Els Opuscles del CREI

num. 35
December 2013

Credit cycles
and systemic risk
José-Luis Peydró
The Centre de Recerca en Economia
Internacional (CREI) is a research centre
Credit cycles
sponsored by the Universitat Pompeu Fabra (UPF) and systemic risk
and the Generalitat de Catalunya. It is located at
the campus of the UPF in Barcelona.
José-Luis Peydró
CREI’s objective is to promote research in
international economics and macroeconomics
subject to the highest standards of academic
scholarship.

Els Opuscles del CREI are the instruments for the

diffusion of CREI’s research beyond the academic
world. Each Opuscle, aimed at a general public,
gathers together conclusions and observations of
specialized papers published, or about to be
published, in professional journals. The opinions
stated in Els Opuscles del CREI are the sole 1. Introduction
responsibility of the authors.
Western Europe and USA suffered recently a
Pdf version of this opuscle and all the others strong banking crisis, followed by a severe eco-
Opuscles del CREI can be downloaded at: nomic recession with important costs in terms of
www.crei.cat/opuscles.php aggregate output and employment. These phe-
nomena are not unique: Banking crises are recur-
rent phenomena, triggering deep and long-lasting
recessions (see Kindelberger (1978) and Reinhart
and Rogoff (2009a) for historical evidence and Fig-
ure 1 based on Laeven and Valencia (2012) on
recent systemic banking crises). The main channel
by which banks’ balance-sheet weaknesses affect
Editorial Board the real economy is via a reduction of the supply
Antonio Ciccone (editor)
Jordi Galí of credit, a credit crunch, and also through some
Teresa Garcia-Milà compositional changes of credit supply as the so-
Jaume Ventura called zombie lending (or loan ever-greening) or
Published by: CREI a significant reduction of appetite for risk. Impor-
Universitat Pompeu Fabra tantly, banking crises are not random events that
Ramon Trias Fargas, 25-27 08005 Barcelona
Tel. 93 542 28 26
come from exogenous risks, but come after peri-
© CREI, 2013 ods of very strong private credit growth (Schular-
© of this edition: José Luis Peydró ick and Taylor, 2012). Therefore, for systemic risk,
Design: Fons Gràfic it is crucial to understand the determinants and
Printed by: Masanas Gràfiques
implications of credit in good and bad times —
ISSN: 1137 - 7828
Dipósito legal: B. 3121-2014 the so-called credit cycles.

1
 This opuscle analyzes the relationship between

rest are systemic events. A banking crisis is defined as systemic if two conditions are met: (1) significant signs of financial distress in the banking sector,
Based on Laeven and Valencia (2012). The sample includes 116 countries, with a total of 147 banking crises, of which 26 are border events and the
credit cycles and systemic risk, where systemic 2009
risk is defined, based on Freixas, Laeven and Pey- 2008
dró (2014), as “the risk of threats to financial sta- 2007
bility that impair the functioning of the financial 2006
system as a whole with significant adverse effects
on the broader economy.” In particular, I analyze
2005

the following questions: Are credit cycles one of 2004

the main determinants of the likelihood and se- 2003

verity of systemic financial crises? What is the role 2002
of the banking sector? Do financial innovation, 2001

and (2) significant banking policy intervention measures in response to significant losses in the banking system.
globalization, deficient corporate governance and 2000
market disciplining, and public policy (includ- 1999
ing macroprudential and monetary policy) affect
credit cycles?1
1998
1997
In the rest of this opuscle, first, I distinguish 1996
between credit cycles caused by financial frictions 1995
in non-financial borrowers (the demand side), 1994
and the ones by frictions in banks (credit supply 1993

Figure 1. Number of banking crises starting each year, 1976-2009

cycles). Second, I analyze credit supply in good
times and its implications for the endogenous
1992

building-up of excessive bank risk-taking. Third,

1991

I analyze credit supply at the time of financial cri- 1990

ses. Finally, I offer some concluding remarks. In 1989
all these cases, I examine the sources of excessive 1988
credit cycles, in particular financial globalization 1987
and deregulation, financial innovation (securitiza- 1986
tion), deficient corporate governance and market
disciplining, and monetary and prudential policy.
1985
1984
1983
1982
1981
1980
1979
1978
1977
1976
25

20

15

10

2 3
2. Financial frictions of aligning the incentives between the principal
(for instance, bank bondholders or the taxpayers)
and credit cycles and the agent (bank managers or shareholders).
First, the basic agency problem stems from the fact
The cycles in credit growth consist of periods that most financial intermediaries have limited li-
during which the economy is performing well and ability (their losses are limited) and invest money
credit growth is robust (on average 7%) and pe- on behalf of others (the final investors). Moreover,
riods when the economy is in recession or crisis they are highly leveraged, notably banks that are
and credit contracts (on average -2% for a sam- funded almost entirely with debt (some banks are
ple of 14 major developed countries over the last funded with 50 units of debt over 1 of equity and
140 years, see Schularick and Taylor (2012) and many have leverage ratios higher than 30 to 1).
their following papers with Oscar Jordà). Figure 2 These frictions create strong incentives for exces-
shows aggregate granted credit and output cycles sive risk-taking as there is little skin in the game
in the USA and Spain since 2000. Granted credit for bank shareholders but high potential upside
is not as forward looking as change in the supply profits. Second, excessive risk-taking notably in-
of committed credit as it is also affected by cred- creases when there are explicit and implicit guar-
it demand, notably drawn of existing credit lines; antees and subsidies from the government (tax-
instead, change in lending standards from lending payers) in case of negative ex-post aggregate risks
surveys from central banks is more forward look- (such as a financial crisis). This increases ex-ante
ing (see Maddaloni and Peydró, 2011). agency problems of financial intermediaries as fi-
nancial gains are privatized, but losses are in great
Credit cycles stem from either: (1) Non-finan- part socialized.3
cial borrowers’ agency frictions and investment
opportunities (credit demand) as in, for example, For example, the excessive credit boom and
Bernanke and Gertler (1989), Kiyotaki and Moore lending standards deterioration in the USA and
(1997), Lorenzoni (2008), and Jeanne and Korinek Spain real estate market before the recent crisis
(2010), where better investment opportunities or has partly been blamed on several factors: (i) the
better firms’ and households’ collateral and net financial innovation that fostered in the USA the
worth imply higher credit, or (2) banks’ agency development of an unregulated shadow bank-
frictions (credit supply) as in, for example, Rajan ing system to arbitrate (evade) bank capital reg-
(1994), Holmstrom and Tirole (1997), Diamond ulation; (ii) in the USA and Spain strong funding
and Rajan (2006), Allen and Gale (2007), and liquidity through securitization sold to foreign fi-
Adrian and Shin (2011), where changes in bank nancial intermediaries due to financial globaliza-
capital, liquidity and competition allow changes tion (in Spain more covered bonds than ABS); (iii)
in credit supply. deficient corporate governance and lack of mar-
ket discipline; (iv) very loose monetary policy (in
The main explanation of credit supply cycles is Spain real interest rates were negative as Germany
based on an agency view. I believe that one can- was having low GDP growth and the ECB had
not understand systemic risk without this agency therefore too low monetary rates for countries like
view of risk-taking.2 The agency view highlights Spain and Ireland); and (v) deficient prudential
agency problems at the core of the build-up of regulation and supervision, both micro and mac-
systemic risk that have to do with the difficulties ro. Moreover, the potential government bailouts

4 5
6
Figure 2. Credit and GDP cycles for the USA and Spain, 2000-2013
US
Credit Growth (left axis) GDP Growth (right axis)
% %
30

25 8

20

15 4

10

5 0

00Q1
00Q3
01Q1
01Q3
02Q1
02Q3
03Q1
03Q3
04Q1
04Q3
05Q1
05Q3
06Q1
06Q3
07Q1
07Q3
08Q1
08Q3
09Q1
09Q3
10Q1
10Q3
11Q1
11Q3
12Q1
12Q3
13Q1

-5 -4

-10

-15 -8

USA: ‘credit’ is defined as ‘Loans and leases in bank credit, all commercial banks’. GDP is in real terms. Both variables are seasonally adjusted
and taken from the website of the Federal Reserve Bank of Saint Louis. They are expressed in annual growth rates.

Spain
Credit Growth (left axis) GDP Growth (right axis)
% %
30

25 8

20

15 4

10

5 0

00Q1
00Q3
01Q1
01Q3
02Q1
02Q3
03Q1
03Q3
04Q1
04Q3
05Q1
05Q3
06Q1
06Q3
07Q1
07Q3
08Q1
08Q3
09Q1
09Q3
10Q1
10Q3
11Q1
11Q3
12Q1
12Q3
13Q1

-5 -4

-10

-15 -8

Spain: ‘credit’ is ‘Credit by credit institutions to other resident sectors’, from the Bank of Spain. GDP is in real terms from the INE. Both variables

7
are seasonally adjusted and expressed in annual growth rates.
imply a lack of market disciplining by bank credi- small and medium banks become systemic since
tors by not imposing losses on these debt-holders the government may ex-post bailout them, as oth-
(as in the case of Ireland), which creates ex-ante erwise there would be too many to fail (Acharya
moral hazard and appetite for excessive risk. Im- and Yorulmazer, 2007). Deficient corporate gov-
portantly, this view is based on agency problems ernance where bank managers maximize only
in both the private sector (financial intermediar- bank shareholder value (a small part of the bank
ies and their investors and managers) and in the total assets) with executive compensation based
public sector (central banks, supervisory agencies, on relative performance with stock options and
and regulatory bodies).4 lack of claw-back options also encourage this type
of excessive risk-taking.
The theory suggests that financial intermedi-
aries might take excessive ex-ante risks, increas-
ing collectively the systemic risk in the financial 3. Credit and imbalances
system. But what are the specific factors and de-
cisions that will cause excessive risks? The main in good times
channel is excessive credit and leverage. In fact,
these variables show the strongest ex-ante corre- The recent financial crisis has come after a pe-
late with the incidence of financial crises as shown riod of significant credit expansion. In order to un-
in the empirical literature analyzing large historical derstand systemic risk, we need to know whether
and cross-country episodes of systemic financial this fact is unique to this crisis or shared among
crises. Private credit (debt and leverage) acceler- many financial crises. However, financial crises are
ation notably increases the likelihood of financial not frequent events, and hence in order to study
crises, and conditional on a crisis occurring, it in- the determinants of such crises it is necessary to
creases its systemic nature and the negative effects use long time series for several countries. A set of
on the real economy associated with the crisis.5 papers by Reinhart and Rogoff and Schularick and
Taylor focuses precisely on this issue.7 Both sets
Credit booms, however, may also result from identify periods of strong debt growth preceding
(and promote) sound economic fundamentals banking crises. Reinhart and Rogoff’s focus is on
(demand-driven credit) and, therefore, could be government and private debt, while for Schularick
benign for systemic risk.6 For example, since 1970s and Taylor focus on bank credit. Not only does
across a broad range of countries, research has the likelihood of systemic crises increases with
shown that two thirds of credit booms did not end ex-ante debt, but they also show that when a crisis
up in a financial crisis (IMF, 2012). Therefore, a occurs, the real costs are higher if the prior debt
key question that we analyze is what are the deter- increase has been higher. Moreover, Jordà et al.
minants of the bad credit booms, in particular the (2011 and 2013) show that the main determinant
ones associated with credit supply (i.e. based on is ex-ante private (bank) credit rather than public
pervasive bank incentives). Credit supply booms debt or external debt.
that are negative for systemic risk generally stem
from correlated risk-exposures by the financial in- Reinhart and Rogoff cover almost eight centu-
termediaries that end up developing asset-price ries for 66 countries, both advanced and develop-
bubbles in real estate or in other asset classes. This ing ones. They find: First, external debt increas-
herding by financial intermediaries may also make es sharply in advance of banking crises. Second,

8 9
banking crises tend to lead sovereign-debt crises Their findings suggest that the prior evolution
(with an increase in domestic government debt). of credit shapes the business cycle. This has im-
Their results suggest that banking crises increase portant implications for macroeconomic models: if
broader debt crises (Kaminsky and Reinhart, 1999). credit were to just follow economic fundamentals
All in all, the results point out that ex-ante lever- and had little impact on the business cycle, then
age is crucial to explain banking crises, and that models omitting the frictions in the financial sector
public (sovereign) debt crises tend to be a con- might be sufficient. Nevertheless, these findings
sequence, rather than a cause, of banking crises. suggest that more sophisticated macro models fea-
turing financial intermediation are needed.8
Schularick and Taylor (2012) analyze the re-
lationship of financial crises with aggregate bank Using the same dataset, Jordà et al. (2011)
credit growth. They build a 140-year panel data analyze whether ex-ante external imbalances in-
set for fourteen developed countries and construct crease the risk of financial crises. In other words,
bank credit and total asset series. Before the Great are external imbalances associated with higher
Depression, money and credit aggregates have a costs in the recession, or are credit booms the
stable relationship with GDP, increasing before the only important variable? Their overall finding is
crisis and decreasing afterwards. After the 1940s, that ex-ante credit growth emerges as the single
credit itself decoupled from broad money by in- best predictor of financial instability; however, the
creasing leverage and funding via nonmonetary correlation between current account imbalances
liabilities of banks. and credit booms has increased significantly in
the recent decades, which indicates that financial
Schularick and Taylor (2012) also analyze the globalization plays a role as well. In a globalized
likelihood and severity of financial crises and show economy, with free capital mobility, credit cycles
that changes in bank loans are a strong predictor and foreign capital flows have the potential to re-
of financial crises. Furthermore, broad money ag- inforce each other more strongly than otherwise
gregates do not have the same predictive power, (on this argument, see also Shin, 2012). Clearly, a
particularly in the post-1940 period. Jordà, Schu- strong and sustained credit boom cannot typically
larick, and Taylor (2014) study the role of credit in be financed with an increase of domestic depos-
the whole business cycle, not only around finan- its and wealth (especially if not driven by very
cial crises. They find that financial-crisis recessions strong fundamentals); therefore, foreign liquidity,
are more costly than normal recessions in terms of or liquidity stemming from expansive monetary
lost output, and for both types of recessions, they policy or financial innovation (e.g. securitization),
show that the financial imbalances built up in the need to be present and interact with credit cycles.9
period preceding the crisis (bank credit) are im- Finally, Jordà et al. (2013) show that the main de-
portant drivers of the strong negative real effects terminant is ex-ante private (bank) credit rather
to the broad economy during the crisis. Specifical- than public debt.
ly, not only does ex-ante credit growth affect the
likelihood of a financial crisis, but conditional on a The historical evidence clearly suggests that
crisis, the real effects are worse when the crisis is high rates of credit growth coupled with widen-
preceded by a credit boom. Therefore, this histori- ing imbalances pose financial stability risks that
cal analysis shows that ex-ante financial imbalanc- policy makers and academics should not ignore.
es are a first-order determinant of systemic risk. Moreover, in the recent crisis, the credit booms

10 11
and large current account imbalances in many Determinants of credit supply booms
countries, low levels of short-term (monetary) and and other financial imbalances
long-term rates, and increasing recourse to secu-
ritization, all seem to confirm that credit growth We have seen that credit booms appear to
and capital inflows and other forms of liquidity precede financial crises, but only one third ends
nowadays interact in a stronger way. Maddaloni up in a crisis, thus an important part of credit
and Peydró (2011) analyze these issues for the re- booms are driven by strong economic fundamen-
cent crisis. Using the survey of lending conditions tals and do not pose a risk for systemic risk. What
and standards for the Euro area countries and the are the determinants of credit supply booms and
USA that the national central banks and regional other financial imbalances?
Feds request from banks, they analyze the deter-
minants of lending conditions and standards for One of the key questions to understand prob-
the financial crisis that started in 2007. They find lems with pervasive bank incentives is whether
that countries with worse economic performance bankers were aware of the excessive risk-taking
during the crisis are those with ex-ante softer of their institutions. In Akin, Marin and Peydró
lending conditions. They also find that lower mon- (2013), we test for US banks whether banks that
etary policy rates imply softer lending conditions performed the worst in the 2007–08 crisis were
and standards. However, after controlling for key correlated to bank insiders’ net sale of shares in
factors (such as country fixed effects and busi- the period prior to the initial fall in house prices
ness cycle conditions), current account deficits in the middle of 2006. If that is the case, then it
or lower long-term interest rates do not correlate would suggest that insiders knew about the risks
with softer lending conditions. Finally, Maddaloni in their institutions. We find robust evidence that
and Peydró (2011) find that lending standards are on average ex-ante insiders’ net sale of shares im-
pro-cyclical (in the upside of the business cycle, plies worse bank performance during the crisis.
lending conditions are softer and banks take on Importantly, the negative relationship becomes
higher risk), a result consistent with Jiménez and more significant for top officers such as CEO and
Saurina (2006) for Spain. CFO, i.e. the ones with the highest set of infor-
mation, for bigger banks (i.e. banks with higher
Credit booms are therefore a crucial ex-ante agency problems) and for banks more engaged
correlate of financial crises. Yet, all of these em- in real estate. In other words, results suggest that
pirical analyses condition on the occurrence of a bank insiders knew that they were taking exces-
crisis and ask what its determinants are. But, do sive risks.10
all credit booms end up in a crisis? The IMF (2012)
analyzes credit booms for 170 countries over the One of the main agency channels, especially
last 40 years of data. They show three important highlighted by the media, is bank compensation.
results: credit booms have become more frequent Compensation for the top officers may not be as
after the 1980s (a period of significant financial important for the build-up of systemic risk as the
deregulation); most booms happen in relatively remuneration structure of loan officers. Contracts
underdeveloped financial systems; and only one to loan officers that maximize loan volume imply
in three credit booms ends up in a financial crisis. excessive risk-taking in lending (as it was the case
in Spain). But why do banks choose compensa-
tion structures that promote excessive risk-taking

12 13
of bank top managers and middle managers as mediaries and by the associated phenomenon of
loan officers? Why were these compensation struc- relative performance evaluation. Furthermore, as
tures designed in the first place? Why is corporate explained in Section 2 of this opuscle and in Frei­
governance not working well in banking? xas et al. (2014), corporate governance in banking
is not working well, as maximizing bank share-
One cannot explain well the heterogeneity in holders’ value can lead to excessive risk in banks
lending standards both during the cycle and across — i.e., negative externalities to bank debtholders
financial intermediaries without focusing on the and depositors, to taxpayers and to non-financial
role of incentives and institutions. Since credit borrowers (firms and households).
decisions are usually delegated to agents inside
financial institutions, if one wants to understand A classical example in banking is Rajan’s (1994)
what causes the changes in lending standards one model, which develops a theory why bank cred-
needs to understand the incentives that financial it policies fluctuate over the business cycle. In a
intermediaries face. But these incentives are also rational profit-maximizing world, banks should
influenced by regulations, accounting standards, maintain a credit policy of lending if and only if
financial competition, innovation, central bank borrowers have positive net present value (NPV)
policies, corporate governance, and compensation projects. Therefore, a change in the level of bank
structures (see Stein, 2013). Many of these prob- credit should be a consequence only of a change
lems had already been mentioned before the cri- in the credit quality of borrowers (a change in the
sis. As explained by Rajan (2005) at the Jackson economic fundamentals). Bank supply of credit
Hole conference when he was the chief econo- should not exert an independent influence on the
mist at the IMF, a fundamental challenge in finan- level of credit.
cial intermediation (notably delegated investment
management) is that many quantitative rules are Bank managers try to maximize bank’s earn-
vulnerable to agents who act to boost measured ings, but they also care about their reputation:
returns by selling insurance against unlikely events they care about stock and labor market’s percep-
— what is known as tail risk. tion of their abilities. However, the composition of
the bank portfolio and the specific performance
Since credit risk by its nature involves an ele- of borrowers are not immediately observable by
ment of put-writing (initial fees and interest pay- the market, as it can only observe banks’ earnings.
ments during the life of the loan but a medium Therefore, bank managers attempt to manipulate
term risk of default on the principal) it is always current earnings to shape market’s perceptions,
going to be challenging in an agency context, es- for instance by extending the maturity of a bad
pecially to the extent that the risks associated with loan: Extending the term of loans, lending new
the put-writing can be structured to partially evade money, and weakening covenants so as to avoid
the relevant measurement scheme. Even more in recognizing default — i.e., loan ever-greening or
bank loans, one can avoid defaults by simply re- zombie lending. Similarly, the bank may attempt
newing or lengthening the loan maturity (i.e., a to convince the market of the profitability of its
practice known as loan ever-greening or zombie lending by promoting a soft credit policy that
lending; see Caballero, Hoshi and Kashyap, 2008). generates up-front fees at the expense of future
Moreover, these agency problems may be exac- credit quality (like the put option in the previous
erbated by competitive pressures among inter- example).

14 15
 In general, a liberal credit policy boosts current paioannou, and Peydró (2010) we study the rea-
earnings at the expense of future performance. sons behind the increase in financial integration.
The bank is trapped into this second-best credit We conclude that the elimination of currency risk
policy simply because the market expects it. The is the main driver, although financial regulation
market is more forgiving of a bank’s poor perfor- convergence played an important role as well.
mance if it knows that the entire borrowing sector However, as we show in Kalemli-Ozcan, Papaio-
has been hit by a systematic and unpredictable ad- annou, and Peydró (2013), financial globalization
verse shock. When multiple banks lend to a sector, has reduced the synchronization in economic ac-
the market learns something about the systematic tivity in normal times. This suggests that when a
component of uncertainty from each bank’s earn- country is hit by a positive shock — or it is having
ings. This informational externality makes bank a credit bubble — foreign capital will flow into
credit policies interdependent. A bank’s reputa- the country, thereby augmenting even more the
tion is less sensitive to poor earnings when other amount of funds available at the expense of other
banks admit to poor earnings. Because true earn- countries.
ings are less likely to be high when the borrowing
sector is distressed, banks collectively coordinate As we have seen in the previous section, in
(herd) on an adverse shock to borrowers to tight- the second part of the 20th century monetary and
en credit policy. In addition, banks are more ea- credit aggregates started to behave very differ-
ger to declare loan defaults when other banks do ently. This points to another element that allows
so because there may be too many banks to fail increasing credit even if no additional funds are
and thus a bailout is easier to obtain. This theory available: financial innovation. Securitization is
therefore yields systemic risk through credit sup- an activity that allows the bank to sell illiquid
ply cycles that stem from financial intermediaries’ balance sheet items such as loans to third parties,
limited liability, compensation structure and gov- thereby increasing their liquidity and decreas-
ernments’ policies. ing capital requirements. In Jiménez, Mian et al.
(2013), we focus on the effects of securitization
Once we understand the agency problems in on lending from Spanish banks, where securitiza-
the banking system, we need a ‘trigger’ for a par- tion was done using real-estate loans.11 We find
ticular credit boom. The worst credit booms are very interesting results that shed light on how
usually funded by more than just the local sav- credit booms are characterized. Securitization
ings of the economy. I am going to focus here on did not affect firms with already strong access to
four additional distortions: too much reliance on the banking sector. However, it was the exten-
wholesale funding, the increase in available funds sive margin (i.e., new borrowers) the ones that
due to globalization, the effects of financial inno- were more affected by credit volume supplied.
vation such as securitization, and the effects of Banks engaged in real estate activities were able,
monetary policy on the credit expansion. through securitization, to expand their credit to
new borrowers, which are usually riskier and, in
One of the main reasons why it was easy to fact, defaulted more. In Maddaloni and Peydró
finance a credit boom, especially for countries that (2011) we find that securitization also implied a
have suffered the crisis more (i.e., peripheral euro softening of lending conditions and standards in
zone), was financial globalization. This is clearly Europe and the USA.
the case for the euro area. In Kalemli-Ozcan, Pa-

16 17
 Monetary policy can also worsen the credit 4. Credit in bad times and the real
boom, both by making it bigger and riskier. In
Jiménez et al. (2012) we analyze the bank lending
costs of financial crises
channel of monetary policy transmission. By stud-
ying detailed loan-level and loan-application data, As we explain in detail in Freixas et al. (2014),
we find that, exactly as predicted by the theory, the real costs of systemic financial crises are signifi-
softer monetary conditions increase lending, espe- cant.13 As the financial system performs several im-
cially for banks with weak balance sheets in terms portant functions for the real sector, its impairment
of capital and liquidity. creates strong costs as, for example, the smooth
functioning of the payment system, risk-sharing,
But not only too low short term rates can and saving products. The main transmission chan-
expand credit supply, but also can increase the nel from the impairment of the financial sector is
risk-taking incentives of banks. This is what Adri- the disruption of the flow of funds from savers to
an and Shin call ‘the risk-taking channel of mone- firms for investment and to households for con-
tary policy’ in the last Handbook of Monetary Eco- sumption purposes; in particular, the credit crunch
nomics (Adrian and Shin, 2011).12 In Jiménez et al. is one of the most important channels that transmit
(2014) we study this channel. The very detailed the financial distress from banks to the real sector.
data on loan applications and outcomes in Spain
has allowed us to identify the effect of monetary Laeven and Valencia (2012) study the costs of
policy on banks’ risk-taking behavior. We separate banking crises since 1970. They find that, on av-
the changes in the composition of the supply of erage, the cost of the fiscal outlays committed to
credit from the concurrent changes in the volume the financial sector was 12.4% of GDP, with cases
of supply and quality and volume of demand. as high as 56.8% in 1997 in Indonesia. In terms
We employ a two-stage model that analyzes the of output loss, on average the cost was 30.1% of
granting of loan applications in the first stage and GDP. These numbers are huge. While public debt
loan outcomes for the applications granted in the does not increase in all banking crises, in those
second stage, and that controls for both observed countries where it indeed increased the average
and unobserved, time-varying, firm and bank het- increase was 26%. The heterogeneity is high: in
erogeneity through time x firm and time x bank Guinea-Bissau, public debt increased by 108.1% of
fixed effects. We find that a lower overnight inter- GDP, while Ireland (to show the case of a devel-
est rate induces lowly capitalized banks to grant oped country) it increased by 72.8% of GDP in the
more loan applications to ex-ante risky firms and last crisis. Table 1 shows the estimated costs for
to commit larger loan volumes with fewer collat- the recent banking crises in terms of output loss,
eral requirements to these firms, yet with a higher fiscal costs, the liquidity support, and the resulting
ex-post likelihood of default. A lower long-term increase in public debt. In the last column, it also
interest rate and other relevant macroeconomic shows whether the country experienced a credit
variables have no such effects. boom prior to the crisis.

Schularick and Taylor (2012) analyze financial

crises for the past 140 years.14 A general lesson is
that a speedy resolution of the crisis is crucial to
mitigate its costs, even if accompanied by high fis-

18 19
Table 1. Costs of the recent banking crises cal outlays to support the financial sector. A post-
ponement of such action will not only delay the
recovery but risks adding to the real costs by pro-

Fiscal Costs

public debt
Increase in
(% of GDP)
Liquidity
longing the credit crunch and the negative spirals

support
Output

Credit
boom
associated with slow growth and debt overhang.
Start
Country
loss
Turning to real effects, despite the much more ag-
Austria 2008 14.0 4.9 7.7 14.8 0 gressive policy response in the postwar period,
Belgium 2008 19.0 6.0 14.1 18.7 1 the cumulative real effects have been somewhat
Denmark 2008 36.0 3.1 11.4 24.9 0 stronger in the postwar period. In the aftermath of
France * 2008 23.0 1.0 7.4 17.3 0 postwar financial crises, output dropped a cumu-
Germany 2008 11.0 1.8 3.6 17.8 0 lative 6.2% relative to trend, and real investment
Greece 2008 43.0 27.3 42.3 44.5 1 by more than 22%. The pre-war output decline
Hungary * 2008 40.0 2.7 1.3 -0.3 1 effect, however, is largely an artifact of the massive
Iceland 2008 43.0 44.2 16.8 72.2 1 financial implosions of the 1930s. Excluding the
Ireland 2008 106.0 40.7 16.3 72.8 1 1930s, the cumulative real output and investment
Italy 2008 32.0 0.3 5.7 8.6 0
declines after crises were substantially smaller.
The finding of limited losses prior to the 1930s
Kazakhstan * 2008 0.0 3.7 5.0 9.1 0
would be consistent with the idea that financial
Latvia 2008 106.0 5.6 3.4 28.1 1
sectors played a less central role in the econo-
Luxembourg 2008 36.0 7.7 4.1 14.6 ...
my, and financial crises were thus less costly in
Mongolia 2008 0.0 4.2 9.4 -5.0 0
the earlier decades of the Schularick and Taylor
Netherlands 2008 23.0 12.7 3.7 26.8 0
(2012)’s sample.
Nigeria 2009 14.0 11.8 11.7 7.7 0
Portugal * 2008 37.0 0.0 16.7 33.6 0 Why are output losses so large today, despite
Russia * 2008 0.0 2.3 23.9 6.4 1 more activist policies? Governments and central
Slovenia * 2008 38.0 3.6 9.6 18.0 1 banks have tried since the 1930s to prevent neg-
Spain 2008 39.0 3.8 6.4 30.7 1 ative feedback loops in the economy and have
Sweden * 2008 25.0 0.7 13.0 11.1 0 sought to cushion the real and nominal impact of
Switzerland * 2008 0.0 1.1 3.0 -0.2 0 financial crises through policy activism. But, at the
Ukraine 2008 2.0 4.5 9.2 28.9 1 same time, the financial sector has grown in size
United Kingdom 2007 25.0 8.8 5.6 24.4 1 and increased leverage. As a result, the shocks hit-
United States 2007 31.0 4.5 4.7 23.6 0 ting the financial sector might now have a poten-
* Borderline cases tially larger impact on the real economy, without
Source: Laeven and Valencia (2012). Output loss is defined as the policy response. Moreover, explicit and implic-
the sum of the differences between trend real GDP and actual it government insurance and other public guar-
real GDP during the first three years of crisis, expressed as per-
centage of trend real GDP. The trend is computed by applying antees may have in turn contributed to the spec-
the HP filter to the real GDP series for the 20 years before the cri- tacular growth of finance and leverage within the
sis. Fiscal costs refer to the direct fiscal outlays due to the finan- system, creating excessive risk-taking incentives.
cial sector rescue packages as a percentage of GDP. Liquidity
support is defined as the ratio of central bank claims on deposit
money banks to total deposits and liabilities to non-residents. Reinhart and Rogoff (2009b) analyze the after-
Increase in public debt is measured as the difference between
pre- and post-crisis debt projections as a percentage of GDP. math of financial crises, which share three charac-
teristics. First, asset market collapses are deep and

20 21
prolonged; real housing price declines average more needed (because firm cash flows and thus
35% (over six years), while equity price collapses internal finance are low).
on average 55% (over three and a half years). Sec-
ond, the aftermath of banking crises is associated Nevertheless, from an empirical perspective, it
with profound declines in output and employ- is difficult to see whether there is a credit crunch
ment; the unemployment rate rises an average of 7 or not. When analyzing aggregate credit dynam-
percentage points over the down phase of the cy- ics, one cannot distinguish between supply and
cle (lasts on average over four years), while output demand. Even when comparing strong and weak
falls (from peak to trough) an average of over 9% banks (in terms of balance sheet strength) before
(averaging two years, which is significantly short- and during a crisis, one cannot be sure whether
er than for unemployment). Third, the real value the two banks are lending to the same type of
of government debt tends to explode, rising an firms (maybe weak banks are lending to riskier
average of 86% in the major post–1945 episodes. firms and thus the reduction in credit is entirely
Interestingly, a substantial part of debt explosions demand driven). Hence, one needs very detailed
is not due to the costs of bailing out and recapital- data to estimate the effects of a credit crunch. This
izing the banking system, but to the collapse in tax is what we do in Jiménez et al. (2012), where we
revenues that governments suffer in the wake of analyze the credit crunch in the 2008–2010 crisis
deep and prolonged output contractions, as well in Spain using a dataset consisting of loan applica-
as the countercyclical fiscal policies aimed at miti- tions and outcomes.
gating the downturn and the rise in automatic sta-
bilizers such as unemployment benefits (Laeven To achieve identification we focus on the set of
and Valencia, 2012). loan applications made in the same month by the
same borrower or for the same loan to different
Credit crunch banks of varying balance-sheet strengths (by in-
cluding in the specifications firm x month or alter-
A reduction in credit is an important negative natively loan fixed effects). Within this set of loan
spillover from financial crises to the economy at applications, for which the quality of potential
large. This does not imply that credit supply is borrowers is constant, we study how economic
being cut: in a recession, credit demand also goes conditions affect the granting of loans depending
down. Economic perspectives look worse, so firms on bank capital and liquidity. Moreover, we ana-
cut back investment decisions. Moreover, collater- lyze whether firms that get rejected in their initial
al also falls in value, which allows non-financial loan application can undo the resultant reduction
firms and households to borrow less — this is in credit availability by successfully applying to
known as the firm and household balance-sheet other banks.
channel.
We find that lower GDP growth reduces the
In order to have a credit crunch, then, we need probability that a loan application is granted, par-
banks to be reducing the supply of credit for rea- ticularly during crisis times. The negative effect on
sons unrelated to the borrowers. Hence, firms may loan granting is statistically stronger for banks with
still have a positive demand for credit but be un- low capital. We also find that firms that get reject-
able to obtain it. Credit crunches are very costly, ed in their initial loan application cannot undo the
since they happen precisely when bank funds are resultant reduction in credit availability by apply-

22 23
ing to other banks, especially in periods of tighter policy and the increased provision of long-term
economic conditions. Therefore, credit constraints refinancing).
seem to be binding for these firms during a cri-
sis. This shows, moreover, that heterogeneity of The transmission through the credit channel
financial conditions of banks matter for borrowers’ is identified using the responses of the euro area
decisions (on top of the aggregate banking condi- Bank Lending Survey (BLS) at the country level.
tions) and they have real consequences. Specifically, the different channels of transmission
are identified by looking at the factors affecting
But why is that other stronger banks do not the decision of banks to change lending condi-
compensate for the cut in credit from more affect- tions and standards for their borrowers. Factors
ed banks? In other words, why do healthy banks related to bank balance sheet capacity and com-
— if they exist during a crisis — not expand the petitive pressures identify the bank lending chan-
supply of credit? The main reason comes from nel, since the decisions to change these lending
the value of relationship lending: banks collect conditions apply to all borrowers independently
soft information on borrowers that allow them to of their credit quality. The factors linked to bor-
make valuable relationship loans. For example, rowers’ creditworthiness and net worth character-
Bae, Kang and Lim (2002) show that firms with ize the (borrower) balance sheet channel. Finally
closer relationships to their banks benefited from the BLS information on loan demand helps to fur-
easier access to credit from their banks during the ther isolate the credit demand channel.
Korean financial crisis of 1997. When a lending
relationship is cut, this soft information is lost, and The analysis suggests that the effect of the bank
so healthy banks cannot compensate this. Hence, lending channel has been partly mitigated espe-
problems in the banking system — both aggregate cially in 2010–2011 by the policy actions. By pro-
and heterogeneity — imply strong real effects.15 viding ample liquidity through the full allotment
policy and the longer-term refinancing operations
In Ciccarelli et al. (2013a and 2013b) we ana- (LTROs), the ECB was able to reduce the costs
lyze the credit supply effects on the real econo- arising to banks from the restrictions to private
my and the effect of the standard monetary policy liquidity funding by effectively substituting the in-
(changes in the overnight rate) during the crisis terbank market and inducing a softening of lend-
and gauge whether the functioning of the trans- ing conditions. At the same time, when looking at
mission mechanism is smooth across the euro the transmission through banks of different sizes,
area. We study how financial fragility of financial it seems that, until the end of 2011, the impact of
intermediaries and of borrowers (the non-financial credit frictions of borrowers has not been signifi-
sector) has affected the monetary policy transmis- cantly reduced, especially in distressed countries.
sion in the euro area, in particular through the Since small banks tend to lend primarily to small
credit channel, exploiting several dimensions of and medium enterprises (SMEs), we infer that the
heterogeneity, in a vector autoregression (VAR) policy framework until the end of 2011 might have
model estimated recursively over the sample been insufficient to reduce credit availability prob-
2002Q4–2011Q3 for a panel of 12 euro area coun- lems stemming from deteriorated firm net worth
tries. The model accounts also for the non-stand- and risk conditions, especially for small firms in
ard monetary policy measures implemented until countries under stress.
the end of 2011 (in particular the full allotment

24 25
 The previous analysis therefore supports the need it most. This is precisely what we find in Iyer
complementary actions that have been put in et al. (2014). We study the effect of the freeze in
place successively, and in particular those specif- the international interbank markets in August 2007
ically targeted at increasing credit to small firms on credit supply by Portuguese banks. With data
to reduce their external finance premia and cred- on all commercial and industrial loans in Portugal,
it rationing. In fact, the decision to enlarge the we find that the freeze caused a credit crunch:
collateral framework of the Eurosystem — in those banks that relied more on interbank borrow-
particular by accepting loans to SMEs as eligible ings cut lending more (compared to banks with
collateral — had the explicit objective of meeting less interbank funds) to the same borrower. More-
the demand for liquidity from banks in order to over, firms could not substitute easily the credit by
support lending to all type of firms. going to less affected banks. There was no credit
crunch for large firms, indicating that small, young
Using a similar dataset in Maddaloni and Pey- and entrepreneurial firms are the ones most at risk
dró (2013), we analyze the impact on lending in a financial crisis.
standards of short-term interest rates and mac-
ro-prudential policy before the 2008 crisis, and of Moreover, we find some evidence of zombie
the provision of central bank liquidity during the lending or loan evergreening, since banks cut
crisis. After the start of the 2008 crisis, we find that lending less to weaker firms in terms of loan cov-
low monetary rates helped to soften lending con- erage. Caballero et al. (2008) show that the zom-
ditions that were tightened because of bank capi- bie lending could have caused extremely negative
tal and liquidity constraints, especially for business effects in Japan, not only in the size but also in
loans. Importantly, this softening effect is stronger the time (a very long period) to overcome the fi-
for banks that borrow more long-term liquidity nancial crisis that started in 1990. Some even ar-
from the Eurosystem. Therefore, the results sug- gue that the recent policies in the Euro area are
gest that monetary policy rates and central bank keeping “zombie” banks artificially alive through
provision of long-term liquidity complement each a combination of regulatory forbearance and easy
other in working against a possible credit crunch lending by the Eurosystem of central banks.
for firms.
Bank failures also generate negative externali-
A key feature of the recent financial crisis was ties for other banks, i.e. contagion. These negative
the ‘excessive’ reliance on the market to obtain externalities associated with bank failures offer
liquidity. In a boom, banks tend to hold little li- the main rationale for financial regulation: to pre-
quidity since the opportunity cost is very high and vent socially costly bank failures (see Freixas et
there is plenty of liquidity in the market, so it is al., 2014). In Iyer and Peydró (2011), we study the
easy and fast to obtain it. However, by behaving effect of the failure of a big bank in India in 2001.
this way, banks create a negative externality: if the This failure was unrelated to economic fundamen-
banking sector is hit by a negative shock, many tals. With detailed data on interbank exposures,
of them will go to the interbank market to obtain we are able to see that banks more exposed to the
liquidity, creating a dry-up that will worsen the failed bank experienced higher deposit withdraw-
initial shock. By keeping too few liquid assets in als. Moreover, interbank linkages with other banks
their balance sheets, banks make it harder to ob- further propagated the shock. We also find real
tain liquidity for other banks precisely when they effects of this failure were significant, driving an

26 27
overall reduction in loan supply. In other words, their individual solvency, banks may be imposing
interbank contagion can have important real ef- negative externalities on the rest of the system,
fects in a credit crunch. thus decreasing the overall financial stability. Mi-
cro-prudential policy, hence, is not well equipped
An overhang of illiquid assets, often associated to deal with systemic risk. Financial regulation is
with banking crises, can also cause a credit market now becoming more macro-focused, focusing on
freeze. Banks that hold large quantities of illiquid the risks of the financial system as a whole, both
assets may trigger sales at fire sale prices when the build-up of financial imbalances, and the ex-
faced with negative liquidity shocks. Diamond and ternalities within the financial sector and from the
Rajan (2011) argue that, although the prospect of financial to the real sector. In other words, going
such fire sales depresses the bank’s current value, forward prudential regulation should also focus
banks may actually prefer to hold on to the illiquid on systemic risk (see Freixas et al., 2014).
assets because the bank’s survival is positively cor-
related with a recovery in asset prices. This creates Examples of this type of macroprudential pol-
high demand by banks for liquid assets, causing icy approach are the so-called countercyclical
banks to cut back on loans. Holmstrom and Tirole capital buffers. This new policy requires banks to
(1998) show that economies may suffer efficien- hold additional capital when aggregate credit is
cy losses when credit markets are disrupted such expanding fast. The intention is twofold: by re-
as during banking crises and are no longer able quiring this additional capital, banks will be more
to provide funds to entrepreneurs that are hit by capitalized when a recession comes, thus allevi-
liquidity shocks and need to raise funds to avoid ating the credit crunch problem. On top of that,
bankruptcy. Such bankruptcies cause a significant the additional capital may cool down the credit
loss in welfare. Consistent with this, Jiménez, Mian, expansion, thus lowering both the probability of a
et al. (2013) find that banks with higher propor- crisis and the cost of it if it occurs. Dynamic provi-
tion of illiquid assets such as real estate provide sioning, a regulation introduced in 2000 in Spain,
less credit supply to firms. is also an example of a macroprudential policy on
countercyclical capital requirements.
Macroprudential policy and credit
In Jiménez, Ongena, et al. (2013) we analyze
The 2008–09 global economic and financial the impact of countercyclical capital buffers held
crises have changed the consensus on how to by banks on the supply of credit to firms and their
conduct prudential regulation. Before it was ‘mi- subsequent performance, exploiting the Spanish
cro-oriented’, focused on ensuring the solvency dynamic provisioning. Spain introduced dynamic
of individual financial institutions and paying little provisioning unrelated to specific bank loan loss-
attention to the financial system as a whole. The es in 2000 and modified its formula parameters
consensus was that by ensuring adequate capital in 2005 and 2008. In each case, individual banks
ratios at the individual level, the whole system were impacted differently. The resultant bank-spe-
would be solvent. However, after a negative shock, cific shocks to capital buffers, coupled with com-
banks may try to increase their capital buffers by prehensive bank-, firm-, loan-, and loan appli-
decreasing their lending, which can create a credit cation-level data, allow us to identify its impact
crunch and fire sales that can in fact worsen the on the supply of credit and on real activity. Our
initial shock. In other words, by trying to increase estimates show that countercyclical dynamic pro-

28 29
visioning smoothes cycles in the supply of credit Notes
and in bad times upholds firm financing and per- (1) This opuscle is mainly based on my book with Xavier
formance.16 Freixas and Luc Laeven on systemic risk: Freixas, Laeven and
Peydró (2014) and on my papers on credit, liquidity and
systemic risk, in particular: Iyer, and Peydró (2011), Jiménez
The estimates are also economically relevant. et al. (2012), Jiménez et al. (2014), Jiménez, Ongena, Peydró
Firms borrowing from banks with a 1 percentage and Saurina (2013), Jiménez, Mian, Peydró and Saurina
point higher dynamic provision funds (over loans) (2013), Iyer et al. (2014), Maddaloni and Peydró (2011 and
2013), Ciccarelli, Maddaloni and Peydró (2013a and 2013b),
prior to the crisis get a 6 percentage points high- Kalemli-Ozcan, Papaioannou, and Peydró (2010 and 2013).
er credit growth, a 2.5 percentage points higher I thank an anonymous referee for very helpful comments
asset growth, a 2.7 percentage points higher em- and suggestions. I thank Francesc R. Tous for his excellent
research assistance. I also thank Antonio Ciccone and Michael
ployment growth, and a 1 percentage point higher Greenacre for helpful comments and suggestions.
likelihood of survival.
(2) Another important view is the preference channel, in
particular behavioral biases. See Freixas et al. (2014) and
Shleifer and Vishny (2010). It is important to stress that
5. Conclusions depending on which is the correct view of the determinants
of excessive risk-taking (preferences vs. agency channel) in
financial intermediaries, optimal prudential policy will be
different. For example, higher capital requirements would be
All in all, the evidence described in this opuscle positive by increasing buffers in a crisis under both channels,
shows that either looking at the history of financial but under the agency channel they may also reduce ex-ante
excessive risk-taking by making the financial intermediaries
crises or looking at the recent global crisis, ex-an- have more own skin in the game.
te credit growth (debt and leverage) is the main
(3) Despite the fact that, once the financial crisis occurs, these
ex-ante correlate in determining the likelihood of guarantees may reduce the systemic costs.
systemic financial crises, and conditioning on the
(4) See Pagano (2012), Stein (2013) and Freixas et al. (2014).
crisis, ex-ante credit growth increases the severity
of the financial crisis, with stronger negative real (5) See e.g. Schularick and Taylor (2012), Jordà, Schularick
effects. Important channels during the crisis are and Taylor (2013), and Freixas et al. (2014).
credit crunches, (wholesale) illiquidity, flight to (6) This does not mean that all demand-driven credit booms
quality and the so-called zombie lending. Finan- result from fundamentals. For example, they can be driven
by collateral values. Although there is an extensive literature
cial globalization and deregulation, financial inno- focusing on frictions on credit demand, it is not the focus of
vation (notably securitization), deficient corporate this opuscle.
governance and lack of market disciplining, mon- (7) Reinhart and Rogoff (2008, 2009a, 2011), Schularick and
etary and macroprudential policies affect credit Taylor (2012) and Jordà, Schularick and Taylor (2011 and
cycles. Both policymakers and academics should 2013). The initial part of this section is based on Gorton and
Metrick (2012) and Freixas, et al. (2014).
pay more attention to credit cycles, notably supply
driven. (8) See Freixas et al. (2014) for models of finance and macro
to analyze systemic risk.

(9) Of course, some credit booms can be financed by domestic

savings as we are seeing recently with China.

(10) As these banks were investing heavily in the real estate

bubble (lack of diversification and strong softening of lending
standards), risks were excessive from an ex-ante (and an ex-
post) view.

30 31
(11) Another contribution of this paper is to develop a References
technique to identify credit supply. See Freixas et al. (2014).
Acharya, V. V., and T. Yorulmazer (2007), “Too many to fail —
(12) See also Borio and Zhu (2012). An analysis of time-inconsistency in bank closure policies,”
Journal of Financial Intermediation, 16, 1–31.
(13) This section is based on Freixas et al. (2014).
Adrian, T., and H. S. Shin (2011), “Financial intermediaries and
(14) Bordo et al. (2001), analyzing a database spanning 120 monetary policy,” in Handbook of Monetary Economics, 3,
years of financial history, find that crisis frequency since 1973 601–650.
has doubled compared to the Bretton Woods and classical gold
standard periods, and is only related to the period of the 1920s Aiyar, S., C. W. Calomiris, and T. Wieladek (2012), “Does
and 1930s. macropru leak? Evidence from a UK policy experiment,” Bank
of England working paper 445.
(15) A large empirical literature has shown that financial
conditions have real consequences (Rajan and Zingales, Akin, O., J. M. Marin, and J.-L. Peydró (2013), “Anticipation of
1998; Levine, 2005). Bernanke (1983) and Calomiris and the financial crisis : Evidence from insider trading in banks,”
Mason (2003) find that the U.S. banking crisis during the mimeo.
Great Depression reduced the efficiency with which credit
was allocated, and that the resulting higher cost and reduced Allen, F., and D. Gale (2007), Understanding Financial Crises,
availability of credit reduced domestic output. Oxford: Oxford University Press.

(16) Aiyar, Calomiris, and Wieladek (2012) study the case Bae, K.-H., J.-K. Kang, and C.-W. Lim (2002), “The value of
of the UK, where time-varying capital requirements were in durable bank relationships: Evidence from Korean banking
place for all banks under the FSA jurisdiction. They find that shocks,” Journal of Financial Economics, 64, 181–214.
increases in capital requirements are followed by decreases
Bernanke, B. (1983), “Nonmonetary effects of the financial
in credit growth to the real economy. However, part of this
crisis in propagation of the Great Depression,” American
decrease is compensated by an increase in lending of those
Economic Review, 73, 257–76.
banks not falling under the FSA rules: foreign branches.
Therefore, those financial intermediaries not subject to Bernanke, B., and M. Gertler (1989), “Agency costs, net worth,
macroprudential regulation could reduce the effects of this and business fluctuations,” American Economic Review, 79,
policy. 14–31.

Bordo, M., B. Eichengreen, D. Klingebiel, and M. S. Martinez-

Peria (2001), “Is the crisis problem growing more severe?,”
Economic Policy, 16, 51–82.

Borio, C. and H. Zhu (2012), “Capital regulation, risk-taking

and monetary policy: A missing link in the transmission
mechanism?,” Journal of Financial Stability, 8, 236–251.

Caballero, R. J., T. Hoshi, and A. K. Kashyap (2008) “Zombie

lending and depressed restructuring in Japan,” American
Economic Review, 98, 1943–77.

Calomiris, C. W., and J. R. Mason (2003), “Consequences

of bank distress during the Great Depression,” American
Economic Review, 93, 937–47.

Ciccarelli, M., A. Maddaloni, and J.-L. Peydró (2013a),

“Heterogeneous transmission mechanism: Monetary policy
and financial fragility in the Euro Area,” Economic Policy, 28,
459–512.

Ciccarelli, M., A. Maddaloni, and J.-L. Peydró (2013b),

“Trusting the bankers: Another look at the credit channel of
monetary policy,” working paper.

Diamond, D., and R. Rajan (2006), “Money in a theory of

banking,” American Economic Review, 96, 30–53.
32 33
Diamond, D., and R. Rajan (2011), “Fear of fire sales, Jordà, O., M. Schularick, and A. M. Taylor (2011), “Financial
illiquidity seeking, and the credit freeze,” Quarterly Journal crises, credit booms, and external imbalances: 140 years of
of Economics, 126, 557–91. lessons,” IMF Economic Review, 59, 340–78.
Freixas X., L. Laeven and J.-L. Peydró (2014), Systemic Risk Jordà, O., M. Schularick, and A. M. Taylor (2013), “Sovereigns
and Macroprudential Policy, MIT Press, forthcoming. versus banks: Credit, crises, and consequences,” CEPR
Discussion Paper 9678.
Gorton, G., and A. Metrick (2012), “Getting up to speed on
the financial crisis: A one-weekend reader’s guide,” Journal Jordà, O., M. Schularick, and A. M. Taylor (2014), “When
of Economic Literature, 50, 128–50. credit bites back: Leverage, business cycles, and crises,”
Journal of Money, Credit and Banking, forthcoming.
Holmstrom, B., and J. Tirole (1997), “Financial intermediation,
loanable funds, and the real sector,” Quarterly Journal of Kalemli-Ozcan, S., E. Papaioannou, and J.-L. Peydró (2010),
Economics, 112, 663–691. “What lies beneath the Euro’s effect on financial integration?
Currency risk, legal harmonization, or trade?,” Journal of
Holmstrom, B., and J. Tirole (1998), “Private and public International Economics, 81, 75–88.
supply of liquidity,” Journal of Political Economy, 106, 1–40.
Kalemli-Ozcan, S., E. Papaioannou, and J.-L. Peydró (2013),
International Monetary Fund (2012), “Policies for “Financial regulation, financial globalization, and the
macrofinancial stability: How to deal with credit booms,” IMF synchronization of economic activity,” Journal of Finance, 68,
Staff Discussion Note 12/06. 1179–1228.
Iyer, R., and J.-L. Peydró (2011), “Interbank contagion at work: Kaminsky, G., and C. Reinhart (1999), “The twin crises: The
Evidence from a natural experiment,” Review of Financial causes of banking and balance of payments problems,”
Studies, 24, 1337–77. American Economic Review, 89, 473–500.
Iyer, R., J.-L. Peydró, S. da-Rocha-Lopes, and A. Schoar (2014), Kiyotaki, N., and J. Moore (1997), “Credit cycles,” Journal of
“The interbank liquidity crunch and the firm credit crunch: Political Economy, 105, 211–48.
Evidence from the 2007-09 crisis,” Review of Financial
Studies, forthcoming. Kindleberger, C. (1978), Manias, Panics, and Crashes: A
History of Financial Crises. First Edition, New York: Basic
Jeanne, O., and A. Korinek (2010), “Managing credit booms Books.
and busts: A Pigouvian taxation approach,” NBER Working
Paper 16377. Laeven, L., and F. Valencia (2008), “Systemic banking crises: A
new database,” IMF Working Paper 08/224.
Jiménez, G., and J. Saurina (2006), “Credit cycles, credit risk,
and prudential regulation,” International Journal of Central Laeven, L., and F. Valencia (2012), “Systemic banking crises
Banking, 2, 65–98. database: an update,” IMF Working Paper 12/163.
Jiménez, G., S. Ongena, J.-L. Peydró, and J. Saurina (2012), Levine, R. (2005), “Finance and growth: Theory and evidence,”
“Credit supply and monetary policy: Identifying the bank in Handbook of Economic Growth, 1A, 865–934.
balance-sheet channel with loan applications,” American
Economic Review, 102, 2301–26. Lorenzoni, G. (2008), “Inefficient credit booms,” Review of
Economic Studies, 75, 809–33.
Jiménez, G., S. Ongena, J.-L. Peydró, and J. Saurina (2014),
“Hazardous times for monetary policy: What do twenty-three Maddaloni, A., and J.-L. Peydró (2011), “Bank risk-taking,
million bank loans say about the effects of monetary policy securitization, supervision, and low interest rates: Evidence
on credit risk-taking?,” Econometrica, forthcoming. from the Euro-area and the US lending standards,” Review of
Financial Studies, 24, 2121–65.
Jiménez, G., A. Mian, J.-L. Peydró, and J. Saurina (2013),
“Local versus aggregate lending channels: The effects of Maddaloni, A., and J.-L. Peydró (2013), “Monetary policy,
securitization on corporate credit supply,” working paper. macroprudential policy, and banking stability: Evidence from
the Euro Area,” International Journal of Central Banking, 9,
Jiménez, G., S. Ongena, J.-L. Peydró, and J. Saurina (2013), 121–169.
“Macroprudential policy, countercyclical bank capital buffers
and credit supply: Evidence from the Spanish dynamic Pagano, M. (2012), “Finance: economic toxin or lifeblood?”
provisioning experiments,” working paper. mimeo.

34 35
Rajan, R. G. (1994), “Why bank credit policies fluctuate: A
theory and some evidence,” Quarterly Journal of Economics,
109, 399–441.

Rajan, R. G. (2005), “Has financial development made the

world riskier?,” NBER Working Paper 11728.

Rajan, R. G., and L. Zingales (1998), “Finance dependence and

growth,” American Economic Review, 88, 559–86.

Reinhart, C. M., and K. S. Rogoff (2008), “Is the 2007 US sub-

prime financial crisis so different? An international historical
comparison,” American Economic Review, 98, 339–44.

Reinhart, C. M., and K. S. Rogoff (2009a), This Time is

Different: Eight Centuries of Financial Folly, Princeton
University Press.

Reinhart, C. M., and K. S. Rogoff (2009b), “The aftermath of

financial crises,” American Economic Review, 99, 466–72.

Reinhart, C. M., and K. S. Rogoff (2011), “From financial crash

to debt crisis,” American Economic Review, 101, 1676–706.

Schularick, M., and A. M. Taylor (2012), “Credit booms gone

bust: Monetary policy, leverage cycles, and financial crises,
1870–2008,” American Economic Review, 102, 1029–61.

Shin, H. S. (2012), “Global banking glut and loan risk

premium,” Mundell-Fleming Lecture, IMF Economic Review,
60, 155–92.

Shleifer, A., and R. Vishny (2010), “Asset fire sales and credit
easing,” American Economic Review, 100, 46–50.

Stein, J. C. (2013), “Overheating in credit markets: Origins,

measurement, and policy responses,” speech delivered
at “Restoring household financial stability after the Great
Recession: Why household balance sheets matter”,
symposium sponsored by Federal Reserve Bank of St. Louis.

36
 Opuscles already published
1. Reconsidering Spanish Unemployment 19. Urban Sprawl: Causes and Consequences
Ramon Marimon (June 97) Diego Puga (January 2008)
2. Reducing Unemployment. At Any Cost? 20. Western European Long Term Growth,
Fabrizio Zilibotti (December 97) 1830-2000: Facts and Issues
Albert Carreras and Xavier Tafunell (June 2008)
3. Capital and Labor Taxes, Macroeconomic
Activity, and Redistribution 21. Overcoming Coordination Failure in Firms
Albert Marcet (November 98) and Organizations: Experimental Evidence
Jordi Brandts (March 2009)
4. The Lender of Last Resort in Today’s
Financial Environment 22. The Misallocation of Talent
Xavier Freixas (November 99) José V. Rodríguez Mora (May 2009)
5. Why does the Public Sector Grow? The 23. Complementarities in Innovation Strategy and
Role of Economic Development, Trade the Link to Science
and Democracy Bruno Cassiman (September 2009)
Carles Boix (November 99)
24. Simple Mechanisms to Resolve Conflicting
6. Gerontocracy and Social Security Interests and to Share the Gains
Xavier Sala-i-Martin (July 2000) David Pérez-Castrillo (November 2009)
7. The Political Viability of Labour Market Reform 25. Transfer of University Innovations
Gilles Saint-Paul (December 2000) Inés Macho-Stadler (January 2010)
8. Are EU Policies Fostering Growth and 26. Firing Costs, Dismissal Conflicts and Labour
Reducing Regional Inequalities? Market Outcomes
Fabio Canova (May 2001)
Maia Güell (June 2010)
9. Agglomeration Effects in Europe and the USA 27. Inequality and Tax Progressivity
Antonio Ciccone (September 2001)
Juan Carlos Conesa (October 2010)
10. Economic Polarization in the 28. Happiness Economics
Mediterranean Basin
Ada Ferrer-i-Carbonell (May 2011)
Joan Esteban (May 2002)
11. How do Households Invest their Wealth? 29. School Choice in Spain: Theory and Evidence
Miquel Faig (October 2002)
Caterina Calsamiglia (September 2011)

12. Macroeconomic and Distributional Effects 30. Corporate Social Responsibility

of Social Security and Social Welfare. How to Promote Socially
Luisa Fuster (April 2003) Responsible Business Strategies

Juan-José Ganuza (March 2012)
13. Educating Intuition: A Challenge for
the 21st Century 31. The Effectiveness of Foreign Aid:
Robin M. Hogarth (September 2003) from the Macroeconomic Perspective
14. Capital Controls in Post-War Europe to the Experimental Evaluation
José G. Montalvo / Marta Reynal-Querol (June 2012)
Hans-Joachim Voth (April 2004)
15. Taxation of Financial Intermediaries 32. Fiscal Policy in the European Monetary Union
Ramon Caminal (September 2004) Evi Pappa (September 2012)

16. Ready to Take Risks? Experimental 33. How Important are Capital Markets
Evidence on Risk Aversion and Attraction Imperfections in Determining Firm Decisions
Antoni Bosch-Domènech / Joaquim Silvestre i Benach and Aggregate Fluctuations?
(November 2005) Andrea Caggese (November 2012)
17. Social Networks and Labour Market Outcomes 34. Globalization, Technology and Inequality
Antoni Calvó-Armengol (January 2006) Gino Gancia (April 2013)
18. The Effects of Employment Protection in
35. Credit cycles and systemic risk
Europe and the USA José-Luis Peydró (December 2013)
Adriana D. Kugler (February 2007)
José Luis Peydró
José Luis Peydró is ICREA Professor and Catedràtic at
Universitat Pompeu Fabra, a Research Fellow at the CEPR,
and an Associate Editor for the Review of Finance, the
journal of the European Finance Association. He has been
consultant in 2012–14 for the International Monetary Fund,
the Federal Reserve Board, the European Central Bank, and
advisor in the Financial Stability department of Bank of
Spain. His research interests are on the intersection between
Finance and Macroeconomics, including systemic risk and
central bank policies. He has written a book on Systemic Risk
and Macroprudential Policy, forthcoming in MIT Press. He
has published his academic research in the leading journals
as Review of Financial Studies, Journal of Finance, American
Economic Review and Econometrica. José Luis has a PhD
in Finance from INSEAD and a Master in Economics from
CEMFI. He won the National Award of Bachelor Studies in
Economics (Premio Nacional) given by the Government of
Spain for the highest GPA in Spain in Economics, 1997.

Ramon Trias Fargas, 25-27 - 08005 Barcelona

Tel: 93 542 13 88 - Fax: 93 542 28 26
E-mail: [email protected]
http://www.crei.cat