مفاهیمی در سطح بنگاه از سیاست های پولی ورودی و خروجی وام بانک

Standard models of the Bank Lending Channel are unable to yield predictions on the differential impact of monetary policy shocks over heterogeneous borrowers. This inability has made researchers doubt about the role played by bank credit as a transmission mechanism of monetary policy. Moreover, it has made them reject those models in favor of the Balance Sheet Channel as a transmission mechanism. In this paper we show that an “augmented” version of the Bank Lending Channel that allows for firm heterogeneity (but without any role for firms’ balance sheets) reproduces well the dynamics of firm-level data. Our contribution is to show that it is not clear that the Bank Lending Channel should be rejected in favor of alternative theories on the basis of its inability to reproduce firm-level data. Thus, there is additional room for econometric tests that can provide support to the Bank Lending Channel.

The theory of the Bank Lending Channel holds that the transmission mechanism of monetary policy shocks operates through adjustments to the assets side of banks’ balance sheets ( Bernanke and Blinder, 1988; see also Kashyap and Stein, 1993 and Stein, 1998). The fall in bank reserves that follows a monetary policy contraction directly limits banks’ access to loanable funds, which makes the supply of bank credit fall. Two necessary conditions must be met for this channel to be operative. First, banks must find it costly to use alternative non-reservable sources of funds and/or to re-balance their asset portfolio after the change in reserves. Second, bank borrowers must not be able to perfectly substitute bank loans with alternative financing methods. There are important reasons to believe that these two conditions hold in the US economy.2,3 However, some authors have raised the concern that the Bank Lending Channel hypothesis—in its standard formulation—cannot account for the rich dynamics of firm-level data (see Gertler and Gilchrist, 1994, Oliner and Rudebusch, 1995 and Oliner and Rudebusch, 1996a). This standard formulation of the bank lending channel (as in Bernanke and Blinder, 1988; Stein, 1998 and the literature that followed) assigns a role to banks’/lenders’ heterogeneity, but not to borrowers’ heterogeneity in the transmission of monetary policy. On the contrary, an alternative credit-based explanation of the real effects of monetary policy, the Balance Sheet Channel, has been shown to match well the dynamics of firm-level data on sales, short-term debt and financing choices between open-market credit and bank credit (see Gertler and Gilchrist, 1994, Oliner and Rudebusch, 1996a, Oliner and Rudebusch, 1996b and Bernanke et al., 1996). Therefore, the empirical literature on the credit channels of monetary policy transmission has recently changed its focus of attention towards the Balance Sheet Channel. This theory assigns no special role to bank credit in the transmission of monetary policy. Rather, it is borrowers’ balance sheets that limit their access to external financing. Due to informational asymmetries between borrowers and lenders, borrowing usually requires the former to pledge their net worth as collateral. As a result, a contractionary monetary policy that adversely affects firms’ balance sheet positions also limits their access to all forms of credit. Furthermore, if one believes that small firms have a weaker net worth position in their balance sheets compared to larger firms, then the following key predictions of the Balance Sheet Channel have been confirmed in the data: 1. The financing mix used by firms (defined as the ratio of bank credit to total short-term external financing) does not respond to monetary policy shocks, for neither small nor large firms (see Oliner and Rudebusch, 1996a, Fig. 1, p. 306 and Gertler and Gilchrist, 1994, Fig. 3, p. 322). The financing mix for the economy as a whole does change after a monetary policy shock, but only due to compositional effects: small firms that rely disproportionately more on bank credit are more severely affected by the policy shock than large firms (see Oliner and Rudebusch, 1996a, Fig. 1, p. 306). 3. After a monetary contraction there is a “Flight to Quality”, i.e. a general redirection of all forms of credit from small firms towards large firms (see Gertler and Gilchrist, 1994, Fig. 5, p. 330; Bernanke et al., 1996, Fig. 1, p. 8). 4. The “Flight to Quality” phenomenon holds even within the group of bank-dependent borrowers (Gertler and Gilchrist, 1994 and Oliner and Rudebusch, 1995). 5. There is an asymmetric response over the business cycle of small firm variables to a monetary policy contraction, with a stronger effect during recessions. Large firm variables do not display such an asymmetry (see Gertler and Gilchrist, 1994, Fig. 4, p. 333). To summarize, the criticism to the Bank Lending Channel is that, in its standard formulation, it cannot account for the dynamics of firm-level data described in 1–5 above. However, the argument we pursue in this paper is that with the Bank Lending Channel being silent about the dynamics of firm-level variables, it would be misleading to use firm-level evidence to reject it. That is, the empirical evidence summarized in 1–5 would provide a test to the bank lending view only in relation to an appropriate falsifiable hypothesis of the Bank Lending Channel that can yield predictions on the dynamics of firm-level variables. Since the lending view amounts to a supply-side shock in the market for bank credit, the standard story has little to say about how this shock impacts over different types of borrowers. Our goal in this paper is to build a model suitable to study this falsifiable hypothesis. We augment a general equilibrium model of the Bank Lending Channel with firm heterogeneity, but still without any role for firms’ balance sheets. We then use the model to study the impact of a monetary policy shock over heterogeneous firms. Specifically, the model allows us to study the qualitative dynamics of disaggregated-level variables such as firms sales or production, capital stock, short-term debt and the financing mix. Our model can reproduce the differential effects of monetary policy on heterogeneous borrowers for two main reasons. First, the model features heterogeneous firms that make endogenous investment and financing decisions. Because of financial frictions that make the costs of borrowing vary across financing sources, companies are not indifferent between bank credit and alternative financing sources. Specifically, small high-risk firms tap bank credit, while large low-risk companies can access direct finance. Second, a loan supply shock that increases the marginal cost of funds for banks makes bank-dependent borrowers change their optimal investment and financing strategies, with some of them switching into bond financing. In a general equilibrium setting, this endogenously affects interest rates in all credit markets in the economy, such that monetary policy has real effects not only on small bank-dependent borrowers (i.e. on the ones initially hit by the shock), but also on larger firms that use alternative financing sources. Our main contribution is to show that the qualitative predictions of this “augmented” model of the Bank Lending Channel are mostly in line with the real effects of monetary policy on small and large firms summarized in 1–5. Thus, we show that there is additional room for econometric tests that can provide support to the Bank Lending Channel, and that it is not clear that it should be rejected in favor of alternative mechanisms of monetary policy transmission. Worthy of note, it is by no means our goal to challenge the validity of the alternative Balance Sheet Channel, or to suggest problems in the empirical research on this topic. Following this introduction, the structure of the paper is as follows. Section 2 contains a brief background literature review of the work on the Bank Lending Channel. Section 3 presents a partial equilibrium model of firms’ investment and financing decisions. Section 4 embeds this model into an otherwise standard dynamic general equilibrium model of a closed economy with heterogeneous borrowers. Section 5 simulates the model numerically to study the response to a monetary policy tightening of both macroeconomic aggregates and firm-level variables. Section 6 concludes. The appendix presents the details on the treatment of idiosyncratic risk in the model.

The inability of a benchmark version of the Bank Lending Channel model to yield predictions in line with firm-level data has cast doubts among researchers about the role played by bank credit in the transmission of monetary policy. At the same time, an alternative mechanism, the Balance Sheet Channel, which relies on shocks to individual borrowers’ net worth rather than on shifts of the aggregate supply of credit, has proven successful at replicating these data. It has been stressed that only the Balance Sheet Channel can account for the rich dynamics of several firm-level variables (see Oliner and Rudebusch, 1996b). However, the main argument we make in this paper is that using firm-level data to judge the validity of the Bank Lending story is misleading because standard versions of the Bank Lending Channel are silent about the dynamics of firm-level variables. The existing empirical evidence could provide a test of the bank lending view only in relation to an appropriate falsifiable hypothesis. In this sense, the model we build in this paper provides an adequate framework to confront the Bank Lending Channel hypothesis to the firm-level evidence gathered by previous work. Our “heterogeneous borrower-augmented” version of the Bank Lending Channel embedded into an otherwise standard dynamic general equilibrium model (with no role for firms’ balance sheets) allows us to study the qualitative dynamics of firm-level variables. Our main contribution is to successfully reproduce the qualitative dynamics of firm-level data. Our model predictions can track well firms’ financing choices, the differential access to credit markets across firm sizes and the compositional effects in changes of the financing mix after a monetary policy tightening, all of them empirical regularities attributed to the Balance Sheet Channel and the “Flight to Quality” hypothesis. It is important to note that our goal was by no means to challenge the validity of the alternative Balance Sheet Channel, or to suggest identification problems in previous empirical research. Rather, the idea is to study the firm-level implications of monetary policy operating through the Bank Lending Channel, and to confirm the feasibility of bank lending as a transmission mechanism of monetary policy. Further empirical research is needed to formulate a rigorous econometric test that can provide support to this hypothesis, and possibly give an indication of the relative importance of the two alternative credit channels of monetary policy. We believe the topic we study in this paper is specially relevant because despite the rapid pace at which the disintermediation process has been unfolding in the banking sector during the last three decades, bank credit still remains an important source of external financing for firms (particularly smaller ones). Thus, it is important to assess the extent to which even small shocks to the supply of bank loans can still provide an effective transmission channel of monetary policy.