عدم تقارن وام و سرمایه بانک در انتقال سیاست های پولی

Utilizing a bank-lending channel framework, we investigate the effects of expansionary and contractionary policy separately on the loan behavior of low-capital and high-capital banks, and between pre-Basel/FDICIA and post-Basel/FDICIA periods. Our results show that low-capital banks are adversely affected by contractionary policy. Expansionary policy, however, is not effective in stimulating the loan growth of low-capital banks. These results are consistent with lending channel predictions, but only hold in the post-Basel/FDICIA period when the capital constraint is stringent relative to the pre-Basel/FDICIA period. These asymmetric policy results have implications for the interaction of monetary and capital regulatory policies.

The bank-lending channel (BLC) focuses on bank behavior as an integral part of the monetary transmission mechanism. According to the lending channel, monetary policy must be able to shift the loan supply of some banks and some of the borrowers of these banks must be bank dependent. Contractionary monetary policy can decrease loan supply due to credit market imperfections faced by some banks. Expansionary policy increases loan supply due to a lack of constraints on the lending of some banks. Thus, for a given stance of monetary policy, evidence of a BLC arises from the cross-sectional asymmetric loan responses by constrained and unconstrained banks. These effects of policy on loan growth, and eventually on expenditures, complement the usual interest rate/money channel. A growing subset of BLC models emphasizes bank capital as a relevant constraint that distinguishes a BLC. Contractionary monetary policy is argued to have severe adverse effects on the loans of capital-constrained banks relative to unconstrained banks. For expansionary policy, the lack of a binding capital constraint allows unconstrained banks to expand loans relative to constrained banks. Although the BLC literature stresses these cross-sectional effects resulting from a given policy stance, it also implies a loan asymmetry over different policy stances for a given capital constraint. In particular, the above arguments suggest that the loan supply of constrained banks responds more strongly to contractionary policy than to expansionary policy.2 Identifying this policy-stance effect on capital-constrained banks may aid in better understanding the effects of monetary policy over the business cycle For example, if a significant portion of the banking system is capital-constrained, contractionary policy may have too severe of an effect on loans and expansionary policy may not be successful in increasing bank loans. This asymmetry could have consequences for the mitigating effects of policy over the business cycle. This emphasis is important since policy-stance asymmetry could explain the well-documented asymmetric effects of monetary policy on nominal GDP. Empirical studies of the BLC have two major shortcomings. First, while the theoretical literature clearly argues that the effects of contractionary policy on the loans of the constrained banks differ from those of expansionary policy; empirical studies fail to make this distinction. Thus, it is unclear whether contractionary policy, expansionary policy or both drive the empirical results that distinguish a BLC. No empirical study has examined the separate effects of each policy stance in identifying a BLC. Second, the regulatory literature shows strong evidence that the effective capital constraint changed for banks after the implementation of Basel I and FDICIA. Although these changes and their effects on loans through expansionary policy have been recently emphasized in the BLC literature, no empirical studies have considered these effects. A change in the effective capital constraint could affect the measured existence and strength of a BLC, and therefore has consequences for how monetary policy works through the BLC. Our paper explores the effects of monetary policy on bank loan growth when low-capital and high-capital banks are used to distinguish a BLC. We include three innovations that contribute to the BLC literature. First, we explore the differences between the loan responses of the low and high-capital banks to monetary policy, where policy is divided into expansionary and contractionary stances. Next, we compare the separate effects of contractionary and expansionary monetary policy on the bank loans of the low-capital banks and then on the loans of the high-capital banks. Finally, we consider the effects of a possible change in the way bank loans respond to policy between the pre-Basel/FDICIA and post-Basel/FDICIA periods, where possible changes in the effective capital constraint may have taken place because of a change in either regulatory requirements or enforcement of these requirements or both. In particular, we are concerned with answering the following five questions: (1) Do low-capital banks reduce loan growth by more than high-capital banks during contractionary policy? (2) Do low-capital banks grow loans less than high-capital banks during expansionary policy? (3) Do low-capital banks respond more to contractionary policy than to expansionary policy? (4) Do high-capital banks respond more to expansionary policy than to contractionary policy? (5) Do the answers to the above questions change between the two periods over which regulatory capital requirements or enforcement change? Our results indicate an asymmetric loan response by the low and high-capital banks for each policy stance, which is consistent with the empirical literature. Additionally, our results show an asymmetric loan response to contractionary and expansionary policy by the low-capital banks, which is consistent with implications of the theoretical literature. However, we find that the character of these asymmetries is affected by apparent differences in the effective capital constraint between pre-Basel/FDICIA and post-Basel/FDICIA periods. In the pre-Basel/FDICIA period, when the capital constraint is weak, contractionary policy is ineffective for low-capital banks, but expansionary policy increases the loans of these banks. In the post-Basel/FDICIA period, marked by a stronger effective constraint, contractionary policy decreases the loan growth of the low-capital banks, but expansionary policy is weak. The post-Basel/FDICIA results are consistent with the theoretical models of the BLC literature. The pre-Basel/FDICIA results are not consistent with these arguments, but are consistent with a weak capital regulatory constraint in this period. The pre-Basel/FDICIA results offer an explanation as to why it may be difficult to find convincing evidence of a BLC through capital in samples that include this first period.3 Our results are also consistent with the credit-crunch literature showing weak loan growth by low-capital banks during the recovery from the 1990–1991 recession. Although the BLC literature stresses the behavior of the capital constrained banks, our results for the high-capital banks in the post-Basel/FDICIA period are also interesting. The high-capital banks do not decrease loans in response to contractionary policy, but they do increase loans in response to expansionary policy. This result is especially strong for real estate loans and is consistent with the recent surge in real estate lending following interest rate cuts by the Fed at a time when banks were fairly well capitalized. These results also imply that expansionary monetary policy may be most effective in pulling the economy out of a recession when the banking system is well capitalized. Our evidence emphasizes a mutually beneficial relationship between regulatory and monetary policies. The next section presents a literature review, which is divided into three parts, reflecting the three literatures we employ to argue our point: (1) contractionary/expansionary monetary policy and the BLC, with an emphasis on the role of capital in distinguishing a BLC, (2) the implications of loan asymmetry arising from a capital constraint associated with the BLC and (3) the effects of a changing regulatory capital constraint on the BLC. The third section lays out data considerations and outlines our empirical approach. The third section also presents our results on the loan responses of low and high-capital banks to monetary policy. The fourth section draws policy conclusions from our results.

Recent models associated with the bank-lending channel (BLC) emphasize bank capital as a constraint that allows contractionary monetary policy to severely curtail the loan growth of capital-constrained banks. These models also argue that a capital constraint can shut down the BLC, disabling the ability of expansionary policy to increase the loan growth of constrained banks. These policy effects emphasize an asymmetry between the loan responses of capital-constrained and unconstrained banks for a given stance of policy (cross-sectional asymmetry). Considered over both policy stances, these models also imply an asymmetry between the effects of contractionary and expansionary policy for a given level of capital (policy-stance asymmetry). These two types of asymmetry may change in nature and intensity due to changes in regulatory and market capital constraints. Our paper empirically explores these asymmetries. We find that monetary policy has the expected impact on the loan growth of the small low-capital banks. That is, contractionary monetary policy decreases the loans of the small low-capital banks relative to high-capital banks, and expansionary monetary policy is not able to increase the loan growth of the low-capital banks relative to the high capital banks. However, we find that these results only hold in the post-Basel/FDICIA period when the effective capital constraint was increased relative to the pre-Basel/FDICIA period. These results imply that merely having de jure capital regulations is not sufficient to obtain the results argued in the BLC literature. Rather, they show that the constraint must be of a certain level or stringency, and enforced effectively for these results to hold. Although the literature stresses the behavior of capital-constrained banks, there are important policy implications that follow from the behavior of our high-capital banks. These banks do not respond to contractionary monetary policy, but the small high-capital banks increase loan growth in response to expansionary policy. If small banks are better capitalized in the future, they might play an important role in aiding expansionary policy during economic recoveries.24 The results of the low-capital and high-capital banks have implications for monetary policy over the business cycle. These results are consistent with the conventional wisdom on the effects of expansionary policy on loan and GDP growth following the recoveries from the 1990 and 2001 recessions. In the post-1990 recessionary period, it was argued that expansionary policy could not overcome the capital crunch that existed in the banking system due to the newly enforced Basel capital standards. Following the 2001 recession, the banking system was considered much better capitalized and expansionary monetary policy is argued to have encouraged real estate loan growth. Our results on total loans and, in particular, on real estate loans are consistent with this assessment. Likewise, our results for low-capital banks are consistent with media reports on the effects of contractionary policy proceeding the 1990 recession, when the banking system was thought to be capital constrained. Thus, in so much as loan growth affects aggregate demand; our results of the asymmetric effects of policy on loans could contribute to the well-established asymmetric effect of monetary policy on nominal GDP. Finally, our results have implications for the future operation of the BLC through capital constraints. The above results highlight the interaction between regulatory and monetary policies and between bank structural changes and monetary policy. Capital regulation appears to be moving toward strengthening the capitalization of the banking system to improve its soundness. Additionally, consolidation of the banking system has dramatically altered its size distribution. If the banking system moves towards larger and better-capitalized banks, the effects of the BLC may be weakened. The above-mentioned points imply that monetary policy should carefully consider the overall capitalization of the banking system, the distribution of capital across banks and changing capital regulations in formulating monetary policy. Both system-wide capital and the distribution of capital over different size banks change over the business cycle. In so much as capital is a constraint on loans, the pro-cyclicality of bank capital could encourage loan volatility and contribute to greater GDP volatility (Bliss and Kaufman, 2003). The capital constraints imposed by Basel II, which is currently being implemented by some banks and takes effect in 2007, may encourage even greater volatility in loans over the business cycle (Kashyap and Stein, 2004). If Basel II does create more constrained banks during recessions and better capitalized banks during expansions, our analysis suggests that countercyclical policy may be even more difficult. In general, the changes brought about by Basel II will result in banks reallocating their capital and consequently, adjusting the way they organize their balance sheets and the way they react to monetary policy shocks. Further work should be undertaken on the monetary policy implications of these new regulations.