Knowledge that Transforms

In this paper, we identify several shortcomings in the systemic-risk scoring methodology currently used to identify and regulate Systemically Important Financial Institutions (SIFIs). Using newly-disclosed regulatory data for 119 US and international banks, we show that the current scoring methodology severely distorts the allocation of regulatory capital among banks. We then propose and implement a methodology that corrects for these shortcomings and increases incentives for banks to reduce their risk contributions.

We examine the empirical relation between labor unions and firm indebtedness in the contemporary United States. Our identification strategy exploits two negative exogenous shocks in union power and the threat of unionization. Further, in the context of panel regressions, we develop a novel firm-level proxy for the bargaining power of labor using collective bargaining information from mandatory IRS filings from 1999 to 2013. Across a battery of tests, we document evidence in favor of a crowding-out hypothesis — namely, a substitution effect between labor power and financial leverage. Notably, this effect is more pronounced in firms in labor-intensive and unionized industries.

We develop a model in which bank culture improves upon outcomes attainable with incentive contracting. The bank designs a second-best incentive contract to induce the desired managerial effort allocation across growth and safety, but this induces excessive growth relative to the first best, a distortion exacerbated by interbank competition. Bank culture has two effects: it matches managers to banks with similar beliefs, and a safety-oriented culture reduces the competition-induced excessive growth focus. Culture is also contagious – a safety-oriented culture in some banks causes others to follow suit – this effect strengthens with higher bank capital and weakens with stronger safety nets.

Intermediation chains represent a common pattern of trade in over-the-counter markets. We study a classic problem impeding trade in these markets: an agent uses his market power to inefficiently screen a privately informed counterparty. We show that, generically, if efficient trade is implementable via any incentive-compatible mechanism, it is also implementable via a trading network that takes the form of a sufficiently long intermediation chain. We characterize information sets of intermediaries that ensure this striking result. Sparse trading networks featuring long intermediation chains might thus constitute an efficient market response to frictions, in which case no regulatory action is warranted.

This editorial reviews the papers that were presented at a conference at Washington University in St. Louis, a subset of which were published in a special issue of The Journal of Financial Intermediation. The papers cover a wide range of issues on how banks and financial markets have evolved since the financial crisis and the blurring of boundaries between institutions and markets.

We develop a model of financial intermediation wherein bank managers “reach for yield” – by overinvesting in risky assets and underinvesting in safer assets – provided they do not face much cost from liquidity shortfalls. The managers follow a pecking order in which their first preference is to invest in risky assets; their second preference is to hoard liquid assets; and their last preference is to invest in safer assets. This behavior is conducive to the formation of bubbles and “negative” bubbles in the market for risky and safer assets, respectively. Monetary loosening, by reducing the cost of liquidity shortfalls, induces further reach for yield and amplifies the bubbles.

Since the early 2000s, macroprudential policy has increasingly become part of the regulatory and supervisory framework. Likewise, the housing market has been at the center of the debate on systemic financial risk prevention. Among macroprudential tools, the purpose of the loan-to-value (LTV) ratio is to constrain mortgage loan creation. This paper is unique in that it analyzes the effectiveness of LTV on mortgage lending moderation using a large sample of more than 4000 banks from 46 countries. The analysis suggests mortgage loans have been successfully curbed in countries with a LTV policy. Size and non-performing loans are the two key characteristics to the effectiveness of LTV. When nonlinearities are considered, the average effect of LTV can be very large; however, it becomes much less effective with large banks and banks with bad loans. Our results suggest the inclusion of other macroprudential tools may have complementary effects to LTV, and for large size banks in particular.

This paper provides novel empirical evidence showing that foreign financial developments are a powerful predictor of domestic banking crises. Using a new data set for 38 advanced and emerging economies over 1970–2011, we show that credit growth in the rest of the world has a large positive effect on the probability of banking crises taking place at home, even when controlling for domestic credit growth. Our results suggest that this effect is larger for financially open economies, and is consistent with transmission via cross-border capital flows and market sentiment. Direct contagion from foreign crises plays an important role, but does not account for the whole effect.

In the UK, between 1955 and 1970, dual class shares quickly lost popularity without any regulatory intervention. The decline in the use of dual class shares was positively correlated with the relative valuations of one-share-one-vote and dual class firms, which in turn was related to media pessimism on the use of dual class shares. Following periods with high relative valuations of one-share-one-vote, one-share-one-vote firms exhibited lower returns than dual class firms suggesting that the latter were undervalued. These and other results suggest that investor demand may lead firms to abandon dual class shares.

We analyzed the loan guarantees that the Japanese government provided for banks’ loans to small and medium-sized enterprises (SMEs). We modeled and estimated how much and under what conditions loan guarantees affected banks’ risk-taking and banks’ non-guaranteed lending. In the presence of controls for bank capital and other factors that might affect supplies of bank credit, our estimates supported our model's implications that loan guarantees increased banks’ risk-taking. Consistent with our model, our estimates imply that, when banks initially had fewer guaranteed loans and then got more guaranteed loans, guaranteed loans were complements to, rather than substitutes for, non-guaranteed loans. As complements, loan guarantees could be “high-powered” in that they generated increases not only in guaranteed loans, but also increases in non-guaranteed loans that were a multiple of the increases in guaranteed loans. In addition, banks’ having more capital was associated with doing more non-guaranteed lending.