What can policymakers do to prevent future financial crises? An emerging consensus holds that so-called macroprudential regulation is key: policies that aim to mitigate risks to the financial system as a whole. In a recent paper, Karsten Müller of Princeton shows that such policies were systematically loosened in the run-up to two-hundred seventeen elections across 58 countries. This raises the question of whether regulators can, in practice, withstand political pressures.
In the brave new world of financial regulation after the Great Financial Crisis of 2007–8, a majority of academics and policymakers have put their hopes in a new type of policy: macroprudential regulation (or macropru). While regulators recklessly relied on supervising institutions one-by-one, the argument goes, they missed the bigger picture: an economy-wide build-up of debt that threatened to bring down the financial system as a whole.
To prevent future financial crises, or at least their most severe consequences, policymakers at central banks or other regulators have increasingly been entrusted with the ability to more or less directly restrict bank lending. But politicians have every incentive to pressure regulators into shaping policies in their favor: politicians benefit at the voting booth when credit flows freely, and can face trouble when it does not. This is particularly true because macroprudential policy often directly targets households—who make up the majority of new lending today in both advanced economies and the rest of the world.
Examples of suspicious coincidences abound. In a systematic review of case studies spanning over 300 years, Jihad Dagher documents that procyclical regulation appears to be a recurring theme before financial crises. In Germany, the parliament essentially blocked the introduction of income-based limits for borrowers just eight weeks before the highly competitive 2017 elections—despite clear recommendations from the German Council of Economic Experts. But does this pattern of hands-off regulation around close elections also hold true in the data?
In a recent paper, I investigate changes to prudential regulations around 217 elections across 58 countries from 2000 to 2014. Figure 1 reveals a striking pattern. When upcoming elections are expected to be close, (macro)prudential regulations are systematically more likely to be loosened in the quarters directly preceding them (Panel A). Sector-specific capital buffers, which usually target households, are substantially less likely to be tightened prior to close elections (Panel B). These effects disappear when upcoming elections are not expected to be close—which effectively eliminates the incentive of incumbent politicians to interfere with the credit market.
Financial regulators are considerably less likely to tighten policies in the run-up to elections, particularly when these elections are expected to be close.
Could these correlations be driven by differences in economic fundamentals around election periods? I argue that this is unlikely to be the case for two reasons. First, pre-election and all other quarters do not differ significantly in a wide range of banking sector and macroeconomic variables in my sample—in either purely statistical or quantitative terms. Second, controlling for these variables (including their future and past values), if anything, strengthens the correlations I find. The results even persist when comparing quarterly changes to regulation within the same country in the same year.
This electoral cycle in regulation is more pronounced when economic fundamentals look rosy—exactly the situations in which financial regulators are supposed to “lean against the wind.” More precisely, sector-specific capital requirements are less likely to be tightened when economic growth is higher (and projected to be higher in the future) and banks are more profitable. Regulation is also less likely to tighten when the credit-to-GDP gap is growing. This finding is particularly troubling because the credit-to-GDP gap is arguably the most widely used metric to gauge the buildup of financial vulnerabilities, which suggests that the electoral cycle makes regulation more procyclical.
Central bank independence weakens electoral cycles in monetary policy, but not in macroprudential regulation.
An important reason why central bankers and other regulators have often downplayed political limitations to macroprudential policies is that central bank independence appears to shield countries from election cycles in monetary policy, and thus inflation. But directly restricting voters’ ability to purchase a home, for example, might be a much more politically sensitive tool than changing the central bank’s policy rate by 25 basis points. Indeed, I find no evidence that a country’s degree of central bank independence matters for electoral cycles in prudential regulation. Central bank independence, however, does matter for cycles in monetary policy. Taken at face value, this suggests that macroprudential policy may be too blunt of a tool to be insulated by independent institutions.
Overall, my findings support the idea that political interference in regulatory decisions may be a serious limitation to the post-crisis consensus on financial regulation. While my study purely leverages variation across countries, the evidence I present is consistent with a wealth of country-level studies on electoral cycles that use microeconomic data. It may be high time for regulators to take seriously the political economy of finance.
Karsten Müller is a Ph.D. candidate at Warwick Business School. Starting in fall 2018, he will join Princeton University as a Postdoctoral Research Associate (www.karstenmueller.eu). You can contact him at email@example.com.
For more on the financial crisis, listen to the “Ten Years Later Pt 2: The Aftermath” episode of Capitalisn’t:
Disclaimer: The ProMarket blog is dedicated to discussing how competition tends to be subverted by special interests. The posts represent the opinions of their writers, not necessarily those of the University of Chicago, the Booth School of Business, or its faculty. For more information, please visit ProMarket Blog Policy.