We aim to study the interaction between bank liquidity and loan monitoring over the business cycle and their distributional impact. The analysis is motivated by various phenomena that in the long run are deepening the markets for bank liquidity. On the regulatory front, also in the wake of the Great Recession, effort has been made to facilitate banks' access to wholesale funding markets. On the banking side, banks' have increased their tendency to accumulate large holdings of sovereign bonds and this has allegedly expanded their collateral and debt capacity in interbank markets, while leaving banks exposed to swings in the price of sovereign bonds. To this end we need to conduct an empirical analysis with bank-level U.S. data to study the existence of pervasive substitution effects between banks' access to liquidity and bank monitoring, especially during recessionary periods. To rationalize these findings, we further need to build a DSGE model with a banking sector that features active loan monitoring and limited access to liquidity, due to borrowing constraints in both the retail and the wholesale funding markets. The model economy also includes a government that issues bonds, used as collateral by banks in the wholesale market. In this framework we can study whether bank monitoring and liquidity tend to comove negatively or positively following recessionary and liquidity shocks. In line of principle, two channels can arise in our model: a "liquidity channel", whereby changes in access to liquidity alter directly banks' incentive to monitor to retain this access; and a "monitoring productivity channel", whereby changes in the value of loans influence the marginal contribution of monitoring in enhancing loan pledgeability.
The aim of this research is to investigate a channel that has been quite neglected by macroeconomist, i.e., the interaction between the access to liquidity and the incentive arising for private banks to monitor their loans. To the best of our knowledge there are several empirical papers that singularly study each of this channel but neither of them investigate the joint role played by bank monitoring and liquidity.
Our novelty is to study how policy and structural factors, traditionally viewed as stabilizing forces in the wake of recessionary shocks, may however have unintended consequences in terms of dilution of bank monitoring incentives during the recessions. Regarding the credit market structure, credit relationships can in fact mitigate the effects of shocks but also diminish the substitutability between bank liquidity and monitoring. On the policy front, credit interventions aimed at sustaining banks' access to liquidity can potentially dilute bank monitoring, at least in the short run.
Thus, the importance of our analysis relies in the construction of a medium-scale DSGE model accounting for a banking sector able to reproduce what observed at empirical level, where we study the relation incurring between the incentive to monitor and the access to liquidity in the wholesale funding market along different phases of the business cycle (the latter as classified by the National Bureau of Economic Research). We not only aim to replicate at theoretical level what obtained at empirical level, but, more ambitiously, we want to study all the economic mechanisms that underline the interplay between the two channels emerging in our model: the "liquidity channel" and the "monitoring productivity channel". Finally, we intend to draw reccomendation for the central bank activity. While in fact, the direct effect of credit policies on the real economy recovery are widely recognized, less words have been spent about how do they affect the bank monitoring incentives. In fact, they can potentially dilute the stabilizing function of monitoring in the wake of tighter liquidity, at least in the short run. Instead, over the medium-long run this side effect of the policy can lose relevance and the depressing effect of the policy on monitoring vanishes. To summarize, it is important to analyse this channel as credit policies may have an unintended consequences in terms of weaker banks monitoring, thus making the soundness of loans more fragile. Yet, such effects are relatively short-lead and in the long-run the credit policy retains only its direct output-stabilizing role. Importantly, however, we observe that the diluting effects on monitoring in the short-run can be either more or less pronounced depending of which kind of credit policy the central bank decides to implement (for example, direct lending, interbank lending and equity injections).
This involves that a policymaker concerned about side effects of credit policies associated with bank monitoring behavior should privilige, for each shock, the policy for which the side effect is less pronounced.