When to Lean against the Wind

• Author: PAUL WACHTEL,MORITZ SCHULARICK,BJÖRN RICHTER
• Published date: 01 February 2021
• DOI: http://doi.org/10.1111/jmcb.12701
• Date: 01 February 2021

DOI: 10.1111/jmcb.12701

BJÖRN RICHTER

MORITZ SCHULARICK

PAUL WACHTEL

When to Lean against the Wind

In this paper, we show that policymakerscan distinguish between good and

bad credit booms with high accuracy and they can do so in real time. Ev-

idence from 17 countries over nearly 150 years of modern nancial his-

tory shows that credit booms that are accompanied by house price booms

and a rising loan-to-deposit ratio are much more likely to end in a systemic

banking crisis than other credit booms. We evaluatethe predictive accuracy

for different classication models and show that characteristics observed in

real time contain valuable information for sorting the data into good and

bad booms.

JEL codes: E32, E44, G01, G21

Keywords: banking crises, crisis prediction, credit booms,

macroprudential policy

P    , credit booms, can have

diverse outcomes. First, there is a large literature that relates credit growth with

improved economic fundamentals and argues that such nancial deepening episodes

are economically benecial. Second, some credit booms end badly; they result in

nancial crises with severe output losses. This means that policymakers eager to

avoid the debilitating effects of banking crises have to walk a ne line between the

two pitfalls of failing to intervene to stop a bad boom and being too activist and

choking off economic growth. Measures to dampen credit booms may reduce the

This work is part of a larger project kindly supported by a research grant from the Bundesministerium

für Bildung und Forschung (BMBF). We are indebted to a large number of researchers who helped with

data on individual countries. Special thanks to the participants at the American Economic Association

meetings in Philadelphia January 2018, the Bank of Finland seminar in June 2018, the 23rd Dubrovnik

Economic Conference in June 2017, the Halle Workshop on Macroeconomics in May 2017, and the LSE

SRC Conference on Financial Crises in April 2018. We are grateful to John Ducca, Evan Kraft, Olivier

Jeanne, Philip Jung, and Alan M. Taylor for helpful comments. All errors are our own.

B R is an Assistant Professor at the Department of Economics and Business, Universitat

Pompeu Fabra and an Afliated Professor at Barcelona GSE (E-mail: bjorn.richter@upf.edu). M

S is a Professor of Economics at the Department of Economics, University of Bonn and CEPR

(E-mail: schularick@uni-bonn.de). P W is a Professorof Economics at Stern School of Busi-

ness, New YorkUniversity (E-mail: pwachtel@stern.nyu.edu).

Journal of Money, Credit and Banking, Vol. 53, No. 1 (February 2021)

© 2020 The Ohio State University

6:MONEY,CREDIT AND BANKING

risk of a banking crisis, but also reduce growth with uncertain costs for the economy

(Svensson 2017, Adrian and Liang 2018).

Policymakers face a problem whenever they observe a period of rapid credit

growth. That is, can they determine whether the boom underway will lead to a crisis?

In this paper, we identify credit boom episodes and demonstrate that policymakers

can distinguish good booms from bad ones and that they can do so with data available

in real time. We show that there are clear markers that policymakers can use to tell

apart good from bad credit booms with considerable accuracy. Financial variables

such as a deteriorating liquidity position (measured by the loan-to-deposit ratio) and

a house price boom are powerful classiers to sort credit boom observations into

malign and benign. Among the set of real economic variables, only changes in the

current-account-to-GDP balance turn out to be a helpful classier.

We arriveat this conclusion by studying long-run data for 17 advanced economies

from 1870 to 2016. We rely on economic and nancial data from the Macrohistory

Database (Jordà, Schularick, and Taylor 2016), as well as the systemic banking

crisis chronology contained therein, which is based on a large number of historical

sources as well as the crisis data set compiled by Laeven and Valencia (2018). We

use the new Hamilton (2018) lter to detrend the data and dene credit booms as

periods when the log of real private credit per capita exceeds its predicted value by

a country-specic threshold. We identify 113 credit boom episodes and 90 systemic

banking crises in our sample of advanced economies over the past 150 years.

Only a few papers have examined credit booms and their outcomes (Mendoza and

Terrones 2014, Dell’Ariccia et al. 2016, Gorton and Ordoñez 2020). Yet since the

time series examined are short and the country experiences are very heterogeneous,

these studies often face challenges to distinguish good booms from bad booms based

on observable characteristics.1Our long-run historical data have the advantage that

we can analyze within-country experiences as most of the sample countries have

experienced both good and bad credit booms at some point in their history.2

The growth of credit has been of interest to economic historians, development

economists, and students of macronance for at least 30 years. Our paper connects

two important and seemingly contradictory strands in the literature (Wachtel 2018).

On the one hand, there is a literature on the nance-growth nexus that associates

credit deepening and the quality of nancial intermediation with economic growth

(King and Levine 1993, Rancière, Tornell, and Westermann 2008). There is a volu-

minous literature that uses post-World War II panel data that have been surveyed by

Levine (2005). The evidence indicates that countries with deeper nancial markets,

a higher credit to GDP ratio, or larger stock market capitalization, experience more

rapid growth. Evidence for the positive effects of nancial booms is also examined

by Rousseau and Wachtel (2017) who use historical data for the period 1870–1929

1. Dell’Ariccia et al. (2016) conclude that most indicators that have been suggested in the literature

lose signicance once one conditions for the existence of a credit boom.

2. However, this comes at the cost of focusing on a sample of developed, largely industrial,

countries only.

BJÖRN RICHTER, MORITZ SCHULARICK, AND PAULWACHTEL :7

and afrm the positive growth effects of nancial deepening. However, Rousseau

and Wachtel (2009) indicate that positive growth effects of nancial deepening have

weakened since the mid-1980s which coincides with an increase in the incidence of

nancial crises.

On the other hand, there is an equally large literature that associates credit booms

with banking crises. Despite the potential benets of nancial deepening, many

credit booms end in often debilitating banking crises with severe effects on the real

economy (Schularick and Taylor 2012, Jordà, Schularick, and Taylor 2013, Mian and

Su 2016). More recently, Mian, Su, and Verner (2017) have shown that household

credit booms predict bad growth outcomes in more recent cross-country samples. In

short, while some credit booms are a precursor to banking distress and crisis, other

credit booms might represent nancial deepening or be the reaction to a positive

productivity shock.

There is also a large literature on crisis prediction that tries to identify crisis indica-

tors and early warning signals (Borio and Drehmann 2009, Detken et al. 2014, Adrian,

Covitz, and Liang 2015, Aldasoro, Borio, and Drehmann 2018, Kiley 2018). Since

the great nancial crisis there has been increased emphasis on macroprudential risks

and policy and indicators based on detailed nancial sector data that are now available

(Cerutti, Claessens, and Laeven 2017). Crisis prediction is a daunting task because at

any point in time, the probability of a nancial crisis occurring is quite small and un-

likely to attract the attention of policymakers.3The approach that we take here is very

different than the crisis prediction literature. It is motivated by the idea that once a

policymaker observes that a credit boom is underway and the economy is in the credit-

boom state, the possibility of crisis warrants attention. For this reason, we depart from

the crisis prediction literature and use the credit boom as our unit of observation. Our

interest in this paper is not to predict crises but instead to provide information that

would help policymakers determine whether an observed credit boom is likely to end

badly. To the best of our knowledge, ours is the rst paper to show that it is possi-

ble to identify markers that distinguish bad booms from good booms in real time, an

important prerequisite for the decision whether to intervene during a credit boom.

The challenge faced by the policymaker is whether a country in a credit boom can

use available information to determine whether prudential or other policies should be

used to deal with the risks of a credit boom turning into a crisis (Cerutti, Claessens,

and Laeven 2017, Adrian and Liang 2018). The prevailing opinion prior to the recent

global nancial crisis was that monetary policymakers should focus on growthand in-

ation and rely on micronancial regulation to maintain nancial stability (Bernanke

and Gertler 1999). Federal Reserve Board Chairman Alan Greenspan, commenting

on the possibility of a bubble bursting famously said that “the job of economic policy

makers [is] to mitigate the fallout when it occurs” (Greenspan 1999). Yet even before

the global nancial crisis, some economists, notably at the Bank for International Set-

tlements, suggested that systemic risks warranted the introduction of macroprudential

3. Svensson (2017) argues that the relationship between credit growth and crisis incidence is based

on so many complex interactions that it is impossible to identify a stable and consistent crisis predictor.