How does monetary policy affect firm activities? While there is long-standing literature on this issue, the transmission mechanism of monetary policy is currently attracting renewed attention. The reason is that many central banks – including the US Federal Reserve, the Bank of England, the ECB, and the Bank of Japan – have introduced unconventional monetary policies such as quantitative easing and credit easing in the wake of the Global Crisis, and sooner or later will have to exit from these policies.
Apart from the ‘textbook channels’ of monetary policy transmission via interest rates and exchange rates, monetary policy can potentially affect firm activities through its impact on firms’ and banks’ balance sheets. In a seminal study on this balance sheet channel, Bernanke and Gertler (1989) showed theoretically that negative shocks to borrowers’ net worth increase the agency costs of financing investment, and thus reduce investment. Following their theoretical model, Bernanke et al. (1996, 1999) further showed that monetary tightening leads to a decrease in borrowers’ net worth, and consequently reduces borrowing and investment. They also show that an economic downturn has a greater effect on the availability of credit, investment, and output for firms with lower net worth. This implies that the impact of an aggregate shock on firms differs depending on firms’ net worth.
While the above studies focus on the credit demand side, Holmstrom and Tirole (1997) applied the argument to the banking sector, and showed that banks with lower net worth supply fewer loans. Their argument implies that banks with lower net worth will supply less credit when monetary policy is tightened or economic growth is lower. The reason is that, in addition to taking insured deposits, banks need to raise funds by issuing uninsured debt, which, as in the case of firms, is susceptible to agency costs.
The argument that monetary policy is also transmitted through the supply of credit is often labelled as the ‘bank-lending view’. According to this view, monetary policy shifts banks’ loan supply curve, and thereby affects investment and other activities of bank-dependent borrowers (e.g. Bernanke and Blinder 1988, Kashyap and Stein 1994). Furthermore, shifts in loan supply depend on banks’ balance sheets. For example, when the central bank sells securities to a bank through open market operations, this decreases the bank’s reserves, and the bank may have to decrease its loans unless it makes up for any shortfall in deposits by selling security holdings or by issuing nonreservable debt. Banks with fewer liquid assets will need to decrease loans more if they cannot issue nonreservable debt or can do so only at a higher cost than deposits (Stein 1998).
A substantial number of empirical studies have sought to examine the role of bank net worth in loan supply. They purport to show that bank liquidity or capital have a significant effect on lending, and that these effects are stronger when monetary policy is tight.1 However, these studies tend to use aggregate-level or bank-level data, meaning that they cannot clearly disentangle shocks to loan supply and shocks to loan demand.
Identifying shocks to loan supply and loan demand
Given this identification problem, recent studies have employed loan-level or firm-bank match-level data to isolate loan supply shocks from loan demand shocks (for a survey, see Hosono and Miyakawa 2014a). Khwaja and Mian (2008) were the first to employ the strategy of identifying (bank-specific) loan supply shocks as changes in loans after controlling for firm-level fixed effects, which are assumed to reflect firm-specific loan demand shocks (as well as aggregate loan supply shocks). They examined whether banks that experience a larger withdrawal of deposits due to an exogenous shock (a nuclear experiment in Pakistan) reduce their lending to client firms more than other banks, and found that this is indeed the case.
Meanwhile, using data on loan applications in Spain, Jiménez et al. (2012) examined how changes in aggregate variables such as interest rates and GDP, as well as the interaction between these variables and lender bank characteristics, affect the likelihood of loans being granted. Extending the empirical strategy employed by Khwaja and Mian (2008), they controlled for the time-variant quality of potential borrowers by considering either firm-month or loan-level fixed effects.2 They found that higher short-term interest rates and lower GDP growth reduce the probability for a loan to be granted, and that this tendency is stronger for banks with low capital (in periods of higher short-term interest rates and lower GDP growth) or low liquidity (in periods of higher short-term interest rates).
However, although the study by Jiménez et al. (2012) successfully overcame the identification problems in testing the hypothesis that bank net worth plays a more important role when monetary policy is contractionary or economic growth is low, they do not:
(i) estimate the quantitative impact of bank capital and liquidity on changes in loan supply,
(ii) estimate the impact of the bank balance sheet channel on firms’ total loans and investment, or
(iii) analyse the impact of unconventional monetary policies.3
New evidence on bank lending in Japan
Against this background, in Hosono and Miyakawa (2014b), we set out to test the bank balance sheet channel hypothesis by addressing the identification problem using a unique dataset for Japan. Our dataset is a panel that covers firms listed on Japanese stock exchanges, and spans almost three decades. It contains information on the banks with which each firm transacts, on the amount of outstanding loans that each firm has with each bank, and on the balance sheet variables of each firm and bank. We combine this with information on monetary policy operations, including the beginning and ending of quantitative easing by the Bank of Japan.
Using this dataset, we examine:
- First, whether banks’ net worth and liquidity (measured in terms of capital and liquidity, respectively, relative to total assets) affect changes in loan supply, and
- Second, whether tighter monetary policy or lower economic growth strengthens the effects of bank net worth and liquidity on changes in loan supply.
We focus on the variation in changes in outstanding loans across banks for the same firm and year to disentangle bank loan supply from demand. Furthermore, we focus on the variation in changes in outstanding loans over time for the same bank and firm to control for assortative matching between banks and firms. We can therefore estimate the quantitative impact of monetary policy and business cycles on changes in outstanding loans purely through bank net worth and liquidity. Using data on the amount of loans, we quantitatively assess the bank balance sheet channel, i.e. changes in bank lending caused by changes in the real GDP growth rate and monetary policy through bank balance sheets. Finally, we investigate how the bank balance sheet channel affects firms’ overall borrowing and investment by focusing on the average level of banks’ net worth and liquidity computed over all the lender banks for each firm.
The main findings of our analysis can be summarised as follows.
1. Banks with a higher liquidity-to-asset ratio tend to supply more loans.
Suppose, for example, that the liquidity ratio of lender bank i for firm j declines by one standard deviation (i.e. 0.068) in year t-1, and that monetary policy does not change in year t, while the real GDP growth rate is zero in year t. Given the estimated parameters, the model predicts that the growth rate of loan provision from bank i to firm j will be 2.6% smaller than in the case that bank liquidity had not declined. Considering that the sample mean and the standard deviation of match-level loan growth are -0.3% and 48.8%, respectively, this implies that bank liquidity has an economically sizeable impact on the growth rate of loans.
2. The effects of bank capital on loan supply were significant during the 2000s, when Japan’s regulatory authorities strictly enforced capital adequacy regulations.
3. The effects of bank capital and bank liquidity on loan supply are stronger when the economic growth rate is lower.
Suppose, for example, that the real GDP growth rate is -2% in year t. Assuming a one-standard-deviation decline in the liquidity ratio of lender bank i, the growth rate of loan provision from bank i to firm j will be 3.4% lower than in the absence of a decline in bank liquidity. This decline in the growth rate of loan provision is larger than in the case above (2.6%) where the real GDP growth rate is 0%.
4. The effects of bank liquidity on loan supply are weaker when monetary policy is loosened and stronger when monetary policy is tightened or the central bank exits from quantitative easing. In particular, the impact of bank liquidity more than doubled when quantitative easing was terminated in 2006 compared to the case of no change in monetary policy.4
5. Fluctuations in economic growth and monetary policy are transmitted to capital investment through the bank balance sheet channel in the case of firms with high investment opportunities. For example, firms with better investment opportunities tend to invest more when their lender banks are more liquid, and this link is stronger during an economic downturn.
Overall, our findings support the hypotheses that bank net worth affects loan supply and that the effect depends on monetary policy and economic growth. Moreover, this bank balance sheet channel has a significant impact on firms’ financing and investment. The results imply that from a policy perspective, it is important to carefully choose the timing of the exit from expansionary monetary policy. In particular, exiting from unconventional monetary policies when bank balance sheets are weak could have a severe adverse impact on firms’ activities. It should also be noted that such an adverse impact would potentially be particularly strong for firms with better investment opportunities, which are expected to be the main drivers of economic growth.
Editor's note: This column is based on research conducted by the authors while participating in the Project on Corporate Finance and Firm Dynamics undertaken at the Research Institute of Economy, Trade and Industry (RIETI). Hosono gratefully acknowledges financial support from Grant-in-Aid for Scientific Research (S) No. No.22223004, JSPS.
Amiti, M and D E Weinstein (2013), “How Much Do Bank Shocks Affect Investment? Evidence from Matched Bank-Firm Loan Data”, NBER Working Paper 18890.
Bernanke, B S and A S Blinder (1988), “Credit, Money, and Aggregate Demand”, American Economic Review (Papers and Proceedings), 78(2): 435–439.
Bernanke, B S and A S Blinder (1992), “The Federal Funds Rate and the Channels of Monetary Policy Transmission”, American Economic Review, 82(4): 901–921.
Bernanke, B S and M Gertler (1989), “Agency Costs, Net Worth, and Business Fluctuations”, American Economic Review, 79(1): 14–31.
Bernanke, B S, M Gertler, and S Gilchrist (1996), “The Financial Accelerator and the Flight to Quality”, Review of Economics and Statistics, 78(1): 1–15.
Bernanke, B S, M Gertler, and S Gilchrist (1999), “The Financial Accelerator in a Quantitative Business Cycle Framework”, in J B Taylor and M Woodford (eds.), Handbook of Macroeconomics, Ed 1, Vol 1, Ch 21, Elsevier: 1341–1393.
Bernanke, B S and C S Lown (1991), “The Credit Crunch”, Brookings Papers on Economic Activity, 22(2): 205–248.
Favero, C A, F Giavazzi, and L Flabbi (1999), “The Transmission Mechanism of Monetary Policy in Europe: Evidence from Banks’ Balance Sheets”, NBER Working Paper 7231.
Holmstrom, B and J Tirole (1997), “Financial Intermediation, Loanable Funds, and the Real Sector”, Quarterly Journal of Economics, 112(3): 663–691.
Hoshi, T, D Scharfstein, and K J Singleton (1993), “Japanese Corporate Investment and Bank of Japan Guidance of Commercial Bank Lending”, in K J Singleton (ed.), Japanese Monetary Policy, Chicago: University of Chicago Press: 63–94.
Hosono, K (2006), “The Transmission Mechanism of Monetary Policy in Japan: Evidence from Banks’ Balance Sheets”, Journal of the Japanese and International Economies, 20(3): 380–405.
Hosono, K and D Miyakawa (2014a), “Bank Lending and Firm Activities: Overcoming Identification Problems”, forthcoming in T Watanabe, I Uesugi, and A Ono (eds.), The Economics of Interfirm Networks, Advances in Japanese Business and Economics, Vol 4, Springer.
Hosono, K and D Miyakawa (2014b). “Business Cycles, Monetary Policy, and Bank Lending: Identifying the bank balance sheet channel with firm-bank match-level loan data”, RIETI Discussion Paper 14-E-026.
Ito, T and Y N Sasaki (2002), “Impacts of the Basle Capital Standard on Japanese Banks’ Behavior”, Journal of the Japanese and International Economies, 16(3): 372–397.
Jayaratne, J and D P Morgan (2000), “Capital Market Frictions and Deposit Constraints at Banks”, Journal of Money, Credit and Banking, 32(1): 74–92.
Jiménez, G, S Ongena, J Peydro, and J Saurina (2012), “Credit Supply and Monetary Policy: Identifying the Bank Balance-Sheet Channel with Loan Applications”, American Economic Review, 102(5): 2301–2326.
Kashyap, A K and J C Stein (1994), “Monetary Policy and Bank Lending”, in N G Mankiw (ed.), Monetary Policy, Chicago and London: University of Chicago Press: 221–256.
Kashyap, A K and J C Stein (2000), “What Do a Million Observations on Banks Say about the Transmission of Monetary Policy?”, American Economic Review, 90(3): 407–428.
Kashyap, A K, J C Stein, and D W Wilcox (1993), “Monetary Policy and Credit Conditions: Evidence from the Composition of External Finance”, American Economic Review, 83(1): 78–98.
Khwaja, A I and A Mian (2008), “Tracing the Impact of Bank Liquidity Shocks: Evidence from an Emerging Market”, American Economic Review, 98(4): 1413–1442.
Peek, J and E S Rosengren (1997), “The International Transmission of Financial Shocks: The Case of Japan”, American Economic Review, 87(4): 495–505.
Ramey, V A (1993), “How Important is the Credit Channel in the Transmission of Monetary Policy?”, Carnegie-Rochester Conference Series on Public Policy, 39: 1–45.
Romer, C D and D H Romer (1990), “New Evidence on the Monetary Transmission Mechanism”, Brookings Papers on Economic Activity, 21(1): 149–198.
Stein, J C (1998), “An Adverse Selection Model of Bank Asset and Liability Management with Implications for the Transmission of Monetary Policy”, RAND Journal of Economics, 29(3): 466–486.
Ueda, K (1993), “A Comparative Perspective on Japanese Monetary Policy: Short-run Monetary Control and the Transmission Mechanism”, in K J Singleton (ed.), Japanese Monetary Policy, Chicago: University of Chicago Press: 7–29.
Woo, D (1999), “In Search of ‘Capital Crunch’: Supply Factors behind the Credit Slowdown in Japan”, IMF Working Paper 99/3.
1 Examples include the studies by Romer and Romer (1990), Bernanke and Blinder (1992), Kashyap et al. (1993), Hoshi et al. (1993), Ueda (1993), and Ramey (1993), who use aggregate data to examine the bank lending channel of monetary policy. On the other hand, Jayaratne and Morgan (2000) use bank-level data to study the relationship between bank liquidity and loans, while Bernanke and Lown (1991), Peek and Rosengren (1997), Woo (1999), and Ito and Sasaki (2002) use bank-level data to examine the relationship between bank capital and loans. Meanwhile, Kashyap and Stein (2000), Favero et al. (1999), and Hosono (2006) use bank-level data to investigate the bank lending channel of monetary policy.
2 They used information on a firm’s successive loan applications to different banks when they controlled for loan-level fixed effects.
3 Amiti and Weinstein (2013) also use matched bank-firm loan data to identify idiosyncratic bank shocks, i.e. movements in bank loan supply net of borrower characteristics and general credit conditions. Using Japanese firm-bank data, they showed that idiosyncratic bank shocks have a large impact on firms’ investment. However, they did not examine the bank balance sheet channel of monetary policy and GDP growth rates.
4 In the case that quantitative easing is terminated, the estimated growth rate of loan provision from bank i to firm j is 6.0 percentage points smaller than without a decline in bank liquidity.