Better capitalized banks lend more and lend better

Many people seem to think that when authorities increase capital requirements, banks lend less. The advocates of this view go on to argue that, since credit is essential for economic growth, we should not impose overly tough constraints on banks. Put another way, a number of people believe that we have gone too far in making the financial system safe and the cost is lower growth and employment.

Treasury Secretary-designate Steven Mnuchin appears to share the view that financial regulation has restrained the supply of credit: in a recent interview, he is quoted as saying “The number one problem with Dodd-Frank is that it’s way too complicated and cuts back lending.” One interpretation of this is that Secretary-designate Mnuchin will support proposals like House Financial Services Chair Jeb Hensarling’s Financial CHOICE Act to allow banks to opt for a simple capital standard as an alternative to strict regulatory scrutiny.

Our reaction to this is three-fold. First, for most banks, which are very small and pose little threat to the financial system, a shift toward simpler capital requirements—so long as they are high enough—may be both effective and efficient; for the largest, most systemic intermediaries, higher capital requirements should still be accompanied by strict oversight. Second, we see no evidence that higher bank capital is associated with lower lending. In fact, quite the opposite. Third, given that the 2007-09 financial crisis was the result of too much borrowing—and that over-borrowing is a leading indicator of financial crises—it follows that not all reductions in lending are bad. We take each of these points in turn.

Last week we wrote about the Minneapolis Plan to End Too Big to Fail and its proposal to sharply increase equity in the 13 largest banks in the country. Specifically, the Minneapolis Plan calls for a pure leverage ratio—the ratio of common equity to total assets—of at least 15% and possibly as high as 24%. The current requirement is 6%, and the CHOICE Act would require only 10%. Unlike the CHOICE Act, the Minneapolis Plan also embraces important aspects of current regulation (including stress tests and living wills) to contain the systemic risks of the largest, most complex, and most interconnected banks, while simplifying regulatory compliance for small banks that do not pose a threat to the financial system. Our view is that the Minneapolis Plan provides a solid basis for legislation that would advance the public goal of making the financial system safe in a cost-effective way.

Important in this conclusion is our judgment that higher capital does not hamper the aggregate supply of credit. The alternative view appears to posit a choice between bank balance sheet shrinkage, on the one hand, and credit growth, on the other. In fact, there is no inconsistency between making banks safer and ensuring long-run growth of credit. To see why, recall that from the bank’s perspective, equity capital is one of the sources of funds while loans and securities acquisitions are uses of funds. That is, the former is a liability while the latter are assets. In theory, an increase in bank equity can be used to fund an increase in credit provision.

But that’s theory, what about experience? Here, the evidence is compelling: strong banks lend to healthy borrowers, weak banks don’t. We have written about this on several occasions. First, we noted that those countries with better capitalized banking systems in 2006, prior to the start of the crisis, experienced stronger lending growth during and after the crisis. That is, higher capital did not slow the economy. Second, we reported on research at the BIS establishing that better capitalized banks experience lower funding costs, higher growth of debt funding, and higher growth of lending volumes.

Further evidence comes from scholarly studies of Japan and Europe. Caballero, Hoshi, and Kashyap describe how, in the 1990s, regulatory forbearance delayed a thorough recapitalization of Japan’s banks for more than a decade. As a consequence, insolvent banks made loans to keep insolvent borrowers afloat. In their study of the impact of the ECB’s recent actions, Acharya et al. conclude that extremely accommodative monetary policy had a similar impact. That is, undercapitalized euro-area banks had an incentive to evergreen loans to “low-quality” firms.

These results rely on data from a range of countries. What happens in the United States when bank capitalization rises and falls? To answer this question at an aggregate level, we have plotted below bank credit (relative to GDP) on the vertical axis and bank capital (relative to assets) on the horizontal axis. The filled-red circle at the top right is the most recent observation from the third quarter of 2016.

Commercial bank credit (ratio to GDP) and equity capital (percent of bank assets), 1992-3Q 2016

Source: Federal Reserve Board, H.8 and Bureau of Economic Analysis, National Income and Product Accounts.

Source: Federal Reserve Board, H.8 and Bureau of Economic Analysis, National Income and Product Accounts.

The results are striking: rather than the downward-sloping relationship that critics of higher bank capitalization anticipate, the relationship is strongly positive. That is, greater reliance on equity funding is associated with an increase of credit! More specifically, in the United States since 1992 a one-percentage-point increase in equity relative to assets comes along with a six-percentage-point increase in commercial bank credit relative to GDP. 

Finally, there is the fact that decreases in lending are not necessarily bad. In fact, quite the opposite. That is what the studies from Japan and Europe highlight: loans to zombie firms made by weak banks reflect an inefficient allocation of savings and lead to slower economic growth. And we doubt that anyone wishes to see a return to the over-indebtedness of U.S. households that contributed to the vulnerability of the financial system in 2007.

As we have argued, given the experience of the financial crisis, we need to ensure that borrowers are able to pay, both to protect the financial system and to protect taxpayers. The latter concern applies to mortgage borrowing, which is now effectively guaranteed by the federal government through the government-sponsored enterprises (GSEs). It also applies to student loans, which (in addition to being guaranteed by the federal government) account for nearly half of consumer credit and create decades-long financial burdens that are virtually impossible to escape even through personal bankruptcy. And then there is payday lending, which, while frequently beneficial, can also be exploited to prey on the least well-off in our society and has been a focus of both the Department of Defense and the Consumer Finance Protection Bureau.

So, what has happened to U.S. indebtedness since the financial crisis? According to BIS data, aggregate credit to the U.S. private nonfinancial sector peaked at the height of the crisis (the third quarter of 2008) at 169.3% of GDP; the latest reading (first quarter of 2016) puts it at 150.1%. Virtually all of this drop is accounted for by the decline of credit to households—from 97.1% to 78.4%. And, since only a bit more than one fourth of the decline reflects a fall in bank credit, most of it arises from the behavior of nonbank intermediaries.

Data from the Federal Reserve Bank of New York highlight this changing mix of intermediation that, in our view, has made the financial system as a whole more resilient and less vulnerable since the crisis. The following chart depicts the evolution since 1960 of the liabilities of three important components of the financial system: commercial banks; broker-dealers and bank holding companies (BHCs); and shadow banks (net of their own holdings of other shadow banks’ liabilities). While the commercial banking system has grown since 2007, the other segments have shrunk. Specifically, over this nine-year period, shadow banking has plunged from 120% of GDP to 76% of GDP (that’s from a peak of $18.0 trillion to a current level of $13.4 trillion). A substantial portion of this correction occurred before the July 2010 enactment of the Dodd-Frank Act, reflecting the crisis-driven demise of the underlying business model of wholesale banking without deposit insurance or a lender of last resort.

U.S. intermediation: shadow banking and commercial banking (percent of GDP), 1960-2Q 2016

BHC Bank holding company. Source: Financial Accounts of the United States; Adrian, Tobias, Daniel Covitz, Nellie Liang (2013) Financial Stability Monitoring, Federal Reserve Bank of New York Staff Report 601. Updates courtesy of the FRBNY.

BHC Bank holding company. Source: Financial Accounts of the United States; Adrian, Tobias, Daniel Covitz, Nellie Liang (2013) Financial Stability Monitoring, Federal Reserve Bank of New York Staff Report 601. Updates courtesy of the FRBNY.

From this evidence, we conclude: (1) higher capital levels are associated with more lending, with better lending, or with both; and (2) to the extent that enhanced regulation hampers lending, it is often of the type that makes the financial system less safe and can leave taxpayers on the hook.

To be clear, we do see great scope for improving and simplifying U.S. financial regulation. Among other things, the system could use massive streamlining; the GSEs still need restructuring; living wills and the resolution mechanism should compel systemic intermediaries to self-insure (in order to avoid bailouts); government guarantees need to be properly priced; and the financial infrastructure could be more resilient. (For a broader list and discussion see here.) But, if we’re to make the financial system both safe and efficient, we need not only much higher capital requirements, but also a strict regulatory regime that makes credible the government’s (and legislators’) promise not to bail the most systemic intermediaries.