Lending Pro-Cyclicality and Macro-Prudential Policy: Evidence from Japanese LTV Ratios
Using a large and unique micro dataset compiled from the official real estate registry in Japan, we examine the loan-to-value (LTV) ratios for business loans from 1975 to 2009 to draw some implications for the ongoing debate on the use of LTV ratio caps as a macro-prudential policy measure. We find that the LTV ratio exhibits counter-cyclicality, implying that the increase (decrease) in loan volume is smaller than the increase (decrease) in land values during booms (busts). Most importantly, LTV ratios are at their lowest during the bubble period in the late 1980s and early 1990s. The counter-cyclicality of LTV ratios is robust to controlling for various characteristics of loans, borrowers, and lenders. We also find that borrowers that exhibited high-LTV loans performed no worse ex-post than those with lower LTV loans, and sometimes performed better during the bubble period. Our findings imply that a simple fixed cap on LTV ratios might not only be ineffective in curbing loan volume in boom periods but also inhibit well-performing firms from borrowing. This casts doubt on the efficacy of employing a simple LTV cap as an effective macro-prudential policy measure.
The recent financial crisis with its epicenter in the U.S. followed a disastrous financial crisis in Japan more than a decade before. It is probably not an exaggeration to argue that these crises shattered the illusion that the Basel framework – specifically Basel I and Basel II – had ushered in a new era of financial stability. These two crises centered on bubbles that affected both the business sector (business loans) and the household sector (residential mortgages). In Japan banks mostly suffered from the damage in the business sector, while in the U.S. banks mostly suffered from damage in the household sector. Following the first of these crises, the Japanese crisis, a search began for policy tools that would reduce the probability of future crises and minimize the damage when they occur. Consensus began to build in favor of countercyclical macro-prudential policy levers (e.g., Kashyap and Stein 2004). For example, there was great interest and optimism associated with the introduction by the Bank of Spain of dynamic loan loss provisioning in 2000. Also, Basel III adopted a countercyclical capital buffer to be implemented when regulators sensed that credit growth has become excessive.