational soon, with newsand materials online
address: http://cm.de.iscte.pt/
(Vivaldo M. Mendes ) Macroeconomics September 2009 3 / 60
Useful information
Grading
The grading process includes three major items:
Midterm Exam (30%): There will be one midterm examination extraclass on a date to arrangeFinal Exam (40%): The �nal exam will be on January 2010A group essay (30%) over a subject discussed in the course
Active participation in classes will be welcome, it�s very useful forlearning and grading
The group essay should not be developed under "self�management"
(Vivaldo M. Mendes ) Macroeconomics September 2009 4 / 60
Useful information
Teaching approach
The teaching material requires 10 weeks
Some topics will be covered in just one week
Some topics requires two weeks
The core theoretical parts (the last two topics) will be covered in �veweeks
The teaching follows a step ladder approach
So if you miss one step, it�s more di¢ cult to put your feet on the next...
Oriented towards "how to do"
students are expected to �nish their Master thesisso, they need to master practical tools ...not just descriptive general knowledge
(Vivaldo M. Mendes ) Macroeconomics September 2009 5 / 60
Useful information
Teaching approach (cont.)
We will make use of the computer as much as possible (Matlab)
Good knowledge of mathematics: it helps, however it is not enough
Lack of good knowledge of mathematics: do not worry, you will notbe left behind ...
... if you pay careful attention when the crucial maths concepts areintroduced in classes... if you are willing to do some extra study
The course is intended to be "self�contained"
The only maths that matters are basic knowledge of:
DerivativesDi¤erence equationsOptimization (Lagrangean)Matrices
(Vivaldo M. Mendes ) Macroeconomics September 2009 6 / 60
Useful information
The textbook
There is no adopted textbookA set of publicly available lecture notes will be provided (topic bytopic)Main reasons:
Students save timeLecture notes are "tailored" to each topicMajor available textbooks require a much lengthier course (not just 20hours course)
Some major postgraduate macro textbooks available:
Michael Wickens (2008). Macroeconomic Theory:A Dynamic GeneralEquilibrium Approach, Princeton University PressLars Ljungqvist and Tom Sargent (2004). Recursive MacroeconomicTheory, 2nd edition, MIT Press
Stephen Williamson (2008), Macroeconomics, 3rd Edition, PrenticeHall � a good choice for those with little economics background
(Vivaldo M. Mendes ) Macroeconomics September 2009 7 / 60
Useful information
A quick guided tour
1 Introduction to macroeconomics (1 week)2 Introduction to Matlab (1 week)3 Major stylized facts about business cycles (1 week)4 Solving models with rational expectations (2 weeks)5 The Real Business Cycle model (2 weeks)6 Commitment and credibility in monetary policy (1 week)7 The New Keynesian model (2 weeks)
(Vivaldo M. Mendes ) Macroeconomics September 2009 8 / 60
Useful information
What is macroeconomics?
(Vivaldo M. Mendes ) Macroeconomics September 2009 9 / 60
What is macroeconomics?
What is macroeconomics?
Analyses the interactions among 4 aggregate markets:
Market for goods&servicesMoney and �nancial marketExchange or FX marketLabor market
... to explain how prices and quantities are determined in eachmarket in interaction
... considering di¤erent time periods for the analysis:
Short term: business cyclesLong term: economic growth, sustainability of social security and ofpublic debt, among others
Please, don�t mix up short term and long term phenomena
(Vivaldo M. Mendes ) Macroeconomics September 2009 10 / 60
What is macroeconomics?
What is macroeconomics? (continued)
... considering di¤erent crucial assumptions:
Markets work perfectly well: Classical macroeconomicsMarkets show large imperfections: Keynesian macroeconomics
To give answers about economic policy: how and whether thegovernment should intervene in the functioning of the economy inorder to increase social welfare:
Classicals: Active economic policy is undesirable ... reduces socialwelfareKeynesians: Market failures are pervasive and require active publicintervention ... to boost social welfare
Try to get this clear opposition between the two schools in your mind
(Vivaldo M. Mendes ) Macroeconomics September 2009 11 / 60
What is macroeconomics?
Microeconomics vs Macroeconomics
Microeconomics
The study of how individual agents behave (a household, a �rm)Addresses partial equilibria: how prices and quantities are determinedin individual marketsThere are no agents with political power to determine outcomes
Macroeconomics:
The study of how aggregate economies behave (all households and all�rms, and their interactions in all markets).Addresses general equilibria: how prices and quantities aresimultaneously determined in all marketsIt includes the possibility of considering the political context ofeconomics: Trade unions, Employers associations, etc...
(Vivaldo M. Mendes ) Macroeconomics September 2009 12 / 60
What is macroeconomics?
Major agents in macroeconomics
Private agents
HouseholdsFirmsBanks (money) and �nancial institutions (�nancial assets)
Agents with a political power
GovernmentCentral Bank (in all OECD countries, this bank is totally independentfrom the Government)Trade Unions and Employers Associations
Foreign agents (agents living in foreign economies)
(Vivaldo M. Mendes ) Macroeconomics September 2009 13 / 60
What is macroeconomics?
The current sate of macro
(Vivaldo M. Mendes ) Macroeconomics September 2009 14 / 60
The current state of macro
The current state of macro
Over the last 30 years, macroeconomics was in big turmoilHowever, a new consensus in macroeconomics has emerged ...developed over the last 10 years or so ...Jordi Gali (2000). �New Perspectives on Monetary Policy, In�ation and theBusiness Cycle�, Dep. Economics, Universitat Pompeu Fabra, Barcelona.
�The �eld of macroeconomics has witnessed in recent years thedevelopment of a new generation of small-scale monetary businesscycle models, generally referred to as New Keynesian (NK) models orNew Neoclassical Synthesis models ... [integrating] Keynesian elements(imperfect competition, and nominal rigidities) into a dynamic generalequilibrium framework that until recently was largely associated withthe Real Business Cycle (RBC) paradigm. They can be used (and arebeing used) to analyze the connection between money, in�ation, andthe business cycle, and to assess the desirability of alternative monetarypolicies�. (page 1)
Crucial words: New Synthesis, Keynesian, RBC, dynamic generalequilibrium, business cycles
(Vivaldo M. Mendes ) Macroeconomics September 2009 15 / 60
The current state of macro
The First (Old) Neoclassical Synthesis
"Macroeconomics" is a relatively young subject: "born" in the mid1940�s
1946: the �rst time the term "macroeconomics" were used in onetitle (vide Fig 1)Macroeconomics were dominated by Keynesian ideas up to the early1070�s
The �rst Neoclassical Synthesis: Keynesian/Classical dichotomy
The economy "is" Keynesian in the short term: there is a permanenttrade-o¤ between in�ation and unemployment that can be exploited bypolicy makersThe economy "is" Classical in the long term: no such permanenttrade-o¤ exists
In the late 1960�s, serious problems with the Synthesis becameevident: empirically and conceptually
(Vivaldo M. Mendes ) Macroeconomics September 2009 16 / 60
The current state of macro
Figure 1: �rst time "macroeconomics" used in a title
(Vivaldo M. Mendes ) Macroeconomics September 2009 17 / 60
The current state of macro
Conceptual problems with the Old Synthesis
No microeconomic foundations: most functions in the model weretotally ad-hocThe model had an intrinsically linear structure, which led to manyinteresting issues to be aside from the model-economy (multipleequilibria, self-ful�lling prophecies, etc..)Backward looking expectations: private agents produce systematicmistakes in their forecasting exercisesIrrationality: policy makers were fully-rational agents and knew howthe economy works; private agents were "irrational" with littleknowledge of how the economy worksTotal nonsense to admit that the Central Bank could managemonetary policy to permanently exploit the trade-o¤ betweenin�ation and unemploymentVulnerable to the Lucas critique: if policy makers intervene in theeconomy, private agents react by changing their choices, so thestructure of the economy changes and the public intervention hasperverse e¤ects(Vivaldo M. Mendes ) Macroeconomics September 2009 18 / 60
The current state of macro
Empirical problems with the Old Synthesis
1 Real wages are countercyclical in the model, but procyclical in theeconomy
2 The early 1070�s put in evidence a very unpleasant reality to whichthe model could provide no remedy: higher and higher unemploymentand in�ation rates (stag�ation)
3 Public debt increased permanently in almost all OECD countries, withlittle evidence of a decline in unemployment
4 Central Banks lost the control of monetary aggregates5 The model could hardly reproduce the basic stylized facts from thebusiness cycles (variances, covariances, etc..)
(Vivaldo M. Mendes ) Macroeconomics September 2009 19 / 60
The current state of macro
30 years of revolutions and counter-revolutions
The Old Synthesis stand for the 1950�s and the golden 1960�sSargent and Lucas: launched the New-Classical model in the early1970�s
Macro with microeconomic foundations
Problems with New-Classical model ... led to the Real BusinessCycles (RBC) model in the early 1980�s
Finn Kydland and Edward Prescott (1982), Time to Build andAggregate Fluctuations, Econometrica, 50, 1345�1370)
Problems with the RBC led to the development of the NewKeynesian Model (or the New Synthesis) in the mid 1990�s:
Yun, T. (1996). Nominal Price Rigidity, Money Supply Endogeneity,and Business Cycles, Journal of Monetary Economics, 37 (April),345�70King, R. (2000). The New IS-LM Model: Language, Logic, and Limits,Federal Reserve Bank of Richmond Economic Quarterly, 86 (3),45�104.
(Vivaldo M. Mendes ) Macroeconomics September 2009 20 / 60
The current state of macro
Main ingredients of the New Synthesis
It is largely built upon the Old Keynesian framework
... with the usual nominal and real rigidities in price setting
... but without the problems that pushed the model to seriousproblems in the early 70s
It has the same functions (IS, LM, Agregate Supply) ...With some new arguments, like "forward looking or rationalexpectations" instead of "adaptive expectations", "Calvo pricing",maximization of utility, and so on ...
Built upon a general equilibrium framework, based on soundmicroeconomic principles,
Relies a lot on quantitative simulations like the Real Business Cycleliterature
But contrary to RBC, it ends up having a key role to monetarypolicy and a signi�cant role for �scal policy
(Vivaldo M. Mendes ) Macroeconomics September 2009 21 / 60
The current state of macro
Major predictions of the New Synthesis
Four basic predictions:
the instrument of monetary policy ought to be the short term interestrate,policy should be focused on the control of in�ationin�ation can be reduced by aggressively increasing short term interestratesthe central bank should conduct monetary policy adopting a strategy ofcommitment in a forward-looking environment, instead of discretion
The Old model�s predictions up-side-down!!!
See Figure 2.
(Vivaldo M. Mendes ) Macroeconomics September 2009 22 / 60
The current state of macro
Active interest rate policy by central banks
The FED now reacts much more aggressively to in�ation than in the "oldtimes"
1960 1965 1970 1975 1980 1985 1990 1995 2000
.02
.04
.06
.08
.1
.12
"FED Funds Rate": 9.7%"FED Funds Rate": 4.8%
Indice Preços Implicito no PIB (trimestral)IPIPIB
(Vivaldo M. Mendes ) Macroeconomics September 2009 23 / 60
The current state of macro
A picture of the success of the receipt (I)
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000
7.5
8
8.5
9 LGDP hpLGDP
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000
.05
.025
0
.025GDPResid
(Vivaldo M. Mendes ) Macroeconomics September 2009 24 / 60
The current state of macro
A picture of the success of the receipt (II)
1985 1990 1995 2000
0
.01
.02G(RealGDP)
1985 1990 1995 2000.01
0
.01
.02 G(RealWages)
1950 1960 1970 1980 1990 2000
.02
0
.02
.04 G(RealGDP)
1950 1960 1970 1980 1990 2000
.02
0
.02
.04G(RealWages)
(Vivaldo M. Mendes ) Macroeconomics September 2009 25 / 60
The current state of macro
Macroeconomic modelsvs
economic reality
(Vivaldo M. Mendes ) Macroeconomics September 2009 26 / 60
The current state of macro
The need for simplicity
A modern market economy is a hugely complex structureThere are thousands of �rms, millions of households, thousands ofgoods and prices, there is a government, a central bank, trade unionsand employers associations, a foreign sector
All this occurs in 4 major markets in permanent interaction
Moreover, agents are rational and try to anticipate the future (theyformulate expectations)
Self-ful�lling expectations (in the good and the bad sense)Herding behavior (euphoria and panic)
It is totally impossible to include all this complexity in a singleeconomic model
So, any macroeconomic model is necessarily an abstractsimpli�cation of economic reality
(Vivaldo M. Mendes ) Macroeconomics September 2009 27 / 60
The current state of macro
The need for simplicity (cont.)
Despite being necessarily an abstract simpli�cation of economicreality
However, models should not omit de�ning features of that reality ...for the sake of simplicity, e.g.:
if prices are sticky, it looks questionable to assume that they areperfectly �exible in the modelif agents make mistakes when forecasting the future, it�s no good toassume in the model that expectations are always ful�lledif there are agents which have market power, it looks bad if the modelignores that important feature
What makes a good model? Good assumptions and good predictions... and simplicity
(Vivaldo M. Mendes ) Macroeconomics September 2009 28 / 60
The current state of macro
The main ingredients of a macroeconomic model
The agents that are crucial players in the aggregate economyThe behavioral assumptions about the macro agents: they try tooptimize their objectives given the information they have and theconstraints they face:
Consumers: max utility based on their preferences over goodsFirms: max pro�tsGovernment: max social utility (or votes in the next election??)Central Bank: min in�ation; min unemploymentTrade Unions: max real wages
The constraints that the crucial players faceAn assumption about the information set available to the agentsThe set of markets and assets produced in the economyThe technology that is available for producing goods and servicesA de�nition of equilibrium in order to close the model
(Vivaldo M. Mendes ) Macroeconomics September 2009 29 / 60
The current state of macro
Two notions of equilibrium
A competitive or Walrasian equilibrium occurs when:Firms and consumers take all prices as given (no market power)There is perfect information concerning all relevant issues in the marketThere are no externalitiesTherefore, prices convey all relevant information, are fully �exible, suchthat supply = demand in all markets (market clearing).
An imperfectly competitive or Non�Walrasian equilibrium occurswhen:
Firms are price makers (they have market power)Information is imperfectly disseminated throughout the marketsThere may be externalitiesTherefore, prices may not convey all relevant information, adjust slowly,and thus supply may be di¤erent from demand in all markets(non�market clearing).
Classicals: the economy tends to work under competitive conditionsKeynesians: the economy tends to work under imperfectlycompetitive conditions(Vivaldo M. Mendes ) Macroeconomics September 2009 30 / 60
The current state of macro
Perfect Markets: example
An increase in demand in a perfectly competitive market (prices are fully�exible to clear the market)
P
P1
Qs
Q
Qd1Qd
0
Qd1Q1Q0
P0 B
C
A
(Vivaldo M. Mendes ) Macroeconomics September 2009 31 / 60
The current state of macro
Imperfect Markets: example
An increase in demand in an imperfectly competitive market (prices arerigid and the market is not cleared without the intervention ofGovernment)
PQs
Q
Qd1Qd
0
Qd1Q0
P0 BA
Rationing due togovernment intervention
Qs + governmentsupply
Government intervention
(Vivaldo M. Mendes ) Macroeconomics September 2009 32 / 60
The current state of macro
Some useful concepts: a review
(Vivaldo M. Mendes ) Macroeconomics September 2009 33 / 60
The Phillips Curve
The Phillips Curve in the short term
In the 1950�s, Almarin Phillips discovered a major stylized fact, widelywidespread over all OECD economies
This fact (known as the Phillips Curve) may be put like this:There is a short term negative trade-o¤ between unemployment andin�ationIf a government wants to reduce unemployment, has to accept higherrates of in�ation
However, this relationship seems questionable over the long term.Why?If the government intervention leads to higher in�ation, the nextperiod workers will ask for higher wages:
costs will move up for �rms, and prices will move up again
So, the higher are the expectations about in�ation, the higher will bethe level of in�ation for a given level of unemployment
(Vivaldo M. Mendes ) Macroeconomics September 2009 34 / 60
The Phillips Curve
The short term Phillips Curve
π
uu0
6%
A
B
Phillips Curve:short term shifts
• C
u*
4%
e%5π
e%4π
e%6π
•
•
D
5%
•
(Vivaldo M. Mendes ) Macroeconomics September 2009 35 / 60
The Phillips Curve
The long term Phillips Curve
The reasoning that we applied to the short term can not be applied tothe long termIn the long term
even if the government tries to exploit the trade-o¤ betweenunemployment and in�ationunemployment can not go lower than its natural or full employmentlevel (u�)
Therefore, the curve becomes verticalMajor message:
even if in the short term there is a trade-o¤ between unemploymentand in�ation,and governments may try to exploit this trade-o¤ for economic orpolitical reasonsin the long term, no such trade-o¤ exists
See next �gure
(Vivaldo M. Mendes ) Macroeconomics September 2009 36 / 60
The Phillips Curve
The long term Phillips Curve
π
uu0
6%
A
B
Phillips Curve:short term shifts
• C
u*
4%
e%5π
e%4π
e%6π
•
•
D
5%
•
Phillips Curve inthe long term
(Vivaldo M. Mendes ) Macroeconomics September 2009 37 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
Dynamics and how to model it
In modern macroeconomics ... everything is discussed using dynamicframeworksDynamics can be modelled under three major ways:
Di¤erence equationsDi¤erential equationsPartial di¤erential equations
Examples (following the previous order)
kt � kt�1| {z }∆k
= a � kt�1 � c � kαt�1 , a, c, α as parameters (1)
∂kt
∂t|{z}∆k
= a � kt � c � kαt (2)
∂kt
∂t|{z}∆k
= a� c � ∂kt
∂ut(3)
Solutions: most di¤erence and di¤erential eq. can be easily solved,partial di¤erential equations very hard
(Vivaldo M. Mendes ) Macroeconomics September 2009 38 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
Major ingredients of a dynamic model
The initial conditionThe transitional dynamics processThe steady state, or the long term equilibrium (or the �xed point)
De�nition
Long term equilibrium (LTE). Este equilíbrio é de�nido como um estadoem que as variáveis endógenas crescem a uma taxa constante, a qual podeser positiva, nula, ou negativa, e em que o período temporal considerado éo longo prazo (tempo pode variar entre 0 e ∞).
De�nitionTransitional dynamics. Representa o ajustamento das variáveisendógenas entre a condição inicial e o equilíbrio de longo prazo � ouentre dois equilíbrios de longo prazo, se houver uma alteração numa (ouvárias) das variáveis exógenas � . Neste processo de transição, as variáveiseconómicas endógenas podem crescer a taxas crescentes ou decrescentes.
(Vivaldo M. Mendes ) Macroeconomics September 2009 39 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
Major ingredients of a dynamic model: graphically
altitudeinicial
altitude
tempo
E
F
t1
•
•
•
t0
Equilíbrio de longoprazo (vôo)
Situaçãoinicial (vôo)
Processo de transição dinâmicaaltitudede vôo
(Vivaldo M. Mendes ) Macroeconomics September 2009 40 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
The three major questions about the LTE
In the analyses of any kind of dynamic process there are three crucialquestions:
Does the process has a LTE?If the LTE exists, is it stable or unstable?If the LTE exists, it is unique or there are many equilibria?
To answer the �rst question, we have just to impose the followingcondition to our previous equations
∆k = 0
For our di¤erence equation (x? is the LTE level of xt) we have
xt+1 = xt = x?
Intuition:xt+1 > xt =) xt "xt+1 < xt =) xt #xt+1 = xt =) xt = x? (constant)
(Vivaldo M. Mendes ) Macroeconomics September 2009 41 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
The three major questions about the LTE (cont.)
... three crucial questions:
............................................................?If the LTE exists, is it stable or unstable?If the LTE exists, is it unique or there are many equilibria?
To provide formal answers to the second and third questions wouldtake us a long time
So we will provide examples which will give intuition about the majorpoints
As you will see no major maths is required to understand thefollowing points
(Vivaldo M. Mendes ) Macroeconomics September 2009 42 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
The steady state or the LTE
Assume that our system can be represented by a rather simpleequation
xt+1 = 10+ 0.5 � xt
The answer to the three questions depend on parameter a :
jaj < 1jaj > 1jaj = 1
First case: a = 0.5.Second case: a = 1.5Third case: a = 1.
(Vivaldo M. Mendes ) Macroeconomics September 2009 43 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
Our �rst case: a=0.5
If we havext+1 = 10+ 0.5 � xt (4)
The LTE can be easily calculated
xt+1 = xt = x?
x? = 10+ 0.5 � x?
x? = 20
Answers:
The LTE exists: x? = 20It is unique (see next �gure)It is stable: no matter what the initial conditions may be, it tendstowards the LTE
(Vivaldo M. Mendes ) Macroeconomics September 2009 44 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
LTE: exists, unique and stable
0 5 10 15 20 25 30 35 400
5
10
15
20
25
30
35
40 45ºx(t+1)
x(t)
Equilíbrio estável
Condição inicial ACondição inicial B
(Vivaldo M. Mendes ) Macroeconomics September 2009 45 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
Our second case: a=1.5
Now, our system can be represented by
xt+1 = 10+ 1.5 � xt (5)
The LTE can be easily calculated
xt+1 = xt = x?
x? = 10+ 1.5 � x?
x? = �20
Answers:The LTE exists: x? = �20It is unique (see next �gure)It is unstable: the LTE only exists if the initial condition equals exactlythe value of x? = �20. If there is a minor shock that forces theprocess to move away (even if only in�nitesimally) from the LTE, theprocess diverges further and further away from the LTE.
(Vivaldo M. Mendes ) Macroeconomics September 2009 46 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
LTE: exists, unique but unstable
100 80 60 40 20 0 20 40 60 80 100150
100
50
0
50
100
150
Equilíbrio instável
5
20
Condição inicial A
Condição inicial B
O
x(t+1)
x(t)
45º
(Vivaldo M. Mendes ) Macroeconomics September 2009 47 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
Our third case: a=1
Now, our system can be represented by
xt+1 = 10+ 1 � xt (6)
The LTE can be easily calculated
xt+1 = xt = x?
x? = 10+ 1 � x?
0 � x? 6= �20
Answers:
The LTE does not exists: no value of x? satis�es 0 � x? = �20Is it unique? Redundant answer (see next �gure)Is it stable? Redundant answer
(Vivaldo M. Mendes ) Macroeconomics September 2009 48 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
LTE: it does not exist
20 10 0 10 20 30 40 5020
10
0
10
20
30
40
50
60
Não existe equilíbrio neste processo dinâmico
x(t+1)
x(t)
Condição inicial A
Condição inicial B
15
45º
(Vivaldo M. Mendes ) Macroeconomics September 2009 49 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
A fourth case: multiple equilibria
Consider that a process can be modelled by the nonlinear di¤erenceequation
yt =a1y3
t�1 � a2y2t�1
2(7)
Assume that the parameters are: a1 = �1, a2 = 3. The LTE iscalculated as
yt = yt�1 = y?
Answers:
The LTE exists? YES! The three roots from that equation are :y?A = 0, y?B = 1, y?C = 2Is it unique? NO!! We have three values that satisfy the LTE condition(see next �gure)Is it unstable:we get two stabe LTE and one unstable (y?B = 1)
(Vivaldo M. Mendes ) Macroeconomics September 2009 50 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
LTE: exists, it is not unique, some stable and someunstable
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 20
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
2
y(t1)
y(t)
5 10 15 20
0
1
2
tempo
y(t)
(Vivaldo M. Mendes ) Macroeconomics September 2009 51 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
A recipe for tragedy: mixing up short term with long term
Please, never mix up short term (transitional dynamics) with longterm phenomena (LTE)This is probably the most common and dangerous mistake inmacroeconomicsA simple example. The evolution of the public debt as a percentageof real GDP (zt)
zt = ψ+ τ � zt�1 , t = 0, 1, 2, ... (8)
ψ is the primary de�cit as a percentage of real GDP; τ = 1+r1+g , with g
as the growth rate of real GDP, and r as the real interest rateAssume g = 3.5%, r = 3%, and two scenarios: ψA = 0.01, ψB = 0.02.(see next �gure)Assume also that between t = 600 and t = 620 the governmentincreases the primary de�cit: what happens to the LTE?Looking at panel (1) we will get a di¤erent answer to that of panel(2), wich applies also to (3) and (4).(Vivaldo M. Mendes ) Macroeconomics September 2009 52 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
Don not mix up short term with long term
0 4 8 12 16 20 24 28 32
.1
.2
.3
.4
.5 (1) Cenário B
Cenário A
0 150 300 450 600 750
.5
1
1.5
2
(2)
Cenário B
Cenário A
590 600 610 620 6301
1.4
1.8
2.2(3)
Cenário A um choque
0 150 300 450 600 750 900
.5
1
1.5
2
2.5
3 (4)
Cenário B
Cenário A
(Vivaldo M. Mendes ) Macroeconomics September 2009 53 / 60
The "Steady Sate" or Long Term Equilibrium (LTE)
The importance of sustainable long termgrowth
(Vivaldo M. Mendes ) Macroeconomics September 2009 54 / 60
The importance of long term growth
Growth rates: basics
Assume the following problem:
I have 10e todayI receive an interest of 4% per year on themHow much are they worth o¤ in t years time?
Mathematically speaking:
y1 = y0(1+ g) (9)
y2 = y1(1+ g) = y0(1+ g)(1+ g) = y0(1+ g)2
.
.yt = y0(1+ g)t (10)
(Vivaldo M. Mendes ) Macroeconomics September 2009 55 / 60
The importance of long term growth
Exponential growth and its log face: g = 4%, y(0) = 10
0 30 60 90 1200
200
400
600
800
1000
1200
0 30 60 90 1202
2.5
3
3.5
4
4.5
5
5.5
6
6.5
7
ln y(t)y(t) = y(0)*(1+r) t , y(0)= 10, r =4%
(Vivaldo M. Mendes ) Macroeconomics September 2009 56 / 60
The importance of long term growth
Exponential growth and its log face (cont.)
When we have exponential growth, this can be linearized just byapplying natural logsLogs are applied everywhere in macro ... makes live a lot easierApply logs to equation (9), and get
ln y1 = ln y0 + 1� ln(1+ g)
As ln(1+ g) � g for small values of g (try it with your calculator for�1 < g < 1), then
ln y1 � ln y0 � g
What about average growth rates? Apply logs to equation (10) and
ln yt = ln y0 + t� ln(1+ g)
orln yt � ln y0
t� g
(Vivaldo M. Mendes ) Macroeconomics September 2009 57 / 60
The importance of long term growth
The importance of sustainable growth
Assume 3 economies (A,B,C) all starting at t(0) with an income percapita of 10e
What happens to their income per capita in 60 years, if the annualgrowth rates of their income per capita are as follows:
g(A) = 2%, g(B) = 4%, g(C) = 6%
Do the calculations, and look at the next �gure
Growth miracles and growth disasters: no mere rhetoric ... look at a�gure and compare
Argentina with JapanBotswana and Benin
(Vivaldo M. Mendes ) Macroeconomics September 2009 58 / 60
The importance of long term growth
The importance of sustainable growth (cont.)
5 10 15 20 25 30 35 40 45 50 55 600
50
100
150
200
250
300
350
g = 2%
g = 4%
g = 6%
y(0) = 10
(Vivaldo M. Mendes ) Macroeconomics September 2009 59 / 60
The importance of long term growth
Growth miracles and growth disasters
(Vivaldo M. Mendes ) Macroeconomics September 2009 60 / 60
Recommended