Literature review of past crises · 1 FINANCIAL CRISES —A REVIEW OF THE LITERATURE Main Findings z While a major fi nancial crisis is a novelty for policy makers in the US and major

InternationalLabourOrganization

European Union

InternationalInstitute forLabour Studies

Literature reviewof past crises

EC-IILS JOINT DISCUSSION PAPER SERIES No. 1

LITERATURE REVIEW OF PAST CRISES

INTERNATIONAL LABOUR ORGANIZATIONINTERNATIONAL INSTITUTE FOR LABOUR STUDIES

LITERATURE REVIEW OF PAST CRISESFinancial crises – A review of literature

Abstract This paper is part of a series of discussion papers that have been prepared by the International Institute for Labour Studies (IILS) within the framework of the joint project “Addressing European labour market and social challenges for a sustainable globalization”, which has been carried out by the European Commission (EC) and the International Labour Organization (ILO). The discussion paper series provides background information and in-depth analysis for two concluding synthesis reports that summarize the main findings of the project. This paper relates to first part of the project “Addressing the short- and medium-term labour market and social challenges of the current economic and financial crisis” and the concluding synthesis report “Building a sustainable job-rich recovery”.

There is a robust modern literature on financial crises that policy makers in newly affected countries can potentially draw from. The existing literature has yet to fully account for the large size of the crises and the long run effects, but still there are a number of findings that may help guide policy in the context of the current crisis. This paper is an attempt to draw the main lessons from this literature, with a special focus for the macro-economic impact of financial crises.

vii

TABLE OF CONTENTS

Main fi ndings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

A. Defi ning and identifying fi nancial crises . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4B. Th e boom before the bust . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6C. Th e bust in the fi nancial sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11D. Supply side eff ects: Misallocation in capital and labour markets . . . . . . . . 14E. Demand side channels and macroeconomic policy . . . . . . . . . . . . . . . . . . . . 18F. Policy considerations and rationale for the next chapters . . . . . . . . . . . . . . 21References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

List of fi guresFigure 1.1 Origins of the fi nancial and economic crisis, 2007-09 . . . . . . . . . . . . . 7Figure 1.2 Boom before the bust: PE ratio and long-term interest rates in the US . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Figure 1.3 Ratio of debt to GDP among select advanced economies . . . . . . . . . . 9Figure 1.4 CEO compensation based on Forbes annual survey . . . . . . . . . . . . . . . 10Figure 1.5 Corporate spreads in the US and EU (basis points) . . . . . . . . . . . . . . . 16

vii

1

FINANCIAL CRISES—A REVIEW OF THELITERATURE

Main FindingsWhile a major fi nancial crisis is a novelty for policy makers in the US and major

European Countries, other parts of the world, mainly emerging economies, have been regularly hit by fi nancial crises since at least the 1980s (see chapters 2 and 3).

Th ese crises have had severe impacts on labour markets, mainly through the cap-

ital allocation channel but also because fi nancial crises are larger versions of usual economic downturns. Available empirical evidence suggests that labour market outcomes during fi nancial crises are worse than in regular downturns, even aft er controlling for the size of the recession. In particular, studies of past crises show that when cost of fi nancing goes up, job creation suff ers.

In terms of empirical evidence for misallocation, studies show that there is a

relatively large drop in manufacturing and mining, which have high produc-tivity, compared to services and agriculture, which tend to have low produc-tivity. Furthermore, misallocation can persist in the medium term, especially if the problems in the fi nancial system are not properly addressed.

Th e recent crisis (2007-09) showed that both during fi nancial crises and in

the recovery phase, unemployment is lower than what would usually be war-ranted by the observed drop in output. Furthermore, studies from past crises show that while in the normal business cycle, old jobs were kept in place, but in fi nancial crises there was a large degree of job destruction without a commen-surate increase in job creation.

Th e large impact of fi nancial crises on the labour market can generate struc-

tural adjustments, like increase in long-term unemployment. Since it is diffi cult

LITERATURE REVIEW OF PAST CRISIS

2

to reintegrate these workers back into the labour force (skills erosion, labour market detachment, etc.), it could lead to “hysteresis,” that is, the tendency of large cyclical unemployment outcomes to translate into a permanent increase in the natural rate of unemployment.

Not surprisingly, there is a vast literature on fi nancial crisis and one story that

emerges clearly is that financial crisis is a reasonably well-defined economic problem, and it is avoidable with the right set of policies. In order to understand a fi nancial crisis, it is important to look at the “boom” period that usually precedes a crisis – two main strands of the literature on crises that attempt to account for this period. Th e fi rst view states that the “boom-bust” cycle is evidence of excessive investment and risk taking (facilitated by easy monetary policy). Th e second per-spective presupposes that infl ated prices in assets increases liquidity – rather than excessive investment – and facilitates investment.

Whichever strand of the literature one subscribes to, what is clear is that “boom”

periods are usually followed by “bust”. Th at is, “bubbles” tend to ultimately lead to “bust” when asset prices drop precipitously and fi nancial markets are no longer functional. Financial markets become dysfunctional because of liquidity prob-lems (when fi rms cancel good projects in the face of liquidity constraints) and the problem of contagion (fi nancial institutions and markets are interconnected, hence become the carrier of the crisis). Th ese problems suggest that government has a role to play in mitigating the fall from a “bust.”

Th e primary role of fi nancial markets is to direct resources to their most productive

uses, and when this ability is compromised, productivity suff ers and could have long-term impact. Th e most direct real eff ects of disruptions in fi nancial markets are on capital (re)allocation during and aft er the crisis. Th e potential role of misal-location in delaying the recovery in Japan and Mexico in the 1990s underscores the need for authorities to intervene quickly and thoroughly in the banking system. Furthermore, Sweden in 1991-93 serves as an example of swift government inter-vention that held banks accountable while providing much needed government support.

In order to prevent the potential problem of moral hazard that comes with bank

bailouts, government should lend freely to only the solvent banks, but only do so against good collateral and high interest rate. If interest rates are high, banks are incentivized to repay the government as quickly as they can, withdraw from risky projects, and attract private capital. Eff ective government policies will revive the credit markets, restore confi dence in the fi nancial system, and enable the fi nancial sector to perform its task of allocating scarce resources to their most productive usage.

In this respect, fi nancial regulation can play an important role. Boom-bust cycles

often follow periods of financial liberalization reforms. This is consistent with the excessive risk taking view of booms, as fi nancial de-regulation reforms have

3

Financial crises – A review of the literature

typically facilitated risk taking. But returning to the previous regulatory frame-works is probably not advisable either, as the heavy intervention of the government in the allocation of credit was also distortionary. Regulatory reforms were suc-cessful in generating higher growth, and now the challenge is to retain these gains while recovering the stability that existed before the reforms.

In light of the recent crisis, studies suggest that fi nancial regulation should

become “macro-prudential,” that is, capital adequacy ratios should respond not only to the individual risk of securities held by banks, but by how they cor-relate with macroeconomic variables and, in particular, with how they behave during fi nancial crises. Macro-prudential regulation should also be cyclical, tighter in the boom and weaker in the downturn.

Even well designed fi nancial regulations are constrained by a boundary. Any fi -

nancial regulation has to specify which assets and institutions fall into its rules and which ones do not. Th is creates an incentive for funds to fl ow towards the unregulated institutions in the boom periods and back to the regulated ones in the crisis, potentially amplifying the cycle. Finding the balance thus in terms of excessive or inadequate fi nancial regulation is challenge for policy makers.

IntroductionFinancial markets are a crucial part of a well functioning economy. Th ey enable the real economy to function properly by channelling capital resources to its most produc-tive use. Th ey provide insurance against idiosyncratic shocks by allowing for pooling of risks. Furthermore, the combination of fi nancial intermediation and risk sharing allows for liquidity transformation, that is, individuals can place their savings in the bank knowing that they can use it whenever they please, while getting the high returns only available for long term, illiquid investments. Th e benefi ts of well-developed fi nan-cial markets are not just theoretical. King and Levine (1993) and Rajan and Zingales (1999) provide empirical evidence that countries with highly developed fi nancial mar-kets grow faster and that fi nancial development is particularly helpful in enhancing the growth of industries that, for technological reasons, rely heavily on external funds to enable their investment. In short, fi nancial markets are important to the well-being of the real economy, especially businesses and enterprises.

Th e past few decades leading up to the crisis were characterized as a period of relative macroeconomic stability, especially in the United States and major European countries. What became known as the “Great Moderation” – the combination of credible mon-etary policy with rapid development of the fi nancial system – seemed to have delivered a persistent drop in the volatility of infl ation and output (Bernanke, 2004). While crises were a possibility, the key to managing their macroeconomic impact appeared to be well understood, given the successful recovery engineered by the Federal Reserve


4

aft er the dot-com bubble (Blinder and Reis, 2005). However, the events in the last two years (2008-2010) have made it clear that the outlook was too optimistic.

While a major financial crisis is a novelty for policy makers in the US and major European Countries, other parts of the world have been regularly hit by fi nancial crises (see chapters 2 and 3). For example in Chile, Finland, and the Republic of Korea, output collapsed by 14 per cent (1982-1983), 10 per cent (1991 and 1992) and 7 per cent (1998) respectively – losses that proved fairly persistent (Cerra and Saxena, 2005; IMF 2009).1 Th e eff ects of crises on growth are oft en persistent. For example, following the banking crisis in 1990, the Sweden went into a major recession, with GDP falling for three consecutive years, a total of –5.1 per cent in 1991–3, and private investment plummeting by 35 per cent during the same period. Furthermore, unemployment rate quadrupled from 3 per cent in 1990 to around 12 per cent in 1993.

As a consequence, there is a robust modern literature on fi nancial crises that policy makers in newly aff ected countries can potentially draw from. Th e existing literature has yet to fully account for the large size of the crises and the long run eff ects, but still there are a number of fi ndings that may help guide policy in the context of the current crisis. Th is chapter is an attempt to draw the main lessons from this literature, with a special focus for the macro-economic impact of fi nancial crises.

A. Defi ning and identifying fi nancial crises

Th e downturn that started in 2007 and was amplifi ed by the bankruptcy of Lehman Brothers in September 2008. Th e current downturn is distinct from a normal down-turn for two notably reasons. First, it was a crisis, i.e. it was sudden, unexpected and with particularly large consequences in so far as key economic variables are concerned. Second, it was fi nancial in nature meaning that it was triggered or amplifi ed by a dis-ruption in the fi nancial sector, the key event being the bankruptcy of Lehman Brothers (see Eichengreen, 2008 for how real variables in the current crisis tracked the Great Depression and Zingales, 2008 on the Lehman Brothers collapse).2

Th e literature has used diff erent criteria to identify crisis episodes, many of which fi t directly or indirectly in parts of the defi nition proposed above. A full and uncontro-versial identifi cation is diffi cult, since it involves a counter-factual exercise: what would have happened in each particular episode if the fi nancial sector had remained intact throughout? Instead, literature has identifi ed episodes in which there are signs of dis-ruption in the fi nancial system, in fi nancial variables, in macroeconomic variables or in some combination of these without a stronger causal claim. For example, a direct sign of a large-scale disruption in fi nancial markets is the presence of widespread bank runs,

1. Th e crisis episode of the Republic of Korea is examined in more detail in Chapter 3.2. Chapter 4 will demonstrate that there are also a number of other intrinsic factors that lie behind the

onset of the current fi nancial crisis.

5

bank failures or bank insolvencies. While it is easy to determine whether there was a bank run or a bank failure, insolvencies are much harder to spot. For this reason, the identifi cation of banking crises oft en relies on the assessments of specialists (Caprio and Klingebiel, 1999; Laeven and Valencia 2008).

Another way to detect disruptions in fi nancial markets is to look at the behaviour of fi nancial fl ows and stocks (see also Chapter 4). An important example of this strategy is Calvo (1991), who looks for sudden stops in the infl ow of foreign capital. Mendoza and Terrones (2008) focus instead on credit booms, defi ned as large departures of credit to the private sector from its long-term trend. As it turns out, the peak of these booms oft en coincides with fi nancial crises, especially when they happen in less devel-oped countries.

A third approach is to look for loss in the value of important classes of assets such as government debt (Reinhart and Rogoff , 2010), stocks and housing (Bordo and Jeanne, 2002). Th ese are assets that represent an important part of the balance sheet of house-holds and fi rms, so that a drop in their value may lead to an interruption in the fl ow of fi nance as lenders fear for the value of the collateral that the borrowers have to off er.

Exchange rate crises can trigger or amplify a fi nancial crisis if fi nancial institutions have liabilities in foreign currency (Diaz-Alejandro 1984; Calvo and Talvi 2008).3 However, not all exchange rate crises turn into fi nancial crisis. For instance, while the collapse of the European Exchange Rate Mechanism in the early 90’s was something policy makers at the time did not desire, in most cases there was no spill over to the broader fi nancial system. In this respect, Kaminsky and Reinhart (2000) propose that it is useful to focus on episodes in which an exchange rate and a banking crisis take place simultaneously. Th e combination of criteria fi lters out episodes such as the ERM crisis as well as epi-sodes in which banking crises did not have any real eff ect.

Lastly, Kehoe and Prescott (2007) defi ne episodes that they call “Great Depressions of the 20th Century”. Th ese are occasions in which a country has suff ered a precipitous and persistent output drop. Th eir defi nition lacks any reference to disruptions in the fi nancial sector but, as it turns out, there is a substantial degree of overlap with fi nancial crises identifi ed in other studies.

In spite of the diff erent defi nitions, there is a striking amount of agreement on the rel-evant events. Apart from the Great Depression, most studies include observations from the Latin American sovereign debt crisis in the early 80s, the Scandinavian collapse in the early 90s, Japan throughout the 90s, the Asian crisis in the late 90s and Argentina in 1998-2001 (see Chapter 2). Th is coincidence strengthens the presumption that fi nancial crisis represent a reasonably well-defi ned economic problem, which is amenable to data col-lection and systematic research. Th is is the subject of the remaining sections.

3. Exchange rate crises are defi ned with reference to price of an asset.



6

B. The boom before the bust

Before discussing what happens during a crisis, it is important to understand the period preceding a crisis. As the literature shows, this proves to be crucial in designing and im-plementing the right set of policies. Th e identifi cation of clear antecedent patterns can provide a warning signal to policy makers and suggest corrections to be taken in order to avoid the worse. Figure 1.1 illustrates the origins of the fi nancial and economic crisis of 2007-09, and while there are several factors that led to the crisis, the most notable pattern is a period of “boom” in economic and fi nancial activity that gave rise to stock-market and housing bubbles. Th is section explores this “boom” period in detail.

A stylized account of the “typical” boom can be reconstructed from the fi ndings of different papers in the literature. There are regulatory changes, which allow banks to lend more freely and take more risk (Kaminsky and Reinhart, 2000; Tornell and Westermann, 2002). What follows is an increase in the supply of credit by banks, as they lend more relative to their assets and to their capital (Mendoza and Terrones, 2008). To this increased supply of credit there is a matching increase in the demand as fi rms increase their leverage and the government borrows more heavily (Rogoff and Reinhart, 2010). At the aggregate level, these trends are apparent in an increase in domestic credit (Mendoza and Terrones, 2008) and in capital infl ows from abroad (Rogoff and Reinhart, 2010). Prices of key assets react as house and equity prices in-crease (Bordo and Jeanne, 2002; Rogoff and Reinhart 2010) and as the exchange rate appreciates (Tornell and Westermann, 2002). All the while there is a boom in eco-nomic activity, with an increase in output and investment. As the real economy starts to lose steam, these trends revert abruptly, and the boom turns into a crisis (Kaminsky and Reinhart, 2002; Mendoza and Terrones 2008). Figure 1.1 illustrates these fi ndings by examining the fi nancial and economic crisis of 2007-09.

Two main strands of the literature on crises attempt to account for the boom pre-ceding the bust. Th e fi rst view states that the boom-bust cycle is evidence of excessive investment and risk taking. In the second view, asset price boom increases liquidity and facilitates investment. In particular, the boom in asset prices may stem from self-fulfi lling expectations about their value, rendering the boom fragile. Examining both these strands of economic literature provides a more balanced view of boom periods. However, it should be noted that both strands of the literature point out that booms eventually lead to busts.

1. Overinvestment

Th e fi rst set of theories suggests that there is overinvestment and excessive risk taking in the boom phase. Excessive investment and risk taking may take place because of (i) distorted incentives put in place by bail-out guarantees and other forms of limited li-ability or (ii) externalities induced by limitations in the workings of credit markets; or

7

(iii) because of bounded rationality which sophisticated investors cannot or will not arbitrage away.

Distortive incentives: Agents may overinvest if government guarantees or limited

liability protect them from the worse consequences of their risk taking behaviour. Th e theoretical case is simple enough (see Ranciere et al. for an example of a model with that feature) and there is some compelling empirical evidence about the as-sociation between banking crises and the existence of deposit insurance or other forms of government guarantees to banks in distress.4 For example, Kane (1989) points to deposit insurance as a key driver of the Savings and Loans crisis in the 1980s and Demirguc-Kunt and Detragiache (2002) fi nd a correlation between the availability of such schemes and the probability of banking crises. Lastly Calomiris (2009) makes a forceful case, based on historical accounts, that banking crises in 19th century U.S. and England were associated with some form of deposit insur-ance or liquidity guarantee to fi nancial institutions.5

Credit market externalities: Even if banks and fi rms do not expect to be rescued

in case their investment projects turn bad, they may not realize that their actions

4. Demirguc-Kunt and Detragiache (2002) fi nd that deposit insurance schemes are, in most countries, run and/or funded by the government.

5. Calomiris (2009) distinguishes “banking crises”, situations in which banks go broke, from “banking panics”, situations in which there are bank runs but no bankruptcies. He also distinguishes these epi-sodes from broader fi nancial crises in which fi nancial variables other than bank’s balance sheets show signs of reversals.

Underestimated risksin financial markets;over investment

Loose monetarypolicy (2003-05)

Failures ofcorporategovernance

U.S. Householdssaving too littleand borrowingtoo much

Excess leverage infinancial institutions;opaque financialproducts

HousingBubble

Stock-marketBubble

Govt policiesencouraging homeownership; predatory lending

Stock market crash; housing market crash

Recession, 2007-09

Figure 1.1: Origins of the financial and economic crisis, 2007-09



8

have negative repercussions on others. Th is may occur if credit markets are incom-plete, so that fi rms depend on the current market value of their assets to fi nance new investment. If there is a crisis, large investment will lead to larger drops in asset prices and hence less new investment. In addition, because fi rms will not re-alize the impact of their decisions on asset prices, generating the possibility of an ineffi ciently large crash (Stiglitz, 1982; Geanakoplos and Polemarchakis, 1985; Lorenzoni, 2008).

Irrational exuberance: Investors may overinvest because they underestimate risks

or overestimate returns. Individuals can miscalculate because they base their eco-nomic decisions on current prices, with little regard to future evolutions. In such a setup, by keeping interest rates artifi cially low, overly lax monetary policy can lead to large, persistent but ultimately unsustainable booms in investment and con-sumption (Garrison, 2001). As Figure 1.2 shows, long-term interest rates decreased sharply in the last 25 years, and that the price to earnings ratio increased sharply

in the same period. Th e increase in P/E ratio signals the boom period that was marked by over-investment, when leverage among fi nancial institutions exploded (see fi gure 1.3). More generally, the behavioural fi nance literature has been suc-cessful in explaining a variety of asset price bubbles with appeal to bounded rational

0

2

4

6

8

10

12

14

16

18

20

0

5

10

15

20

25

30

35

40

45

50

1950 1960 1970 1980 1990 2000 2010

Pric

e-E

arni

ngs

ratio

Price-Eearnings ratio

Figure 1.2: Boom before the bust: PE ratio and long-term interest rates in the US

Source: Shiller, 2010.

Long-Term interest rates

Long

-Ter

m in

tere

st ra

tes

9

behaviour. One example emphasized by the literature is for funds to be directed to assets that appear to have performed particularly well in the recent past (see Shiller, iscussion of this strand of the literature), thus propagating booms and busts.

Th at some investors do not behave in a fully rational pattern is not surprising given the complexities involved in fi nancial investing. Th e question is why the more sophisti-cated investors do bet against these “naïve” investors, thus bringing the price of various assets in line to their fundamental value. One answer is that it is hard to bet against an asset by selling it short. Indeed, it is oft en hard to fi nd third parties willing to lend an asset, those who do charge a fee and demand protections for their risk (see d’Avollio, 2002 for a thorough account of the market for shorting stocks). Th e short sale of other, more complex assets is even harder to do, as it may involve the design of new, derivative securities (see Lewis 2010 for a journalist account of the diffi culties faced by contrar-ians who wanted to short CDO’s in the run up to the current crisis). Without short sales, when investors disagree about the value of a security, pessimists are unable to bet against it and the price of the asset refl ects the view of the optimists (Miller, 1977; Hong and Stein, 2003).


50

100

150

200

250

300

350

3/31

/198

7

6/30

/198

8

9/30

/198

9

12/3

1/19

90

3/31

/199

2

6/30

/199

3

9/30

/199

4

12/3

1/19

95

3/31

/199

7

6/30

/199

8

9/30

/199

9

12/3

1/20

00

3/31

/200

2

6/30

/200

3

9/30

/200

4

12/3

1/20

05

3/31

/200

7

6/30

/200

8

Government HouseholdsFinancial institutions Nonf inancial corporations

Figure 1.3: Ratio of debt to GDP among select advanced economies

Note: In percent, GDP-weighted, 1987=100. Source: IMF.


10

Speculation can become destabilizing if sophisticated investors hold on to what they perceive as overpriced assets in the expectation of future price increases (Scheinkman and Xiong, 2003). In fact, even if they perceive an unsustainable bubble, they may choose to “ride the bubble” and count on the probability that they will be able to exit before the crash, reaping the capital gains along the way (Abreu and Brunnermeier, 2003). Executive pay practices coincide with this, as increasingly a higher share of CEO salaries comes from stock gains, and bonus is dependent on quarterly returns (see fi gure 1.4). In Minsky (1986) fi nanciers are willing to pump credit into fi rms even as these have to run Ponzi schemes, acquiring new loans in order to pay for old ones. A “musical chairs” situation ensues, where everyone plays but in the end not all can fi nd a place. Eventually the system unravels, leading to a crisis.

2. Enhanced liquidity: enabler of rapid growth?

It is possible that the boom is not fuelled by overinvestment, but rather enhanced li-quidity which stems from rising asset prices. In particular, a fi rm or a household may need to have immediate access to funds because of, e.g. (i) an investment or acquisition opportunity, (ii) unexpected cost that has to be paid in order to ensure its continued existence or because in order for the business to continue it requires uninterrupted

Figure 1.4: CEO compensation based on Forbes annual survey

Note: Includes 500 largest firms in the US. Source: Forbes, 2009.

0

4

8

12

16

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

$ m

illio

ns

Other compensation Stock gains Salary and bonus

11

investment. But importantly, liquidity is a state in which this demand for immediacy is satisfi ed without much cost. For example, frictionless models feature funding liquidity, since households can borrow as much as they like at a given rate of interest. Th ere is also liquidity if the fi rm or the household has assets in its books that can be marketed immediately at little loss (“market liquidity” or “outside liquidity”, see Brunnermeier and Pedersen, 2009).

A boom in asset prices increases liquidity. In particular, it increases funding liquidity by increasing the value of assets that fi rms hold as collateral. Furthermore if the booming assets are relatively easy to market, asset price booms increase the availability of market liquid assets. As such, with higher asset prices there are more available means to store value for a given capital stock and, hence, higher market liquidity.

It is possible that assets are valued above their intrinsic value without sustainability being an issue. In other words, rational bubbles can emerge. Agents will hold a seem-ingly overpriced asset because they rationally expect other agents to do so in the future (Tirole, 1985). Th e purest example of such an asset is fi at money (see Samuelson 1958), but other assets could acquire similar properties. Rational bubbles can be welfare en-hancing in that they facilitate intergenerational transfers (Samuelson, 1958; Tirole, 1985), international asset fl ows (Ventura, 2004; Hellwig and Lorenzoni, 2008) and the direction of real resources to highly productive but credit constrained entrepreneurs (Caballero and Krishnamurthy, 2005; Farhi and Tirole, 2008; Martin and Ventura, 2010).

Of course, to compete, bubbles have to grow at rate higher than prevailing interest rates but if the value of a bubble rises too fast it may eventually become much larger than what agents are willing to hold and a collapse ensues (Blanchard and Watson, 1982). Importantly, the growth of bubbles may itself lead to higher interest rates, as savings are increasingly channelled to feeding its growth as opposed to the growth in alternative assets, potentially fuelling its own demise. In fact, they might deviate resources away from other more productive investments, such as research and development (Saint Paul 1992; Grossman and Yanagawa, 1993; Olivier, 2001). Th ey may also compete with assets commonly used as collateral, thus reducing their value and constraining further investment (Farhi and Tirole, 2010) and they can fuel excessive risk taking (Caballero and Krishnamurthy, 2005). In sum, while there is a theoretical case that bubbles en-hance liquidity and facilitate investment, they can also be distortionary by competing with other, more socially desirable assets. As such bubbles may enhance liquidity and fuel a boom, but they can equally well lead to overinvestment and misallocation of resources.

C. The bust in the fi nancial sector

No matter what strand of literature one subscribes to, booms are followed by busts. Financial crises occur when the fi nancial market is disrupted and, as a consequence,



12

become unable to function appropriately. Th is section describes two important ways in which fi nancial markets can become dysfunctional. First, liquidity problems may force good projects to be discontinued. Second, there is the problem of contagion given that fi nancial markets bring together other, e.g. trade, key components of economic growth. Both of these problems provide rationales for government action. Th e last section closes with some considerations about the care that policy makers have to take when doing these interventions.

1. Distinguishing insolvent from illiquid institutions

Th e diff erence between solvency and liquidity problems is best understood in the con-text of the Diamond and Dybvig’s (1983) model of bank runs. In this model, banks direct funds to the construction and operation of long term, illiquid projects. A bank is solvent if the projects are worth more if they are allowed to mature than if they are liquidated. Th e Diamond-Dybvig (1983) framework allows for the liquidation of sol-vent banks if there is a run. Depositors may decide to withdraw from the bank because they lose confi dence on its ability to honour the deposits. If the liquidation value of the projects is less than their continuation value, the bank may fi nd itself short on li-quidity and unable to honour the deposits. Th e run is self-fulfi lling and the liquidation ineffi cient.

While Diamond and Dybvig’s (1983) paper centred specifi cally on bank runs, the logic of the argument applies to alternative contexts. Th e potential for self-fulfi lling creditor runs is possible whenever an entity has liabilities which are of shorter maturity than its assets and its assets cannot be liquidated without paying a fi re-sale cost. For example, some interpretations of the Asian crisis view it as a liquidity crisis (Chang and Velasco, 1998). As a second example, the widespread use of overnight loans by fi nancial institu-tions meant that Diamond-Dybvig logic could be at play in the current crisis episode, even if actual run of depositors to the banks were not an important part of the picture (Brunnermeier, 2009).

Th e bank-run view of banking crises is not without controversy. For example, Gorton (1988) fi nds that banking panics in the 1863-1913 period took place when leading indicators of recessions were strong. In his interpretation, depositors were rationally anticipating that the downturn would bring insolvencies and were withdrawing their deposits to make sure they did not get lost. In contrast, the Diamond-Dybvig model allows for bank runs as one of multiple equilibria, without any reference to funda-mentals. He and Xiong (2009) provide a theoretical reconciliation between Gorton’s (1988) results and the Diamond-Dybvig worldview. In their paper, heterogeneity be-tween creditors imply that there is a single equilibrium in each point in time, with runs only occurring if fundamental values are low enough. Still, the Diamond-Dybvig logic applies and the model allows for runs in fundamentally healthy banks. What the model and Gorton’s results point to is that it may be, in practice, hard to distinguish liquidity

13

problems from solvency problems, as liquidity may become scarce exactly when there are reasons for doubts about solvency.

2. Contagion

Th e focus on the solvency or liquidity of individual institutions may be too narrow. Financial crises are oft en systemic events, spreading over multiple markets and regions. “Contagion”, the spread of fi nancial diffi culties from one group of debtors to another is a distinct possibility. Th ere are, broadly speaking, four sources of contagion. Th e fi rst is contagion through real linkages. If Th ailand suff ers a crisis, South Korea is bound to experience a drop in its exports, with negative consequences on its domestic economy. Th e second is through information problems. A collapse in one part of the system may be a signal to investors that other parts are also in danger, leading to a decline in con-fi dence and ensuing consequences. Th ird, contagion can occur if fi nancial links are relatively sparse, so that the disappearance of key nodes may lead to a collapse in the network. Th e last is through the wealth of decision makers. A bank whose capital is depleted by a crisis may become reluctant to give new loans and sell part of its assets, both of which can result in reductions in the capital of all banks.

Real linkages: modern economies are characterized by a multitude of buyer and

supplier linkages. If the collapse of a fi nancial institution aff ects an important node in this supply chain, the impact could spread. Trade links between countries can have a similar eff ect, as Glick and Rose (1999) argued was the case in the Asian crisis. If real linkages are the only source of problems, then policy makers may be justifi ed in ignoring contagion. Aft er all, if a certain business is a bad proposition, then the same is true for those businesses whose viability depend on it. On the other hand, if businesses linked by trade are also linked by fi nancial relationships that cannot be easily substituted, then the disruption of these trade links may lead to fi nancial problems that will lead otherwise healthy businesses to go under, pos-sibly aff ecting other businesses still through the credit chain (Kiyotaki and Moore, 1997b).

Communication or information problems: For example, in Calvo (1999), a few in-

formed investors pick where to invest and a large number of less informed investors free ride on their information by replicating their choices. If informed investors face some liquidity problem and for that reason disinvest in a certain market, the less informed investors may erroneously conclude that this disinvestment refl ects some bad information on their part and that they are better off following suit. Furthermore, once they see the drop in asset prices in one market, uninformed investors may decide to withdraw from other markets as well if they think there



14

are linkages between them (King and Wadhani, 1990; Kodres and Pritsaker, 2002) or if they do not discriminate between information about various risky assets in order to economize on information acquisition costs (Mondria and Quintana-Domeque, 2009).

Key fi nancial links: Firms or countries may not be linked to each other by trade re-

lationships, but by common lenders. For example, when Russia collapsed in 1998, Latin American countries were aff ected even though the trade links were minimal. Th e Russian collapse led to the LTCM default, which in turn put the U.S. credit market into distress, aff ecting Latin American countries.6 Apart from links through common lenders, fi nancial institutions are themselves linked to each other, so that distress in one institution may lead to distress in others. If each institution only has a limited number of counterparties, a domino eff ect can ensue, where the failure of one institution depletes the wealth of its counterparties, and so on (Allen and Gale, 2000). In such a system, individual fi nancial institutions do not realize that, by pre-serving their links to other players, they provide a valuable service to the system as a whole – thus ignoring the important externality (see Zawadowski, 2010).

Wealth of decision makers: Crises can spread and propagate through its eff ect on

asset prices and, consequently, on the wealth of fi rms and banks. Th is channel was emphasized in Kiyotaki and Moore (1997a) and applied specifi cally to fi nancial crises in emerging markets by Mendoza (2010). A negative shock that reduces the value of some fi xed asset such as real estate may spread over the economy as fi rms use real estate as collateral to secure loans. As the value of collateral drops, fi rms are forced to reduce their investment in real estate, leading to a further drop in the value of their collateral. Th ere is a “pecuniary externality” (Stiglitz, 1982; Lorenzoni, 2007), as fi rms do not realize that, by investing when times are good they are increasing the stock of capital that will have to be liquidated when times are bad.

Many of these interact with one another during a crisis, e.g. wealth and information. In Brunnermeier and Pedersen (2009) for example, when sophisticated investors borrow from brokers they need to come up with some capital of their own (the haircut). If brokers perceive an increase in volatility, they will demand higher haircut. Th e require-ment leads investors to reduce their positions, bringing prices down and volatility up. Th e increased volatility generates an increase in the haircut and further price drops. Th e spiral occurs because lenders cannot diff erentiate liquidity shocks from changes in fun-damental values and withdraw their funds, further increasing the liquidity problem.

6. Some studies have found evidence for this kind of channel at the country level (Kaminsky and Reinhart, 2000), and among hedge funds (Boyson et al., 2008), although the authors themselves admit that the evidence is not conclusive.

15

D. Supply side effects: Misallocation in capital and labour marketsTh e most direct real eff ects of disruptions in fi nancial markets or on the supply side, most specifi cally are on capital (re)allocation. Th is is because one of the main functions of fi nancial markets is exactly to direct resources to their most productive uses. If this ability is compromised, then productivity suff ers. Th is is especially problematic since it can have persistent, long-term eff ects, with large welfare consequences.

Financial crises also have an impact on labour markets. In part this is an expression of the allocation problems in capital markets, but also in part due to the fact fi nancial crises are simply larger versions of the “usual downturns”. In this regard, there is a sub-stantial body of theory discussing effi ciency and frictions in labour markets and how fi nancial markets can both amplify the usual frictions and create new ones.

1. Capital misallocation

Th e economic environment changes aft er a crisis, as credit availability, foreign capital infl ows and investment become persistently depressed relative to their pre-crisis levels (Calvo, 2006). Th e now scarcer capital needs to be reallocated in order to refl ect the new environment. Accordingly, Dell’Ariccia et al. (2009) and Kroeszner et al. (2007) fi nd that during banking crises the industries that require large up-front fi xed invest-ments are more severely aff ected; and Schwartzman (2010) fi nds a similar eff ect during emerging market crises for industries whose production processes require them to hold inventories for long periods of time. Recent evidence from the EU and the US shows that the corporate spreads have sky-rocketed during the crisis and they remain elevated (see fi gure 1.5). Th is underscores the problem of high uncertainty and risk in the cor-porate sector, hence probability of default, which makes direct borrowing costly for fi rms.

It is not clear, however, that the reallocation is necessarily in the right direction or in the right order of magnitude. For example, industries that require little up-front investment will do better over and above what is justifi ed by the shorter duration of their capital because they are able to self-fi nance. More broadly, a breakdown in fi nan-cial markets is costly exactly because fi nancial markets are responsible for distributing resources effi ciently across fi rms and sectors. Capital may end up being directed to inferior uses and high productivity fi rms left underfi nanced. Moreover, fi nancial crises can aff ect the high productivity fi rms most heavily, since these are the ones that re-ceived most credit before the crisis and have least debt capacity left (see Kiyotaki 1998; Rampini and Visnawathan, forthcoming).

In terms of empirical evidence for misallocation, Ohanian (2001) investigates cross-sec-toral reallocation during the Great Depression and Benjamin and Meza (2009) during the Korean crisis. Both fi nd that there was a relatively larger drop in manufacturing and



16

mining, which have high productivity, relative to services and agriculture, which have low productivity (for more detail see Chapter 3). Moreover, misallocation can persist in the medium term, especially if problems in the fi nancial system are not properly ad-dressed. Caballero et al. (2008) suggest that, aft er the crisis, Japanese banks had incen-tives to keep on lending to otherwise insolvent fi rms in order to avoid recognizing losses and to enjoy government support. Th ey fi nd that, because of this “zombie lending”, non-favoured fi rms did not improve their productivity or create jobs as quickly as they would otherwise. Bergoeing et al. (2009) make a similar point when comparing the performance of the Chilean and the Mexican economies aft er the 1980’s crisis. Chile recovered faster, they suggest, because it was quicker to return banks nationalized during the crisis to the private sector. Also, Chile reformed bankruptcy procedures to make default easier. Mexico avoided these policies, reaping benefi ts in the short run, but compromising the recovery in the medium term.

While the sources of productivity loss during crises and aft erwards are still not fully understood (Ohanian, 2001), there are enough grounds to believe that misallocation of resources is a major concern, especially in so far as the recovery is concerned. Th e potential role of misallocation in delaying the recovery in Japan and Mexico underline the need for authorities to intervene quickly and thoroughly in the banking system.

US

EU

0

200

400

600

800

1000

1200

1400

1600

1800

3658

6

6/26

/200

0

12/1

5/20

00

3720

1

3699

3

5/30

/200

2

11/1

9/20

02

5/15

/200

3

3772

2

4/27

/200

4

10/1

8/20

04

3869

0

9/30

/200

5

3/27

/200

6

9/18

/200

6

3941

9

3915

0

2/26

/200

8

8/19

/200

8

4005

8

Bas

is p

oint

s

Figure 1.5: Corporate spreads in the US and EU (basis points)

Source: IMF.

17

2. Labour reallocation

Even in the course of normal business cycles, standard models of labour markets cannot fully account for the high variability of unemployment. In previous episodes of crisis, labour market adjustment has oft en occurred via employment, hours worked and/or wages. In some cases, wages may not have fallen because the government intervenes in downturns to protect the jobs and the income of employed workers (McGrattan and Prescott, 2007; Mertens and Ravn, 2008). Even during the Great Depression, the Hoover administration pursued policies to keep nominal wages from dropping (Ohanian, 2009); and in the Japanese crisis in the 90’s, with a mandatory decrease in working hours (Hayashi and Prescott, 2000).

Interestingly, Bernanke and Carey (1996) fi nd evidence that nominal wage rigidity played an important role during the Great Depression, with the highest real wage and the lowest output prevailing in countries that suff ered most from negative demand shocks. As additional evidence for a role for nominal wage rigidity, Fallon and Lucas (2002) fi nd that among the emerging market crises in the 90’s, real wages dropped more in countries where exchange rate depreciation was highest.

Some labour market problems are specifi c to fi nancial crises. One reason why fi rms may be reluctant to hire new workers during a crisis is that part of the wage bill has to be paid in advance because of a preference for liquidity by the workers or of time lags in the production process (Neumeyer and Perri, 2004; Schwartzman, 2010). Th us, when the cost of fi nance goes up, fi rms become reluctant to hire. Hiring is also a credit intensive activity so that job creation suff ers if the cost of fi nancing goes up (Weismer and Weil, 2004). Finally, the allocation problems in capital markets may spill over into labour markets, with the most productive fi rms not being able to hire the best workers (Caballero, 2007).

Th e available empirical evidence does suggest that labour market outcomes during fi -nancial crises are worse than in regular downturns, even aft er controlling for the size of the recession. A recent World Economic Outlook report (IMF, 2010) fi nds that both during fi nancial crises and in the recovery phase, unemployment is lower than what would usually be warranted by the observed drop in output. Furthermore, Caballero (2007) shows that, while in the normal American business cycle old jobs are kept in place, in fi nancial crises there was a large degree of job destruction without a commen-surate increase in job creation. In the latest crisis, Elsby et al. (2010) confi rmed that this same pattern also holds for the United States. Lastly, Schwartzman (2010) fi nds that during emerging market crises the industries with high labour share have under-performed compared to their usual cyclical behaviour, a pattern that also points to a particularly large increase in labour costs.

Th e large impact of fi nancial crises on the labour market can also generate structural adjustments in the labour market, e.g. increased long-term unemployment (Elsby et al., 2010). It is diffi cult to reintegrate such workers into the labour force, as skills erode



18

and labour market detachment rise. Th is gives rise to hysteresis, i.e., the tendency of large cyclical unemployment outcomes to translate into a permanent increase in the natural rate of unemployment (see Ball, 2009 for a recent survey of the literature). Th ere are also issues with respect to role of deregulation and job creation. For example, in Korea, “fl exibilization” of the labour market to spur job growth lead to increased labour market duality (see Chapter 3 and Chapter 6).

E. Demand side channels and macroeconomic policy

On top of the direct supply eff ects, fi nancial crises can feed back into the economy through the demand side. In part, fi nancial crises cause drops in aggregate demand and important problems emerge if the interest rate is constrained to be equal to zero. Th is is more than a theoretical possibility, as zero interest-rate was a feature of two prominent past episodes, the Great Depression and the Japanese crisis, and is a feature of the cur-rent episode.

Not only there are changes in aggregate demand, but also there are shift s in the com-position of demand, generating large shift s in relative prices. Th e most prominent ex-ample is the exchange rate, which can have a negative impact if fi rms have a mismatch between the denomination of their debt and that of their cash fl ow. However, there are also shift s in the relative price of capital, which can be problematic if capital is used as collateral.

1. Aggregate demand and the zero lower bound on interest rates

Financial crises can induce a drop in aggregate demand. Increased uncertainty induces precautionary savings and a portfolio shift away from risky investment towards safe monetary assets (Keynes, 1936). Also, because of the crisis, borrowers face a higher interest rate, so they consume less and invest less (Bernanke, 1983). Finally, corporate debt becomes less liquid relative to cash, increasing the cost of investing (Del Negro et al., 2010). In all of these cases, a lower interest rate is needed for the goods market to be in equilibrium. In Woodford’s (2003) terms, there is a reduction in the natural rate of interest.

If the drop in this natural rate of interest rate is so low that it becomes negative, it can throw the economy in a liquidity trap: the nominal interest rate is constrained to be zero and cannot go any lower because the return on bonds cannot be lower than the return on cash. In such circumstances, a perverse feedback loop can occur: because of the bound on the interest rate, it is higher than what is consistent with equilibrium in the goods market, leading to defl ation. But, given the nominal rate, defl ation raises the real interest rate even more, leading to an even greater disequilibrium.

19

In a liquidity trap there is no opportunity cost of holding cash, so agents satiate their demand for liquidity. Without the liquidity channel, additional cash injections in the economy (quantitative easing) can still have a real eff ect because at some point agents will not want to hold any more cash for fear of being under-diversifi ed. However, this is a much weaker trade-off than the traditional liquidity margin, and the quantities involved have to be an order of magnitude larger. Alternatively, central banks can inter-vene directly in the market for long maturity bonds, as they recognize that these are the ones that aff ect real decisions. Again, the central bank has to rely on agent’s desire for diversifi cation to change the relative returns on short and long-term bonds and, simi-larly, large quantities need to be traded given that bonds of diff erent maturities are close substitutes (see Bernanke and Reinhart, 2002, for a discussion of these policies). As a result, the existing evidence for the eff ectiveness of this kind of policy is mixed, with Bernanke et al. (2004) fi nding that in the U.S. the term structure reacts to changes in the composition of treasury bills outstanding but it does not happen in Japan. On the other hand, Bernanke et al. (2004) do fi nd that the quantitative easing policy in Japan was successful in both changing asset prices and increasing infl ation expectations.

Th e policies just described increase the short-term interest rate consistent with equilib-rium in the goods markets, bringing it closer to the zero lower bound. A second set of proposals focuses on decreasing all nominal interest rates by increasing expected infl a-tion. Th is can be achieved if agents believe that the Central Bank will keep interest rates low even aft er infl ation has picked up. Simple Central Bank announcements may help, as they seem to have an impact on market prices (Bernanke et al, 2004). However, there is a credibility problem, as Central Banks usually have a preference for low infl ation. One alternative is to engineer monetary expansions or an expansion of government debt so large that the government cannot plausibly take them back without recourse to infl ation (Auerbach and Obstfeld, 2005; Eggertson and Woodford, 2002). But just with the policies exploiting the portfolio channel, large interventions are needed. Svensson (2003) notes that the commitment problem would not arise if Central Banks do not commit to an infl ation target, but to a “price level target”. Th is means that if infl ation is below the long-run target in a given year, then in the next year the Central Bank should seek infl ation above the long-run target. With sustained defl ation, Central Banks need even higher infl ation rates in order to reach their proposed target. If the policy is cred-ible, agents would come to expect these, and infl ation would rise accordingly. Svensson (2003) suggests that a “Fool-Proof Way” of implementing such a price level target is through targets for nominal exchange rates.

A fi nal alternative is to use fi scal stimulus. Christiano et al. (2009) use a quantitative model to argue that fi scal policy is likely to be most potent in liquidity trap situations. If nominal interest rates are equal to zero, fi scal policy can increase infl ation without increasing the nominal rate, thus decreasing the real interest rate, in a reversal of the textbook crowing-out eff ect. Th is gives some ground for optimism that fi scal expansion multipliers under liquidity traps are above the usual estimates, which are typically low (see: Barro, 2009; Ramey, 2009; and Mountford and Uhlig, 2009). Still, care is needed. Giavazzi et al. (2000) fi nd that fi scal multipliers are most likely to be negative if the



20

fi scal intervention is large, presumably because this involves higher and more immediate tax increases (for a more detailed examination of fi scal multipliers see Chapter 6).

2. Shifts in the composition of demand and relative price changes

Financial crises aff ect not only aggregate demand, but also its composition. While such compositional shift s may seem to be relatively less relevant, they can have big aggregate impacts on aggregate output if they change the relative prices in a way that makes it hard for fi rms to pay their debts or to issue new one. Two relative prices are of particular importance: the relative price of foreign goods relative to domestic goods (also known as the real exchange rate) and the price of capital goods relative to consumer goods.

Consider fi rst shift s in the relative price of foreign goods (Aghion et al., 2002). Suppose fi rms and/or banks are constrained in their ability to raise funds so that they rely partly on cash fl ow for investment. Furthermore, suppose they borrow heavily from foreigners, who only accept liabilities denominated in foreign currency. If there is an exchange rate devaluation, fi rms will see their liabilities rise but will not see a commensurate increase in their cash infl ows. Given a constraint in their ability to borrow, they will be forced to reduce their investment and output. In this context, increasing the interest rates in order to attract foreign investment and to defend the currency may turn out to be the best course of action. A robust, country level, observation that lends support to this channel is that fi nancial crises are more probable and/or more severe in countries where dollarized liabilities are more prevalent (Calvo et al., 2004; Bordo, 2006; Eichengreen et al., 2005). At the fi rm level, however, the evidence is more mixed. While Aguiar (2005) and Desai et al (2008) do fi nd an important fi rm level eff ect, the quantitative relevance of the mechanism is questionable. Gilchrist and Sim (2007) estimate and simulate a structural model of fi rm behaviour for South Korea and fi nd solid evidence for a currency mismatch eff ect, but fi nd that its impact on aggregate investment was relatively minor. Rather, they argue, the largest drop in investment occurred because of the combination of low cash fl ows with high interest rates and debt overhang, irrespec-tive of the actual denomination of the debt.

Th e quantitative role of balance sheet eff ect matters because there are trade-off s. For example, if fi rms have short-term capital needs, the high interest rate needed to de-fend the currency may end up doing more harm than good by aff ecting fi rm’s ability to raise working capital (Christiano et al, 2004) Also, higher interest rates may amplify the drop in the relative price of capital, including real estate. Th is can exacerbate the crisis, since fi rms use capital as collateral. Mendoza (2010) and Gertler et al. (2007) construct quantitative models to explore the relevance of fl uctuations in the relative price of capital and fi nd large eff ects. In particular, Gertler et al. (2007) argue that an uncompromising defence of the currency may make things worse. In their simulation of the Korean economy, they fi nd that on net it was best to let the currency depreciate. However, the gain is quantitatively small and it could be reverted for diff erent param-eter values. Th e best decision has to be made on a case-by-case basis, with countries

21

where the extent of foreign currency indexation of liabilities is extreme likely to be benefi t most from a strict defence of the currency.

F. Policy considerations and rationale for the next chapters Th is chapter so far has provided a comprehensive look at the literature on fi nancial crises. More specifi cally, it has looked the period preceding the crisis (“boom”), crisis period itself, and the dislocations prevalent during the recovery phase, mainly supply-side eff ects on capital and labour. Monetary policy has an important role to play in avoiding the crisis, and/or mitigating the fall during the crisis. Th ere is a lot that policy makers can draw from the macroeconomic literature. Th is section looks at two periods when economic policy is most important: 1) “boom” period that can be tamed by a right set of policies, and 2) recovery phase during which policy measures can address dislocations in the capital and labour markets.

1. Role of policy during “boom” period

Monetary policy clearly plays a role in facilitating excessive growth in asset prices. Indeed, easy monetary policy has been associated with higher asset prices (Taylor, 2009) and over-leverage in the fi nancial sector (Adrian and Shin, 2010) – both hall-marks of the boom phase. Conversely, monetary policy can, if tightened, reverse these trends, especially where there is a tendency to overinvest (Cechetti et al, 2000).

Th e problem is that monetary policy may be too blunt an instrument and diffi cult to manage. As evidence in the literature, asset price booms generate risks and distor-tions but also enhance liquidity and facilitate investment. Bernanke and Gertler (2000, 2001) address the trade-off . While tight monetary helps keep prices from rising un-sustainably, these reductions aff ect the value of assets used as collateral, potentially leading to an unnecessary recession. Furthermore, policy makers may not have enough information to discern the sources of the asset price movements and how dangerous they are. Bubbles are hard to identify empirically (Gurkaynak, 2008) and it is oft en the case that booms that do not end in a fi nancial crisis (Tornell and Westermann, 2002). Finally, even if central bankers are able to perfectly understand how sustainable and dangerous a boom is, the optimal monetary policy is likely to be highly non-linear in the state of the economy, as the trade-off between liquidity and risk of crash is likely to change with economic conditions (Bordo and Jeanne, 2002). Lastly, the literature on rational bubbles suggests that interest rate hikes may bring about almost discontinuous drops in asset prices – facilitating the onset of the crisis.

In this respect, fi nancial regulation can play an important role. Boom-bust cycles oft en follow periods of financial liberalization reforms (Kaminsky and Reinhart, 2000; Tornell and Westermann, 2002), implying that fi nancial regulation does matter. Th is



22

is consistent with the excessive risk taking view of booms, as fi nancial de-regulation reforms have typically facilitated risk taking (See Barandiarán and Hernandez, 1999; and Drees and Pazarbaşioğlu, 1998 for detailed descriptions of the fi nancial deregula-tion reforms in Chile and the Nordic countries and their aft ermath). Returning to the previous regulatory frameworks is probably not advisable either, as the heavy interven-tion of the government in the allocation of credit was likely very distortionary. Reforms were successful in generating higher growth (Ranciere et al., 2003) and the challenge is to retain these gains while recovering the stability existent before the reforms.

Brunnermeier et al. (2009) investigate some options in light of the recent events. Th ey suggest that fi nancial regulation should become “macro-prudential”. By this the au-thors mean that capital adequacy ratios should respond not only to the individual risk of securities held by banks, but by how they correlate with macroeconomic variables and, in particular, with how they behave during fi nancial crises. As an example, highly rated, e.g. AAA securities, are generally very safe or riskless. However, they fail exactly when there is a crisis and liquidity is scarcest. From a macro-prudential perspective, this may make them more risky than lower rated but diversifi able BBB securities. Macro-prudential regulation should also be cyclical, tighter in the boom and weaker in the downturn. Th is is the opposite of what the Basel II rules do, as they require banks to increase their capital ratios exactly when a crisis leads to the downgrade of their securities.

Even well designed fi nancial regulations are constrained by a boundary problem (see the appendix by Charles Goodhart in Brunnermeier et al., 2009). Any fi nancial regula-tion has to specify which assets and institutions fall into its rules and which ones do not. Th is creates an incentive for funds to fl ow towards the unregulated institutions in the boom periods and back to the regulated ones in the crisis, potentially amplifying the cycle. Finding the balance thus in terms of excessive or inadequate fi nancial regula-tion is challenge for policy makers.

2. Addressing dysfunction in the fi nancial markets

Th e problems of lack of liquidity and of contagion suggest that governments have a role to play in ensuring that fi nancial institutions do not fail. Th is generates an immediate contradiction: bailout guarantees provide incentive for bankers to take excessive risk and for depositors to not monitor banks appropriately. Before any intervention takes place banks will invest in projects with excessively high risk of failure (see above) and aft er it does, the banks under intervention may perceive a “soft budget constraint” (see Kornai et al. 2003), since the government has an increased stake in their survival, and will start taking even more risky positions.

In order to avoid banks taking advantage of government help, it is advisable to avoid helping insolvent banks and to make bailouts costly for the ones that do receive it. Th ese two insights underpin Bagehot’s (1894) classic prescription: First, Central Banks

23

should lend freely during crises, but not to insolvent banks. Second, they should lend only against good collateral and at a punitively high interest rate. If interest rates are high, banks will want to make sure they repay the government loan as quickly as they can, having an incentive to withdraw from risky projects and attract private capital.

Distinguishing which banks are illiquid and which are insolvent is not a simple matter. As discussed above, banking crises occur when fundamentals suggest that banks might be insolvent (Gorton, 1988) but theory implies that liquidity problems may induce in-effi cient liquidations (He and Xiong, 2009). One possibility, suggested by Demirguc-Kunt; and Serven, 2010, is that the government could mobilize the knowledge of the private sector by insisting that any loans to banks in distress must be in part privately fi -nanced, and that its equity in the recapitalized bank be senior to that of private agents. Th e government should also try to make sure it has as much information about the position of fi nancial institutions before the crisis (Demirguc-Kunt and Serven, 2010). Th e information available to the regulators is improved if in normal times transactions are transparent and that regulators focus on detecting problems early on. Ex-post, it helps if the government is capable of intervening on short notice in the event of a crisis (Kane and Klingebiel, 2000).

While some level or form of government intervention therefore seems to be a foregone conclusion, the eff ectiveness of government intervention is in itself a matter of dispute. One form of intervention that has received some attention is the use of blanket guar-antees. Laeven and Valencia (2008) fi nd that blanket guarantees are successful in re-ducing the pressure on banks by depositors, although Honohan and Klingebiel (2003) fi nd that are not correlated with smaller losses in aggregate output. On the cost side, Honohan and Klingebiel (2003) fi nd a correlation between the use of blanket guar-antees and large increases in government debt, although Laeven and Valencia (2008) argue that this may be a refl ection of the severity of the crises in which blanket guaran-tees are used.

Given how costly and uncertain are the instruments to facilitate the recovery, perhaps the main lesson is that it is best to avoid fi nancial crises rather than wait for them to happen. During booms fuelled by large expansion in credit, policy makers should enforce a macro-prudential focus on fi nancial regulation while avoid overly complex regulatory frameworks that constrain economic agents. Once the crisis hits, policies should address the dislocations and capital and labour market.

As this chapter pointed out, financial crises are a fairly common phenomenon in emerging and developing economies, but not so in developed economies. For EU wide policy to respond better to the current crisis, especially in terms of medium to longer-term policies, it is vital to look at the cause of the crisis. One of the causes of the 2008-09 crisis is the global savings imbalance, where massive infl ow of capital from emerging countries has removed the liquidity constraint for corporations and private households in the U.S. and a few other industrial countries, building up large asset price bubbles. Chapter 2 contributes to the debate on the role played by global savings



24

imbalance, focusing in particular on domestic policies related to social security and wage developments.

ReferencesAbreu, D. and B. Brunnermeier “Bubbles and Crashes”, Econometrica, 2003.Adrian, T. and H. Shin “Liquidity and Leverage”, Journal of Financial Intermediation, 2008.Aghion P., P. Bachetta and A. Banerjee “Currency Crises and Monetary Policy in an Economy with Credit

Constraints”, European Economic Review, 2001.Aguiar, M. Investment, devaluation, and foreign currency exposure: The case of Mexico”, Journal of

Development Economics, 2005.Allen, F. and D. Gale, “Financial Contagion”, Journal of Political Economy, 2000.Allen, F. and D. Gale, Competition and fi nancial stability”, Journal of Money, Credit, and Banking, 2004.Angeletos, G., G. Lorenzoni and A. Pavan “Beauty Contests and Irrational Exuberance: A Neoclassical

Approach”, NBER Working Paper W15883, 2010.Auerbach A. and M. Obstfeld “Th e Case for Open Market Purchases in a Liquidity Trap”, Th e American

Economic Review, 2005.Bagehot, W. “Lombard Street”. London: Kegan, Paul, Trench, Trubner & Co., 1894.Ball, L., “Hysteresis in Unemployment: Old and New Evidence”, NBER Working Paper Series, w14818,

2009.Barandiarán E. and L. Hernandéz, “Origins and Resolution of a Banking Crisis: Chile 1982-86”, Central

Bank of Chile working paper, 1999Barro, R. and C. Redlick “Macroeconomic Effects from Government Purchases and Taxes”, Mimeo,

2010.Beck, T., A. Demirgüç-Kunt, R. Levine, “Bank concentration, competition, and crises: First results”,

Journal of Banking & Finance, 2006.Bergoeing, R , PJ Kehoe, TJ Kehoe, R Soto, “A Decade Lost and Found: Mexico and Chile in the 1980s”,

Review of Economic Studies, 2002.Bernanke, B. “Nonmonetary Eff ects of the Financial Crisis in the Propagation of the Great Depression”,

Th e American Economic Review, 1983Bernanke, B. “Essays on Th e Great Depression”, Princeton University Press, Princeton, 2004.Bernanke, B. “Th e Great Moderation”, Remarks at the meetings of the Eastern Economic Association,

Washington, DC February 20, 2004.Bernanke, B. and K. Carey, “Nominal wage stickiness and aggregate supply in the Great Depression”, Th e

Quarterly Journal of Economics, 1996. Bernanke, B. and M. Gertler “Agency Costs, Net Worth and Business Fluctuations”, The American

Economic Review, 1989.Bernanke, B. and M. Gertler “Monetary Policy and Asset Price Volatility”, Federal Reserve Bank of Kansas

City, Kansas City, 2000.Bernanke, B. M. Gertler and S. Gilchrist “Th e Financial Accelerator in a Quantitative Business Cycle

Framework”, in: Woodford and Taylor (eds.), Handbook of Macroeconomics, 1999.Bertola G. and R. Caballero “Irreversibility and Aggregate Investment”, Th e Review of Economic Studies,

1994.

25

Bertola G. and W. Koeniger “Consumption smoothing and the structure of labor and credit markets”, European Economic Review, 2007.

Blanchard, O. “Employment Protection, Sclerosis and the Eff ect of Shocks on Unemployment”, 2000.Blanchard, O. and M. Watson “Bubbles, rational expectations and fi nancial markets”, NBER Working

Paper Series W0945, 1983.Blinder A. and R. Reis “Understanding the Greenspan Standard”, Th e Greenspan Era: Lessons for the

Future, Federal Reserve Bank of Kansas City, 2005.Bordo, M. and O. Jeanne “Boom-Busts in Asset Prices, Economic Instability and Monetary Policy”, NBER

Working Paper Series W8966, 2002.Bordo, M. “Sudden Stops, Financial Crises and Original Sin in Emerging Countries: Déjà Vu?”, NBER

Working Paper Series W12393, 2006.Borenzstein E. and J. Lee, “Financial crisis and credit crunch in Korea: evidence from fi rm-level data”,

Journal of Monetary Economics, 2002.Boyson, N., Stahel, C. and Stulz, R. “Hedge fund contagion and liquidity”, NBER Working Paper Series,

w1406, 2008Brunnermeier, M. “Deciphering the liquidity and credit crunch 2007-2008”, Journal of Economic

Perspectives, 2009.Brunnermeier M. and L. Pedersen “Market Liquidity and Funding Liquidity”, Review of Financial Studies,

2008.Brunnermeier, M., Crockett A., Goodhart C., Persaud, A. and Shin, H. “Th e Fundamental Principles

of Financial Regulation”, International Center for Monetary and Banking Studies and Centre for Economic Policy Research, Geneva, 2009.

Caballero, R. “Specifi city and the Macroeconomics of Restructuring”, MIT Press, Cambridge, 2007.Caballero, R. and A. Krishnamurthy “International and Domestic Collateral Constraints in a Model of

Emerging Market Crises”, Journal of Monetary Economics, 2001.Caballero, R. and A. Krishnamurthy “Bubbles and Capital Flow Volatility: Causes and Risk Management”,

Journal of Monetary Economics, 2006.Caballero, R. and M. Hammour “Jobless Growth: Appropriability, Factor Substitution and

Unemployment”, 1998.Caballero, R., T. Hoshi and A. Kashyap “Zombie Lending and Depressed Restructuring in Japan”,

American Economic Review, 2008.Calomiris, C. “Banking Crises and the Rules of the Game”, mimeo, 2009Calvo, G. “Capital Flows and Capital-Market Crises: Th e Simple Economics of Sudden Stops”, 1998.Calvo, G. and E. Mendoza, “Capital Market Crises and Economic Collapse in Emerging Markets: An

Informational Frictions Approach”, American Economic Review 2002.Calvo, G., A. Izquierdo and L. Mejia, “On the Empirics of Sudden Stops: Th e Relevance of Balance Sheet

Eff ects”, NBER Working Paper Series, 2004.Calvo, G., A. Izquierdo and E. Mejia, “Phoenix Miracles in Emerging Markets: Recovering Without

Credit From Systemic Financial Crises”, 2006.Calvo, G. and E. Talvi, “Sudden stop, fi nancial factors and economic collapse in Latin America: learning

from Argentina and Chile”, in: Th e Washington Consensus reconsidered: towards a new global gov-ernance (Serra, N and Stiglitz J., eds.), Oxford University Press, Oxford, 2008.

Caprio G. and D. Klingebiel, “Bank Insolvencies: Cross Country Experiences”, World Bank Research Working Papers, 1999.



26

Cecchetti, S. H. Genberg, S. Wadhwani, “Asset Prices in a Flexible Infl ation Targeting Framework”, NBER Working Papers, w8970, 2002.

Cerra V. and S. Saxena, “Growth Dynamics: Th e Myth of Economic Recovery”, American Economic Review, 2008.

Chang, R. and A. Velasco, “Th e Asian liquidity crisis”, NBER Working Paper Series, W6796 , 1998Christiano L., M.S. Eichenbaum and S. Rebelo, “When is the Government Spending Multiplier High?”,

2009.Christiano L., C. Gust and J. Ruldos, “Monetary Policy in a Financial Crisis”, Journal of Economic Th eory

2004.D’Avolio, G. “Th e Market for Borrowing Stock”, Journal of Financial Economics, 2002.Del Negro, M., G. Eggertsson, A. Ferrero and N. Kiyotaki “Th e Great Escape?”, Mimeo. 2010.Dell’Ariccia, G., E. Detragiache and R. Rajan “Th e Real Eff ecs of Banking Crises”, Journal of Finance,

2008.Demirguc-Kunt, A. and E. Detragiache “Financial Liberalization and Financial Fragility”, World Bank

Policy Reseaerch Working Paper No. 1917 , 1999.Demirguc-Kunt, A. and E. Detragiache “Does deposit insurance increase banking system stability? An

empirical investigation”, Journal of Monetary Economics, 2002.Demirguc-Kunt, A. and L. Serven “Are all the Sacred Cows Dead? Implications of the Financial Crisis for

Macro and Financial Policies”, World Bank Policy Research Working Paper No. 4807, 2009.Desai, M., C. Foley and K. Forbes, “Financial constraints and growth: Multinational and local fi rm re-

sponses to currency depreciations”, Review of Financial Studies, 2008.Diamond and Dybvig “Bank runs, deposit insurance and liquidity”, Th e Journal of Political Economy,

1983.Diaz-Alejandro, C. “Good-Bye Financial Repression, Hello Financial Crisis”, Journal of Development

Economics, 1985.Drees, B. and C Pazarbaşioğlu, “The Nordic banking crises: pitfalls in financial liberalization?”,

International Monetary Fund Occasional Paper, 1998.Eggertsson and Woodford “Th e Zero-Interest Rate Bound and Optimal Monetary Policy”, Brookings

Papers on Economic Activity, 2003.Eggertson “Th e Paradox of Toil”, Federal Reserve Bank of New York Staff Report, No 433, 2010.Eichengreen, B. “From Great Depression to Great Credit Crisis: Similarities, Differences and

Lessons.” Presented at the 50th Economic Policy Panel Meeting, held in Tilburg on October 23-24, 2009

Eichengreen, B., R. Hausmann and U. Panizza, “Th e Pain of Original Sin”, in: Other People’s Money: Debt Denomination and Financial Stability in Emerging Market Economies”, Eichengreen B. and R. Hausmann (eds.), 2005.

Elsby, M., B. Hobijn, A. Sahin, “Th e Labor Market in the Great Recession”, NBER Working Paper 15979, 2010.

Fallon P. and R.E. Lucas, “Th e Impact of Financial Crises on Labor Markets, Household Incomes, and Poverty: A Review of Evidence”, Th e World Bank Research Observer, vol. 17, no. 1, 2002.

Farhi, E. and J. Tirole “Competing Liquidities: Corporate Securities, Real Bonds and Bubbles”, NBER Working Paper No. 13955, 2008.

Garrison, “Time and Money: the Macroeconomics of Capital Structure”, Routledge, London, 2001.

27

Geanakoplos J. and H. Polemarchakis, “Existence, regularity, and constrained suboptimality of com-petitive allocations when the asset market is incomplete”, in: Essays in honor of Kenneth J. Arrow, Volume 3 (Heller, W., Starr, R., Arrow, K.J. and Starret D. eds.), 1986.

Gertler, M., S. Gilchrist and F. Natalucci, “External constraints on monetary policy and the fi nancial ac-celerator”, Journal of Money, Credit and Banking, 2007.

Giavazzi, F., T. Japelli and M. Pagano, “Searching for non-linear eff ects of fi scal policy: evidence from industrial and developing countries”, European Economic Review, 2000

Gilchrist S. and J. Sim, “Investment During the Korean Financial Crisis: A Structural Econometric Analysis”, NBER Working Paper Series, w13315, 2007.

Glick, R. and A. Rose “Contagion and trade:: Why are currency crises regional?”, Journal of International Money and Finance, 1999.

Goldstein M., Kaminsky G. and C. Reinhart “Assessing Financial Vulnerability: An Early Warning System for Emerging Markets”, Institute for International Economics, 2000.

Gorton, G. “Banking panics and business cycles”, Oxford economic papers, 1988Gourinchas, P., R. Valdés and O. Landerretche, “Lending Booms: Latin America and the World”,

Economia, 2001.Grossman, G. and N. Yanagawa, “Asset Bubbles and Endogenous Growth”, Journal of Monetary

Economics, 1993.Gurkaynak, R. “Econometric tests of asset price bubbles: taking stock Journal of Economic Surveys”,

2008.Hall, R. “Employment fl uctuations with equilibrium wage stickiness”, American Economic Review, 2005.Hayashi, F. and E. Prescott, “Th e 1990’s in Japan: A Lost Decade”, Review of Economic Dynamics, 2002.He, Z. and W. Xiong, “Dynamic Bank Runs”, Mimeo, 2009.Hellman, T., C. Murdock and J. Stiglitz “Liberalization, moral hazard in banking, and prudential regula-

tion: are capital requirements enough?”, American Economic Review 2000.Hellwig, C. and G. Lorenzoni, “Bubbles and Self-Enforcing Debt”, Econometrica, 2009.Hong, H. and J. Stein “Diff erences of Opinion, Short-Sales Constraints, and Market Crashes”, Review of

Financial Studies, 2003.Honohan P. and D. Klingebiel, “Th e fi scal cost implications of an accommodating approach to banking

crises”, Journal of Banking and Finance, 2003.IMF, “World Economic Outlook Report – Rebalancing Growth”, International Monetary Fund, 2010.IMF, “World Economic Outlook Report – Crisis and Recovery”, International Monetary Fund, 2009.Jermann, U. and V. Quadrini, “Financial Innovation and Macroeconomic Volatility”, NBER Working

Paper Series, w12308.Kaminsky G. and C. Reinhart “On Crises, Contagion and Confusion”, Journal of International Economics

1999.Kaminsky G. and C. Reinhart “The Twin Crises: The Causes of Banking and Balance-of-Payments

Problems”, American Economic Review 2000.Kane E. “Th e S&L Insurance Mess, How Did it Happen?”, Th e Urban Institute Press, Washington D.C.,

1989.Kane E. and D. Klingebiel “Alternatives to blanket guarantees for containing a systemic crisis”, Journal of

Financial Stability, 2004.Kehoe T. and E. Prescott (Eds.) “Great Depressions of the Twentieth Century”, Federal Reserve Bank of

Minneapolis, Minneapolis, 2007.



28

Keynes, J.M. “Th e General Th eory of Employment, Interest and Money”, Macmillan Cambridge University Press, for Royal Economic Society, 1936.

King, M. and S. Wadhwani, “Transmission of volatility between stock markets”, Review of Financial Studies, 1990.

King, R. and R. Levine “Finance and growth: Schumpeter might be right”, Th e Quarterly Journal of Economics, 1993.

King, R., C. Plosser and S. Rebelo, “Production, growth and business cycles”, Journal of Monetary Economics, 1988.

Kiyotaki N. and J. Moore “Credit Cycles”, Journal of Political Economy, 1997.Kiyotaki N. and J. Moore “Credit Chains”, Mimeo, 1997.Kiyotaki N. “Credit and Business Cycles”, Japanese Economic Review, 1998.Kiyotaki N. and J. Moore “Evil is the Root of all Money”, Mimeo, 2001.Kiyotaki N. and J. Moore “Liquidity, Business Cycles and Monetary Policy”, Mimeo, 2008.Kiyotaki N. and R. Wright “On Money as a Medium of Exchange”, Journal of Political Economy, 1989.Kodres, L. and M. Pritsker, “A rational expectations model of financial contagion”, Journal of

Finance, 2002.Kornai, J., E. Maskin and G. Roland, “Understanding the soft budget constraint”, Journal of Economic

Literature, 2003.Kroesner, R., L. Laeven and D. Klingebiel, “Banking Crises, Financial Dependence and Growth”, Journal

of Financial Economics 2007.Krueger, A. and A. Tornell “Th e role of bank restructuring in recovering from crises: Mexico 1995-98”,

NBER Working Paper Series, w7042, 1999.Krugman, P. “It’s Baaack: Japan’s slump and the return of the liquidity trap”, Brookings Papers on

Economics, 1998Laeven L. and F. Valencia “Systemic Banking Crises: A New Database”, IMF Working Paper No. 08/224,

2008.Laeven L. and F. Valencia “Th e Use of Blanket Guarantees in Banking Crises”, IMF Working Paper No.

08/250, 2008.Lewis, M. “Th e Big Short: Inside the Doomsday Machine”, W. W. Norton & Company, 2010.Lorenzoni G. “Ineffi cient Credit Booms”, Review of Economic Studies, 2008.Lucas, R. “On the Mechanics of Economic Development”, Journal of Monetary Economics, 1988Martin, A. and Ventura, J. “Economic Growth with Bubbles”, NBER Working Paper 15870, 2010McGrattan, E., and E. Prescott “Technical Appendix: Unmeasured Investment and the Puzzling US Boom

in the 1990s.” Federal Reserve Bank of Minneapolis Staff Report 395, 2007.Mertens, K. and M. Ravn, “The Aggregate Effects of Anticipated and UnanticipatedU.S. Tax Policy

Shocks: Th eory and Empirical Evidence.” Mimeo, 2008.Mendoza, E., “Sudden Stops, Financial Crises and Leverage”, Th e American Economic Review, 2010

(forthcoming).Miller E. “Risk, Uncertainty and Divergence of Opinion”, Journal of Finance, 1977.Minsky, H. “Stabilizing and Unstable Economy”, Yale University Press, 1986.Mondria, J. and Quintana-Domequez, C. “Financial contagion and attention allocation”, Mimeo, 2010.Morris, S. and H. Shin, “Unique equilibrium in a model of self-fulfi lling currency attacks”, American

Economic Review, 1998.Mountford A. and H. Uhlig “What are the effects of fiscal policy shocks?”, Journal of Applied

Econometrics, 2009.

29

Neumeyer, P. and F. Perri, “Business Cycles in Emerging Economies: Th e Role of the Interest Rate”, Journal of Monetary Economics, 2004.

Ohanian, L., “Why Did Productivity Fall So Much During the Great Depression”, American Economic Review, 2001

Ohanian, L. “What or Who Started the Great Depression”, Journal of Economic Th eory, 2009.Olivier, J., Growth-enhancing Bubbles, International Economic Review, 2000.Rajan, R. and L. Zingales “Financial Dependence and Growth”, American Economic Review, 1998.Ramey, V. “Identifying government spending shocks: It’s all in the timing”, NBER Working Paper Series,

w15464, 2009. Rampini, A. and S. Viswanathan, “Collateral, risk management, and the distribution of debt capacity”,

Journal of Finance, forthcoming.Sachs J., A. Tornell and A. Velasco, “Financial Crises in Emerging Markets: Th e Lessons from 1995”,

Brookings Papers in Economic Activity, 1996.Samuelson P. “An Exact Consumption-Loan Model of Interest with and without the Social Contrivance

of Money”, Th e Journal of Political Economy, 1958.Scheinkman J. and W. Xiong, “Overconfi dence and Speculative Bubbles”, Journal of Political Economy,

2003.Shiller, R. “From Effi cient Markets Th eory to Behavioral Finance”, Journal of Economic Perspectives,

2003.Shimer, R. “Th e cyclical behavior of equilibrium unemployment and vacancies”, American Economic

Review, 2005.Svensson, L. “Th e Zero Bound in an Open Economy: a Foolproof Way of Escaping From a Liquidity Trap”,

NBER Working Paper Series W7957, 2000Svensson, L. “Escaping From a Liquidity Trap and Defl ation: Th e Foolproof Way and Others”, Journal of

Economic Perspectives, 2002 Stern, N., Review on the Economics of Climate Change, H.M. Treasury, UK, October, 2006.Stiglitz, J. “Th e ineffi ciency of the stock market equilibrium”, Th e Review of Economic Studies, 1982.Mendoza, E. and M. Terrones, “An Anatomy of Credit Booms: Evidence from Micro and Aggregate Data”,

NBER Working Paper W14049, 2008Mendoza, E. “Sudden Stops, Financial Crises and Leverage”, American Economic Review, 2010Ranciere, R., A. Tornell and F. Westermann “Crises and Growth: A Re-Evaluation,” NBER Working

Papers W10073, 2003.Reinhart C. and K. Rogoff “This Time is Different: Eight Centuries of Financial Folly”, Princeton

University Press, Princeton, 2010.Saint Paul, G., “Fiscal Policy in an Endogenous Growth Model”, Quarterly Journal of Economics, 1992.Schwartzman “Time to Produce and Emerging Market Crises”, Mimeo, 2010.Stiglitz J. “Th e Ineffi ciency of the Stock Market Equilibrium”, Th e Review of Economic Studies, 1982.Tirole J. “On the Possibility of Speculation Under Rational Expectations”, Econometrica, 1982.Tirole J. “Asset bubbles and overlapping generations”, Econometrica, 1985.Tornell, A. and F. Westermann “Boom-Bust Cycles and Financial Liberalization”, MIT Press Books,

Cambridge, 2005. Weismer, E. and Weil, P. “Th e Macroeconomics of Labor and Credit Market Imperfections”, Th e American

Economic Review, 2004.Woodford, M. “Interest and Prices: Foundations of a Th eory of Monetary Policy”, Princeton University

Press, Princeton, 2003.



30

Zawadowski, A. “Entangled Financial Systems”, Mimeo, 2010.Zingales, L., “Causes and Eff ects of the Lehman Brothers Bankruptcy”, Written Testimony Before the

Committee on Oversight and Government Reform, United States House of Representatives, October 6, 2008.

Documents

Literature review of past crises · 1 FINANCIAL CRISES —A REVIEW OF THE LITERATURE Main Findings z While a major fi nancial crisis is a novelty for policy makers in the US and major