Bankers' conflicts of interest in the interwar years: Lessons for today’s regulators

Marc Flandreau, Norbert Gaillard, Ugo Panizza 18 April 2010



There are interesting parallels between the current crisis and the 1931-33 government debt debacles that followed the 1929 stock market crash.

  • Then, as now, bad behaviour in the financial industry was seen at the culprit of the crisis.
  • Then, as now, a number of financial intermediaries were accused of reaping large gains while offloading losses on third parties (investors then, the public sector now).
  • Then, as now, observers emphasised the danger of new financial instruments and of the “originate and distribute” model of finance.
  • Then, as now, banks were accused of knowingly bringing to the market unworthy government securities (think of Goldman Sachs and Greece today).
  • Then, as now, the crisis sparked strong political pressure for tighter financial regulation.

These parallels suggest that there is much that modern policymakers can learn from history (Calomiris 2008 and 2009).

New evidence on old issues

In a recent paper (Flandreau et al. 2010), we provide new evidence on the behaviour of the financial intermediaries that operated in the New York foreign government bond market in the 1920s.

The interwar debt crisis of 1931-1933 was a major event that led nearly 40% of international sovereign and sub-sovereign bonds to default (see Figure 1).

Figure 1. Share of defaulted bonds by year of issuance

The “originate and distribute” model in sovereign lending was held responsible for this failure. The claim that banks had proven unable to manage their conflicts of interest and alleged wrongdoings in the underwriting business of commercial and investment banks played an essential role in justifying the regulatory reform that followed the crisis, including the Glass-Steagall Act that separated commercial and investment banking.

On the basis of anecdotal evidence that drew heavily on the experience of the foreign bond market, US Senate Hearings held in the early 1930s (and known as the US Senate Committee on Finance Hearings and the famous Pecora Hearings that followed) concluded that commercial banks had incentives to underwrite low quality bonds.

In our work, we find that the experience of the 1920s is consistent with theoretical arguments about reputational concerns. In particular, we show that bonds underwritten by less prestigious houses (i.e., banks with smaller market shares) were more likely to default even after controlling for ex ante measures of bond quality (such as rating or spread at issue). This suggests that market leaders worried about retaining their prestige and hence were particularly careful in selecting the bonds they underwrote. Our results are also consistent with the notion that prestigious underwriters did not try to use their reputation to extract anomalous fees from borrowers.

We provide two distinct but complementary ways to think about intermediaries’ incentives.

  • First, we check whether bankers’ concerns over reputation played a role in determining the quality of the bonds they underwrote.
  • Second, we test for bad behaviour (“banksterism”) by exploiting information about underwriting fees charged by bankers. In order to conduct these tests we use a new dataset that covers the universe of bonds issued in New York during the 1920s by foreign sovereign, sub-sovereign, sovereign-owned, and sub-sovereign-owned entities.

We start by looking at whether our data support the idea that commercial banks and investment banks faced different sets of incentives and whether this difference in incentives led the former to less selective underwriting decisions. This claim has assumed renewed political relevance after Paul Volcker’s recent emphasis on the need to divorce commercial and investment banking once again. We find no support for the idea that commercial and investment banks behaved differently when they underwrote sovereign bonds.

Of course, the fact that we do not find any difference between commercial and investment banks does not necessarily mean that banks at large did not face conflicts of interest. We discuss this issue by looking at the relationship between reputation and conflicts of interest and relate our findings to modern theories of signalling and certification, which suggest that reputation interacts with market structure. We find that prestigious underwriters who might have been tempted to overprice low quality securities refrained from doing so, presumably to avoid damaging their reputation. Despite contemporary focus and outrage with allegedly anomalous underwriting fees, we find no evidence that banks misrepresented the securities they underwrote and sold them to the public at inflated prices while pocketing extra fees. In other words, we find no evidence of "banksterism" in the market for international sovereign bonds. As the case for banksterism becomes less convincing, the case for bad luck – meaning an unexpected shock or unpredicted policy errors when dealing with that shock – becomes more plausible.

The fact that we find no evidence of systematic misbehaviour on the part of investment and commercial banks does not necessarily mean that the regulatory reforms implemented in the 1930s were not warranted. However, it does mean (and it is an important finding) that the official motivations for them lacked empirical validity. We think that this is information to ponder at a time when regulatory responses to the subprime crisis are being considered.

Relevance to today?

A more fundamental question is whether the discipline at work during the 1920s also operates today. The irony of the historical experience is that, while we find no evidence of misbehaviour, contemporaries were convinced that banksterism was the main cause of the crisis. Regarding today’s crisis, some observers have expressed their doubt on the validity of arguments suggesting that concerns for reputation or the fear of losing market shares played a role in mitigating those conflicts of interest (see, for instance, Acemoglu 2009). If this is correct, future research ought to understand why size and reputation, which were once moderators of bad behaviour, no longer operate in the way they used to.


Acemoglu, Daron (2009), "The crisis of 2008: Structural lessons for and from economics", CEPR Policy Insight 28, December.

Calomiris, Charles W (2008), “The subprime turmoil: What’s old, what’s new, and what’s next”,, 22 August.

Calomiris, Charles W (2009), “The Subprime Turmoil: What’s Old, What’s New, and What’s Next”, Journal of Structured Finance, 15: 6-52, Spring.

Flandreau, Marc, Norbert Gaillard, Ugo Panizza (2010), “Conflicts of Interest, Reputation, and the Interwar Debt Crisis: Banksters or Bad Luck?”, HEID Working Paper No. 02/2010, February.



Topics:  Economic history Global crisis

Tags:  financial crises, Great Depression, conflicts of interest

Professor of Economic History at the University of Pennsylvania and CEPR Research Fellow

Economist and Independent Consultant

Professor of Economics and Pictet Chair at the Graduate Institute, Geneva; Research Fellow, CEPR