Unconventional monetary policy, funding expectations, and firm decisions

Annalisa Ferrando, Alexander Popov, Gregory F. Udell

Working Paper Series

Unconventional monetary policy, funding expectations and firm decisions

No 2598 / October 2021

Disclaimer: This paper should not be reported as representing the views of the European Central Bank (ECB). The views expressed are those of the authors and do not necessarily reflect those of the ECB.

Abstract

We study the transmission of (unconventional) monetary policy to the real sector when …rm decisions depend on both current and future credit market conditions. For a given level of current credit access, investment and employment increases more at …rms expecting bank credit to improve in the future. Three separate unconventional policies by the ECB- the OMT, the introduction of negative rates, and the CSPP- improved expectations of future credit access for SMEs borrowing from banks that were expected to increase SME lending due to the policy. Our results enhance our understanding of the bank balance sheet channel of monetary policy.

JEL classi…cation: D22, D84, E58, G21, H63.

Keywords: Unconventional monetary policy, funding expectations, corporate investment.

ECB Working Paper Series No 2598 / October 2021

1

Non-technical summary

The "bank lending channel" hypothesis postulates that monetary policy is transmitted to the real economy through changes in the level and composition of bank credit. In this paper, we argue that …rms' expectations about future credit availability play an important- and hitherto undocumented- role in the "bank lending channel" of unconventional monetary policy. By targeting a particular set of banks, monetary policy improves the expectations about future credit availability of …rms borrowing from these banks. In turn, higher expectations about future credit conditions are re‡ected in …rms'investment and employment decisions before an actual improvement in credit conditions takes place. The mechanism we uncover is important because while it can take a year before bank balance sheets adjust to a monetary policy shock, adjustments in expectations can be immediate. It is also novel: while shocks to in‡ation expectations have been studied extensively, to our knowledge we are the …rst to analyze the interplay among monetary policy, …rms' funding expectations, and …rms' real decisions. Because such decisions are central to many macroeconomic models, it is crucial to understand the precise mechanisms driving this interplay.

We examine the impact of three separate unconventional monetary policies enacted by the ECB in the wake of the twin …nancial and sovereign debt crisis in Europe on the variation in small …rms' expectations of future credit availability. The three policies are the OMT Program, announced in the summer of 2012; the introduction of negative policy rates in the summer of 2014; and the announcement of corporate bond purchases in the spring of 2016. We investigate whether after the announcement of each policy, small …rms experienced an improvement in their expectations of future credit availability, especially if their main creditor was plausibly expected to increase small business lending as a result of the policy. We also study how this improvement in expectations a¤ected …rms'real decision, such as investment and employment. We do so for a sample of around 7,000 SMEs in eight euro area countries, using a restricted access dataset containing rich balance sheet information for individual …rms, information on expectations, and the identity of their main bank. As a source of identifying variation, we use the …rm-bank match: while all …rms are subject to the same policy announcement, some …rms know they will bene…t from the new policy because their creditor is going to be a¤ected by it, and others know they will not.

We …nd that the announcement of the OMT Program resulted in a strong improvement (over

ECB Working Paper Series No 2598 / October 2021

2

the next year) in expectations of future credit availability by …rms borrowing from banks with substantial balance sheet exposures to impaired sovereign debt whose funding costs were expected to decline as a result of the announcement. Similarly, the introduction of negative policy rates improved credit expectations for …rms borrowing from high-deposit banks, which were expected to increase risky lending because they would be reluctant to pass on the negative rates to their depositors. Finally, the announcement of the CSPP improved credit expectations for …rms borrowing from banks focusing so far on large-…rm lending, which were expected to have a spare lending capacity as a result of large …rms …nding it cheaper to sell bonds to the ECB than to borrow form banks. While the results are most consistently signi…cant across model speci…cations in the case of the OMT Program, the evidence suggests that all of these three policies broadly work in the same direction. Our results point to the importance of funding expectations in the transmission of unconventional monetary policy through the bank lending channel.

We also …nd that controlling for current credit access, …rms with higher expectations of future credit availability increase investment and employment. This is true both at all times, as well as during the period in question for …rms a¤ected by unconventional monetary policy shocks. This suggests that improvements in funding expectations can impact …rms'real decision over and above …rms'actual access to …nance. Because changes in expectations often lead changes in credit access by 6 months to one year, the mechanism we identify not only complements, but also precedes the bank balance sheet adjustment channel of monetary policy. Our results thus imply that unorthodox monetary policy can have an impact on the real economy faster than previously believed. This has signi…cant implications for the unconventional monetary policy initiatives being deployed to address the COVID-19 crisis- including, for example, programs such as the Primary Market Corporate Credit Facility and the Main Street Facilities deployed by the Federal Reserve, or the Pandemic Emergency Purchase Programme by the ECB. Our results suggest that policy makers and researchers should be aware that the e¢ cacy of these programs could play out not only through the conventional balance sheet channel, but also the expectations channel we identify in this paper.

ECB Working Paper Series No 2598 / October 2021

3

• Introduction

The "bank lending channel" hypothesis postulates that monetary policy is transmitted to the real economy through changes in the level and composition of bank credit.1 The same broadly applies to unconventional monetary policy. For example, the Large Scale Asset Purchases (LSAP) by the Federal Reserve Board increased the price and value of banks'asset holdings, thereby recapitalizing banks and stimulating lending (e.g., Rodnyanski and Darmouni, 2017; Chakraborty, Goldstein, and MacKinley, 2020). Analogously, the Corporate Sector Purchase Program (CSPP) of the European Central Bank (ECB) sought to reduce the cost of debt for …rms relying on bond …nancing, thereby reducing their demand for credit and increasing banks'willingness to lend to SMEs.

In this paper, we argue that …rms'expectations about future credit availability play an important- and hitherto undocumented- role in the "bank lending channel" of unconventional monetary pol- icy. By targeting a particular set of banks, monetary policy improves the expectations about future credit availability of …rms borrowing from these banks. In turn, higher expectations about future credit conditions are re‡ected in …rms'investment and employment decisions before an actual improvement in credit conditions takes place. The mechanism we uncover is important because while it can take a year before bank balance sheets adjust to a monetary policy shock (e.g., Gertler and Karadi, 2015), adjustments in expectations can be immediate. It is also novel: while shocks to in- ‡ation expectations have been studied extensively (e.g., Cogley and Sargant, 2008; Orphanides and Williams, 2008; Guiliano and Spilimbergo, 2014), to our knowledge we are the …rst to analyze the interplay among monetary policy, …rms'funding expectations, and …rms'real decisions. Because such decisions are central to many macroeconomic models, it is crucial to understand the precise mechanisms driving this interplay.

We study three unconventional monetary policies adopted by the ECB in recent years. The …rst one is the Outright Monetary Transactions (OMT) Program announced in the summer of 2012. Under this program, the ECB committed itself- under strict conditionality- to purchasing eligible sovereign bonds issued by euro area governments. The announcement of the OMT program reduced yields on sovereign bonds issued by …scally stressed countries immediately, sharply, and

• See Bernanke and Blinder (1988) and Bernanke and Gertler (1989) for theoretical contributions, and Kashyap and Stein (2000), Gambacorta (2005), and Jimenez, Ongena, Peydro, and Saurina (2012) for empirical evidence, among others.

ECB Working Paper Series No 2598 / October 2021

4