Stealing Deposits: Deposit Insurance, Risk‐Taking, and the Removal of Market Discipline in Early 20th‐Century Banks

• Published date: 01 April 2019
• Author: CHARLES W. CALOMIRIS,MATTHEW JAREMSKI
• Date: 01 April 2019
• DOI: http://doi.org/10.1111/jofi.12753

THE JOURNAL OF FINANCE •VOL. LXXIV, NO. 2 •APRIL 2019

Stealing Deposits: Deposit Insurance,

Risk-Taking, and the Removal of Market

Discipline in Early 20th-Century Banks

CHARLES W. CALOMIRIS and MATTHEW JAREMSKI∗

ABSTRACT

Deposit insurance reduces liquidity risk but can increase insolvency risk by encour-

aging reckless behavior. Several U.S. states installed deposit insurance laws before

the creation of the Federal Deposit Insurance Corporation, and those laws applied

only to some depository institutions within those states. These experiments present a

unique testing ground for investigating the effect of deposit insurance. We show that

deposit insurance removed market discipline constraining uninsured banks. Taking

advantage of World WarI’s rise in world agricultural prices, insured banks increased

their insolvency risk and competed aggressively for deposits. When prices fell after

the war, the insurance systems collapsed and suffered high losses.

DEPOSIT INSURANCE SPREAD THROUGHOUT the world in the latter half of the 20th

century as a result of external and internal political pressures favoring its

adoption (Demirg ¨

uc¸-Kunt, Kane, and Laeven (2008)). The International Mone-

tary Fund, the European Union, and the World Bank have all endorsed deposit

insurance. Despite its overwhelming political support, however, a large em-

pirical literature suggests that the moral-hazard costs of deposit insurance

have outweighed its liquidity-risk-reduction benefits.1These papers show that

deposit insurance is among the most important contributors to the unprece-

dented wave of banking crises that has washed over the world during the past

∗Charles W. Calomiris is with Columbia University and the National Bureau of Economic Re-

search (NBER). Matthew Jaremski is with Utah State University and NBER. The authors would

like to thank the Editor, two referees, Gerard Caprio, and David Wheelock, as well as seminar

participants at Columbia University, University of California–Merced, the Federal Reserve Bank

of Dallas, the Workshop in Macroeconomic Research at Liberal Arts Colleges, the Economic His-

tory Association, and the NBER Development of the American Economy meetings for valuable

comments and suggestions. The authors have read the Journal of Finance’s disclosure policy and

have no conflicts of interest to report. They received no funding support for this article, and have no

paid or unpaid positions in any relevant entity whose policy positions, goals, or financial interests

relate to this article. No party other than the two authors has any right to review this article prior

to its circulation.

1See Brewer (1995), Caprio and Klingebiel (1996), Martinez-Peria and Schmukler (2001),

Calomiris and Powell (2001), Demirg¨

uc¸-Kunt and Detragiache (2002), Honohan and Klingebiel

(2003), Demirg¨

uc¸-Kunt and Huizinga (2004), Cull, Senbet, and Sorge (2005), Barth, Caprio, and

Levine (2006), Demirg¨

uc¸-Kunt, Kane, and Laeven (2008), Beck and Laeven (2008), Laeven and

Valencia (2013), Yan et al. (2014), and Calomiris and Chen (2018).

DOI: 10.1111/jofi.12753

711

712 The Journal of Finance R

four decades. The separation between policy recommendations and economic

studies begs the question of whether empirical studies have failed to properly

control for the other influences that have contributed to both the rise of deposit

insurance and banking instability.

Most studies of deposit insurance are based on cross-country comparisons

or comparisons over time within countries that contrast the behavior of in-

sured banking systems with uninsured banking systems.2Despite attempts by

authors to control for factors that coincide with the creation or expansion of de-

posit insurance through explicit controls or through instruments that explain

the creation of deposit insurance, it is conceivable that some of the positive

association between deposit insurance and increased bank risk may reflect ex-

ogenous increases in risk that encourage the passage of deposit insurance. If

true, then the risk-creating effects of deposit insurance would be exaggerated.

In this study, we examine a near-ideal environment from the standpoint

of identification—the state deposit insurance experiments of the early 20th

century in the United States.3These systems installed deposit insurance for

unit state-chartered commercial banks that operated in parallel to the unin-

sured system of national banks (i.e., unit banks that were chartered by the

Comptroller of the Currency) within a given state and in parallel to uninsured

state and national banks operating in bordering nondeposit insurance states.

Mitigating the omitted variable problem embodied in cross-country studies,

this period allows us to study insured and uninsured depository institutions

operating at the same time and in the same place under the same legal sys-

tem and currency. Using detailed information about banks’ locations, economic

environments, and balance sheet characteristics, we implement a difference-

in-difference-in-difference model that measures the effect of deposit insurance

on insured banks controlling for both the change in uninsured banks in deposit

insurance states and the change in uninsured banks in other states. Moreover,

because several of the laws were passed in the same year but implemented

in different subsequent years, we are able to use placebo tests to determine

2A few exceptions exist. Brewer (1995) achieves identification by comparing the behavior of

institutions that had suffered large losses to those that had not. Yan et al. (2014) contrast insti-

tutions within Australia that were differentially affected by deposit insurance protection. These

studies likely suffer less than others from possible endogeneity bias in identifying the effects of

deposit insurance.

3A recent working paper by Aldunate (2015) uses a similar difference-in-difference-in-difference

approach to ours but with a smaller sample of banks. Aldunate compares the growth of various bank

balance sheet measures (e.g., deposits, loans to assets, investments to assets, cash and due from

banks to assets, and capital to assets) at the bank level within three state pairs: Nebraska/Colorado,

South Dakota/Minnesota, and Mississippi/Alabama. For each pair, the analysis focuses on the year

before deposit insurance was installed and the year after. The study finds that insured banks saw

greater deposit growth but it finds no effect on risk-taking or failure rates. Based on our results

from a larger period and additional states, the small sample could be biasing the results. Also, the

choice of states means that the years when deposit insurance is in place are concentrated during

the World War I agricultural price spike, preventing the analysis from clearly separating the

effect of deposit insurance from the region-wide changes in state bank growth due to World WarI.

Our study differs from Aldunate in numerous other ways, including how we construct uninsured

comparison states and our focus on the timing of deposit insurance’s effect.

Stealing Deposits 713

whether a region-specific economic shock was responsible for changes in banks’

and depositors’ behavior in addition to the passage of deposit insurance, or

whether changes in behavior were the consequence of deposit insurance.

Our findings not only corroborate prior literature on the moral-hazard con-

sequences of deposit insurance, but also show how the introduction of deposit

insurance created systemic risk. We find strong evidence that deposit insurance

caused risk to increase in the banking system by removing the market disci-

pline that had been constraining uninsured banks’ decision making. Depositors

applied strict market discipline on uninsured banks when evaluating whether

to place deposits in those banks, but put relatively little weight on the financial

soundness of insured banks. Insured banks were thus able to use the promise of

insurance to compete away deposits from uninsured banks. Because they were

constrained only by regulatory standards (i.e., a minimum capital-to-deposits

ratio, a minimum reserves-to-deposit ratio, and in some cases, a maximum

interest rate paid on deposits), which often proved inadequate to prevent insol-

vency, insured banks raised their loans, reduced their cash reserves, and kept

their capital ratios close to the regulatory minimum.

Insured banks seemingly bet on the permanence of the agricultural price

increases that occurred during World War I (WWI), and depositors seemingly

believed in the insurance systems’ ability to protect them. Deposits flowed most

strongly into insured banks located in counties where the price rises had the

biggest effect. As insured banks most often used those deposits to fund new

loans, the implementation of deposit insurance allowed an asset price bubble

in farm land to quickly form. When prices reversed in the early 1920s, the

insured banking systems collapsed, leaving depositors with losses.

The rest of the paper proceeds as follows. Section Ireviews the details of the

early 20th-century state-deposit insurance systems and summarizes the aggre-

gate data on the changing allocation of deposits that accompanied the passage

of deposit insurance. Section II describes our data set. Section III reports our

findings on the effects of deposit insurance on insured and uninsured banks’ de-

posits, loans, cash-to-assets ratio, and leverage. Section IV examines whether

deposit insurance led to the relaxation of market discipline and changed the

nature of competition in the deposit market. Section Vdiscusses the collapses

of the insured banking systems during the 1920s. Section VI concludes.

I. State Deposit Insurance Schemes of the Early 20th Century

The United States was a bank liability insurance innovator, installing the

only deposit insurance systems for nearly 150 years. The unique propensity for

liability insurance in the United States reflected political influences that also

produced the nation’s unique single-office (“unit banking”) system (for a review

of the history of liability insurance, see Section Iof the Internet Appendix,

which may be found in the online version of this article at the publisher’s

website). While several states installed liability insurance funds during the

antebellum period, none lasted beyond the Civil War and it was not until

the early 1900s that another wave of laws passed. During that wave, eight