Descriptions of financial frenzies suggest that lenders abandon caution in the midst of a boom and become more aggressive (or careless) in their lending (e.g. Aliber and Kindleberger 2015, Minsky 2008). Such descriptions have two distinctive elements. First, credit quality deteriorates in the boom as lenders search for risk and it improves in the subsequent bust. Second, it deteriorates more when there are more lenders competing for business – the proverbial ‘madness of crowds’. A number of studies (e.g. Mian and Sufi 2009, Madalloni and Peydro 2010, Gianetti and Laeven 2012, Lisowski et al. 2017) demonstrate the cyclicality of credit standards. However, the degree to which competition amongst lenders interacts with cyclicality is relatively unexplored. In Granja et al. (2018),
Granja, Leuz, Rajan considers the following as important:
This could be interesting, too:
Chen, Los, Timmer writes The rise of intangible income: A global value chain perspective
Laeven, McAdam, Popov writes Labour market flexibility during financial crises
Amodio, Di Maio writes How conflict disrupts the economy
Bian, He, Shue, Zhou writes Leverage-induced fire sales and stock market crashes
Descriptions of financial frenzies suggest that lenders abandon caution in the midst of a boom and become more aggressive (or careless) in their lending (e.g. Aliber and Kindleberger 2015, Minsky 2008). Such descriptions have two distinctive elements. First, credit quality deteriorates in the boom as lenders search for risk and it improves in the subsequent bust. Second, it deteriorates more when there are more lenders competing for business – the proverbial ‘madness of crowds’. A number of studies (e.g. Mian and Sufi 2009, Madalloni and Peydro 2010, Gianetti and Laeven 2012, Lisowski et al. 2017) demonstrate the cyclicality of credit standards. However, the degree to which competition amongst lenders interacts with cyclicality is relatively unexplored. In Granja et al. (2018), we get at this issue using a novel measure of the extent to which lenders are willing to stretch their lending portfolio – specifically, the average physical distance of their borrowers from their branches.
The ability to produce information about hard-to-evaluate credits has historically been based on close interactions between bankers and potential borrowers. The firmness of a borrower’s handshake, the cleanliness of her premises, or her punctuality in meetings might all reveal valuable information about the likelihood of repayment. Petersen and Rajan (2002) showed, however, that the adoption of information and credit scoring technologies in the 1980s and 1990s brought fundamental changes to the business models of banks. Slowly, but steadily, information technologies allowed lenders to substitute somewhat for local interactions in lending to small businesses. The average distance between banks and their borrowers increased steadily as communication technologies improved.
Yet, at any point in time, available communication technologies determine the limits of the area within which a bank can lend safely. If a bank stretches to lend beyond this, it will screen and monitor the borrower less effectively, thus taking on more credit risk. In other words, a faster-than-trend expansion of the average distance at which a bank lends is either evidence of a rapid improvement of information technology or suggestive of increased bank risk taking. It is relatively easy to tell the effects of improvements in technology from increased risk taking when a lending boom is followed by a bust. If it is the former, the average distance of loans made should not differ from the trend over the business cycle, even if the trend becomes steeper. If the rapid increase in average distance in the boom reflects risk taking, it should be followed by a rapid drop in average distance in the bust as banks become more conservative in lending. We should also see that more distant loans are associated with higher default rates, especially during the boom.
We use two datasets that, when combined, offer information on the locations of borrowers and respective lenders of most small business loans originated in the US over the last two decades. We find a significant cyclical component in the evolution of lending distances. Distances widen considerably when credit conditions are lax and also shorten considerably when credit conditions become tighter. Between 2004 and 2007, average distances approximately doubled from 175 miles to 350 miles. These distances, however, quickly slipped back to approximately 200 miles following the Global Crisis (see Figure 1). This cyclicality also holds when we look at the median, lower and upper deciles of the distribution of distance. We also confirm that it is not driven by a few large banks but can be seen across different size classes.
Figure 1 Evolution of average lending distance over time
Next, we empirically analyse whether distant lending in the boom is, on average, riskier. We find that the higher the post-crisis average non-performing loan ratio of the bank, the more cyclical is its pattern in lending distance, suggesting that it was risky to go the extra mile during the boom. It would be even more persuasive, however, to show that more distant loans which originated during the credit cycle boom defaulted more often. Towards this end, we use the Small Business Administration loan-level dataset of government-guaranteed loans, which contains information on ex-post defaults (also termed charge-offs). We find that a 1% increase in lending distance in 2006 and 2007 is associated with an increase in the likelihood of charge-off that is between two and three times larger than that of a similar increase in lending distances in 2003 (see Figure 2).
Figure 2 Distance and likelihood of charge-off over the business cycle
Having established a cyclical pattern of risk taking, with distance being a good proxy for that risk, we turn to the conditions under which risk taking behaviour emerges. We predict that banks whose branches are primarily in competitive banking markets see a more pronounced cyclical pattern in average lending distance. Since such banks likely look for borrowers in less competitive areas, we should find a similar cyclical pattern in average borrowing distance for borrowers located in less competitive areas. Finally, distant loans made from a competitive area to a less competitive area should also have a cyclical pattern. We find evidence consistent with these predictions, when we measure competition as the Herfindahl-Hirschman index (HHI) for bank loans made in the county of interest. Specifically, we plot the average lending distances at the bank level for banks below and above the median HHI in their home and destination. We find that the lending distances of banks exposed to below-median concentration in their home markets and above-median concentration in their destination markets are more cyclical than those of other banks (see Figure 3).
Figure 3 Lending distance and market concentration
Overall, our paper suggests that fierce interbank competition in good economic times can lead to deterioration in lending standards, which can be captured by a faster-than-trend expansion of the average distance between lenders and borrowers in an economy. We believe that the findings could be useful to bank regulators. Since distance is easily measurable, it is something that bank supervisors could keep track of (see Meiselman et al. 2018 for another ex-ante measure that might inform supervisors). Of course, the cycle in distance lending, even if risky, may have a silver lining. To the extent that banks push new lending technologies to their limit, it may give them a better understanding of these technologies, and a greater ability to lend at a distance in more normal times. In other words, excess distance lending may expand the normal lending potential of banks, and accelerate the secular trend in lending distance. Until this issue is further explored, any supervisory intervention needs to be measured.
Finally, much of the discussion of the Global Crisis turns on incentive problems in sub-prime mortgage originations, in securitisation, or in bank leverage. Our paper suggests that risk taking was pervasive even in the most unlikely areas, such as stretching to lend at a distance. This would suggest that some underlying common factor that was at its peak in 2006 and 2007 contributed to exacerbating more traditional agency problems. Establishing what this factor is – the suspects include accommodative monetary policy, easy liquidity more generally, and lax supervision – is a question for future research.
Aliber, R and C Kindleberger (2015), Manias, panics and crashes: A history of financial crises, New York: Palgrave Macmillan.
Giannetti, M and L Laeven (2012), “The flight home effect: Evidence from the syndicated loan market during financial crises”, Journal of Financial Economics 104(1): 23-43.
Granja, J, C Leuz and R G Rajan (2018), “Going the extra mile: Distant lending and credit cycles”, NBER working paper w25196.
Lisowsky, P, M Minnis and A Sutherland (2017), “Economic growth and financial statement verification”, Journal of Accounting Research 55(4): 745-794.
Maddaloni, A and J-L Peydró (2010), “Bank risk-taking, securitization, supervision and low interest rates: Evidence from the euro area and the US lending standards”, ECB working paper 1248.
Meiselman, B, S Nagel and A Purnanandam (2018), “Judging banks’ risk by the profits they report”, working paper, University of Chicago.
Minsky, H (2008), Stabilizing and unstable economy, New York: McGraw-Hill.
Mian, A and A Sufi (2009), “The consequences of mortgage credit expansion: Evidence from the US mortgage default crisis”, Quarterly Journal of Economics 124(4): 1449-1496.
Petersen, M A and R G Rajan (2002), “Does distance still matter? The information revolution in small business lending”, Journal of Finance 57(6): 2533-2570.