The recent financial crisis suggested an important connection between real estate investments and bank trouble, especially in the U.S. In fact, this is nothing new. Although financial crises can have multiple and varied causes, real estate investment booms are likely to bode particularly ill for the stability of the financial system. Indeed, in a new paper, I highlight the existence of a strong connection between mortgage lending and bank failure already during the 1930s Great Depression. And there are some important lessons for bank regulation.
The causes of the US Great Depression, which saw the failure of thousands of banks throughout the country, have been the subject of academic debate for decades. No doubt many questions will remain unanswered for a long time to come. But one city deserves our attention: Chicago. Chicago had to face some of the worst episodes of urban banking distress in the US, and it ended up with one of the highest urban bank failure rates in the country.
For a long time, historians have known that Chicago underwent a deep real estate boom in the 1920s, collapsing in a bust in the 1930s (see Figure 1). But no one had actually quantitatively tested the link between mortgage lending and the ensuing bank failures. In this paper, I hand-collected individual bank balance sheet data from 1923 all the way up to 1933 to analyse the link between several types of investments and subsequent bank failure.
Figure 1: Annual number of new building in Chicago.
Failures occurred in three or four major waves throughout 1930-1933, at a time when notions such as “too-big-to-fail” were close to inexistent. Depositors ran en masse to withdraw their money, without much state intervention.
Importantly, what my paper shows is that amidst these repeated banking panics, the banks that were more likely to survive were those that had invested less in mortgages in the decade preceding the crisis. The liquidity crises created a sudden need for cash, and those assets maturing in a long time, such as mortgages, were likely to run against banks’ attempts to meet depositors’ demands. Of course, banks could also sell securities and rediscount some of them at the central bank in exchange for cash. But the higher the proportion of mortgages a bank had on its balance sheet, the more likely it was to be unable to properly face the liquidity crisis. The central bank was warier about accepting those long-term loans as collateral, in part due to the uncertainty tied to their future quality.
Some banks had been more cautious in their total amount of real estate lending throughout the 1920s. Others had indulged in the boom without much regard for the possibility of a crisis. The more cautious banks were rewarded when the crisis came: they were likely to fail later, or not at all (see Figure 2).
Figure 2: Proportion of total assets invested in real estate loans, by four groups of banks (1931 failures, 1932 failures, 1933 failures, and survivors), 1923-1933.
The study does not ignore the possibility that low loan quality may have been at the root of the problem: after all, perhaps borrowers were simply unable to pay back their mortgages on time, leading banks to make losses on these loans through foreclosure (what is usually called credit risk). However, this possibility is unlikely because loans at the time were very small as a proportion of property values (only about 50 percent). Even if borrowers had been unable to pay back their mortgages, banks were unlikely to make any significant losses on these loans, and this even after a significant fall in land values. This fact allows me to isolate the special importance of mortgages’ long maturities as a primary source of weakness during the crisis.
What this suggests is a very simple idea: that banks that are better prepared to face a liquidity crisis, in particular by not overinvesting in long-maturity loans, are more likely to survive one. From a liquidity point of view they are healthier. Of course, deposit insurance and lender-of-last resort facilities were further developed precisely because it was deemed that some otherwise healthy banks were likely to fail unjustifiably if they had to face a run of some kind. Once those facilities were enhanced in the aftermath of the Great Depression, international regulators’ attention became increasingly focused on other sources of ill health, usually those emanating from credit risk (think in particular, the Basel 1 and Basel 2 Accords). Credit risk was deemed the chief possible cause of bank illness, causing possible capital impairment, while liquidity risk was deemed minor, thanks to those facilities.
The problem with this view is that liquidity risk is still with us, and is likely to remain so for some time. Today, securitization, including mortgage securitization, has increased liquidity in the system, and central banks have played an important role in alleviating distress in the recent crisis. Reckless predatory lending to subprime borrowers caused banks to incur significant losses, implying that credit risk was indeed a major component of the crisis.
Nevertheless, liquidity risk was far from inexistent. It dramatically increased in the aftermath of Lehman’s failure, and although people did not witness any visible runs on banks’ deposits, thanks to deposit insurance, there was an invisible run on the uninsured part of banks’ liabilities, documented at length by a number of scholars. During this run, the long maturity of loans underlying mortgage-backed securities was an increased source of liquidity risk.
Because central banks’ job is still a very difficult one, due to uncertainty regarding the future quality of banks’ collateral, liquidity risk will never disappear completely. For this reason, a bank that is not well prepared to face this risk cannot be considered healthy, even if its credit risk is very low. This is perhaps an unusual claim: that banks should be prepared for an out-of-equilibrium event. The question is of course: to what extent? One could not ignore the essential trade-off between low liquidity risk and a bank’s fundamental role in the economy, which is to provide loans, including long-maturity ones. But nor can we ignore the dangers of liquidity risk – as the Great Depression shows, these can be very serious indeed.
Certainly some balance needs to be struck, and in light of this study’s findings, the more recent international banking regulation’s focus on liquidity risk is welcome.
- This article is based on the paper What Caused Chicago Bank Failures in the Great Depression? A Look at the 1920s,
- The post gives the views of its authors, not the position of LSE Business Review or the London School of Economics.
- Featured image credit: Depositors clamor to withdraw their savings from a bank in Berlin, 13 July 1931, Bundesarchiv, Bild 102-12023, Wikimedia Commons
- Before commenting, please read our Comment Policy
Natacha Postel-Vinay is CAGE Postdoctoral Fellow at the University of Warwick, soon-to-be Assistant Professor in Economic History at the LSE. She is an economic and financial historian, and won in 2012 the Economic History Society New Researcher Prize. Her research focuses on the history of financial panics, banking crises, and international banking regulation.