Banking on the Boom, Tripped by the Bust: Banks and the World War I Agricultural Price Shock

• Author: MATTHEW JAREMSKI,DAVID C. WHEELOCK
• Published date: 01 October 2020
• DOI: http://doi.org/10.1111/jmcb.12725
• Date: 01 October 2020

DOI: 10.1111/jmcb.12725

MATTHEW JAREMSKI

DAVID C. WHEELOCK

Banking on the Boom, Tripped by the Bust: Banks

and the World War I Agricultural Price Shock

How do banks respond to asset booms? This paper examines (i) how U.S.

banks responded to the WorldWar I farmland boom; (ii) the impact of regu-

lation; and (iii) how bank closures exacerbated the postwar bust. The boom

encouraged new bank formation and balance sheet expansion (especially by

new banks). Deposit insurance amplied the impact of rising crop prices

on bank portfolios, while higher minimum capital requirements dampened

the effects. Banks that responded most aggressively to the asset boom had a

higher probability of closing in the bust, and counties with more bank clo-

sures experienced larger declines in land prices.

JEL codes: E58, N21, N22

Keywords: Asset booms and busts, banks, bank lending, bank entry, bank

closure, deposit insurance, capital requirements, regulation

A     are often intertwined with

lending booms and busts. Although possibly triggered by a fundamental shock, ris-

ing asset prices can stimulate lending and increased leverage, which in turn causes

asset prices to rise further, generating more lending, and so on. Similarly,falling asset

prices can force debt contraction and deleveraging that reinforce the decline in asset

prices. Large declines in asset prices can be disruptive, especially when preceded by

rapid credit growth or involve highly leveraged assets such as real estate (see, e.g.,

Kindleberger 1978, Minsky 1986, Borio and Lowe 2002, Eichengreen and Mitchener

The views are those of the authors and not necessarily ofcial positions of the Federal ReserveBank of

St. Louis or the Federal Reserve System. The authors thank Lee Alston, Mark Carlson, Matteo Crosignani,

Bob DeYoung,Chris Hanes, two anonymous referees, and participants in the 2018 Cliometrics conference,

Central New YorkEconomic History Conference, Economic History Association Annual Meetings, and

NBER Summer Institute for comments.

M J is an associate professor in the Department of Economics and Finance

at Utah State University (E-mail: matthew.jaremski@usu.edu). D C. W is a group

vice president and deputy director of research at the Federal Reserve Bank of St. Louis (Email:

david.c.wheelock@stls.frb.org).

Received December 3, 2018; and accepted in revised form June 26, 2019.

Journal of Money, Credit and Banking, Vol. 52, No. 7 (October 2020)

© 2020 The Ohio State University

1720 :MONEY,CREDIT AND BANKING

2004, Reinhart and Rogoff 2009, Schularick and Taylor2012).1The interrelationship

between asset prices and lending booms thus raises important questions, including

how various regulations and policies might affect the vulnerability of the banking

system to asset price shocks, and how bank lending and instability can exacerbate

asset price movements.

Many studies have investigated these questions in the context of the U.S. house

price boom of the early-to-mid 2000s and nancial crisis of 2008–09. For example,

researchers have highlighted the outsized role of the “shadow”banking system in sup-

plying credit that fueled the boom (e.g., Loutskina and Strahan 2009, Mian and Su

2009), while others focused on credit demand (e.g., Dell’Ariccia, Igan, and Laeven

2012, Glaeser, Gottlieb, and Gyourko 2012). Still other studies have examined the

impact of the nancial crisis and bank distress on economic activity during and af-

ter the crisis (e.g., Campello, Graham and Harvey 2010, Ivashina and Scharfstein

2010, Chodorow-Reich 2014, Gertler and Gilchrist 2018). However, the complexity

of modern nancial systems, which include banks with international operations and

off-balance sheet activities as well as shadow banks and other forms of intermedia-

tion, can obscure fundamental relationships. Thus, detailed examinations of episodes

when such factors were not in play can be especially useful for identifying key re-

lationships between asset booms and the banking system, as well as the effects of

different policies that are hard to discern in more complex environments.

This paper studies the interplay of bank lending and asset prices in the boom-bust

cycle affecting U.S. agricultural land prices during and after World War I. Unlike

the recent U.S. house price boom, where the underlying shock or set of shocks that

triggered the boom has proved difcult to identify conclusively, the farmland boom

of the 1910s had a clearly identiable trigger.2The wartime collapse of European

agriculture drove commodity prices sharply higher and constituted an external de-

mand shock that sparked the boom in U.S. farmland prices. However, the boom was

short lived. European production bounced back quickly when the war ended, driving

down crop prices and land values in the United States, and initiating a wave of farm

foreclosures and bank failures in the early 1920s (Alston 1983, Alston, Grove and

Wheelock 1994).

The historical episode is particularly advantageous for studying the interrelation-

ships between lending and asset price booms and busts because bank lending at the

time was decidedly local. Federal law prohibited interstate branch banking, and most

states either prohibited or severely restricted branching within their borders. More-

over, with the automobile still in its infancy and paved roads almost nonexistent

in rural areas, it would have been impractical for most farmers to obtain services

from a bank located more than a few miles from their home. Thus, the balance sheet

1. Theoretical descriptions of how credit cycles can amplify real shocks include Rajan (1994),

Kiyotaki and Moore (1997), Geanakoplos (2010), and Nuˇ

no and Thomas (2017).

2. Studies attribute the U.S. housing boom to, among other things, loose monetary policy (Taylor

2010), a savings glut and heightened demand for safe U.S. nancial assets (Bernanke 2005), a bubble

(Case and Shiller 2003), and nancial market innovations including securitization of subprime mortgages

(Loutskina and Strahan 2009, Mian and Su 2009).

MATTHEWJAREMSKI AND DAVIDC. WHEELOCK :1721

information we observe for individual banks reects their lending to local farmers,

and we can approximate local income shocks using detailed information about crop

production in a bank’s county.3Specically, we calculate a county-specic farm out-

put price shock by applying the annual nation-wide price changes of 11 major crops

to the county output shares of each crop before the war. This provides exogenous

variation both across time and within a state to identify the effects of the price shock.

In two important studies of the episode, Rajan and Ramcharan (2015a, 2016) nd

that counties with more banks experienced larger increases in farmland prices and

mortgage debt during the boom and suffered larger price declines and more bank

failures during the bust. Whereas Rajan and Ramcharan examine the effects of credit

availability (as reected in the number of banks present in a county) and bank sus-

pensions on county-level land prices, here we use biennial balance sheet data for in-

dividual banks in 18 agricultural states for 1908–20 to examine how the price shock

affected (i) the establishment of new banks, (ii) bank portfolio decisions, and (iii) the

determinants of bank closures when farm prices and incomes ultimately collapsed.

The bank-level data enable us to test whether state banking policies amplied or

dampened the impact of asset price shocks on the banking system, and to use an

instrumental variables approach to identify the impact of bank closures on county-

level farmland values during the early 1920s. Our study, thus, provides insights about

the channels by which the asset boom-bust affected the banking system and supply

of credit, the role of government policies, and how banking instability contributed to

the collapse of asset prices when commodity prices began to fall.

Our results show that rising crop prices encouraged entry of new banks and balance

sheet expansion of new and previously established banks. While all banks expanded

their assets and loans, those established during the war were especially aggressive

lenders. New banks accounted for some one-third of the total increase in bank loans

in our sample states between 1914 and 1920, and many increased their portfolio risk

in response to rising crop prices. Similar to shadow banks in modern times, state-

chartered banks responded more strongly to the boom than did more tightly regulated

national banks. Moreover, we nd that deposit insurance amplied the effects of ris-

ing crop prices on bank loan volumes, whereas higher minimum capital requirements

deterred entry and dampened the effects of crop prices on loan growth and risk.

When farm prices and incomes collapsed in the early 1920s, banks that had ex-

panded more during the boom had a higher probability of closing. Recently estab-

lished banks were especially likely to close, as were banks with insured deposits,

higher leverage, or larger shares of their portfolios devoted to loans. Further, banks

located in counties with large increases in land values during the boom were more

likely to close when prices fell. Finally, we nd that bank closures exacerbated the

collapse of farmland values during 1920–25. Thus, our research provides new evi-

dence of how banks can both be affected by and contribute to asset price booms and

busts, and how banking policies can inuence the feedback loop around such events.

3. The county is a reasonable approximation of the area constituting a rural banking market at the time.

Most empirical studies dene banking markets at the MSA or rural county level even in modern times.