VoxEU Column Financial Markets Monetary Policy

Does a surprise tightening of monetary policy expand shadow banking?

There has been an extensive debate over whether central banks should raise interest rates to ‘lean against’ the build-up of leverage in the financial system. This column reports on empirical evidence showing that, in contrast to the conventional view, surprise monetary contractions have tended to increase shadow bank asset growth, rather than reduce it in the US. Monetary policy had the opposite effect on commercial bank asset growth. These findings cast some doubt on the idea that monetary policy could be used to “get in all the cracks” of the financial system in a uniform way.

Was monetary policy an important driver of financial intermediaries’ balance sheet dynamics in the run-up to the global financial crisis? Should monetary policy have been ‘leaning against the wind’ of the rapid build-up in financial sector leverage that preceded the crisis – including that in the shadow banking sector? A popular narrative is that low US interest rates post-2001 fuelled leverage growth and prepared the ground for the global calamity of 2007–2008. It is therefore argued that monetary policy should have been tighter, particularly because its effects extend beyond the reach of more targeted regulatory tools, “get[ting] in all of the cracks” (Stein 2013).

US monetary policy and the expansion of balance sheets

Figure 1 shows that US monetary policy in the run-up to the crisis can be characterised by two phases. In the first phase, the Fed cut rates from 5.6% in 2001Q1 to 1% by 2003Q4. In the second phase, the Fed started raising rates in 2004Q2 up to 5.25% by 2006Q3. Commercial bank asset growth increased rapidly during the first phase and declined in the second phase, while shadow bank asset growth remained high in spite of seemingly tighter policy.

This is in line with a recent discussion by Woodford (2010), who points out that spreads were unusually low and credit growth was at its highest after the federal funds rate had already returned to high levels. He argues that the increase in the riskless short-term rate did reduce demand and checkable deposits of households and firms, but this did not prevent a net increase in the overall liabilities of financial intermediaries, including shadow banks, as shown in our Figure 1.

Figure 1. Pre-crisis US monetary policy and the expansion of banks’ balance sheets

Notes: Data are from the Bureau of Economic Analysis and the Flow of Funds. The definition of shadow banks follows Adrian et al. (2010), and it incorporates issuers of asset-backed securities (ABS), finance companies, and funding corporations.

Of course, Figure 1 does not necessarily imply a causal relationship, and at least three possible stories would be consistent with these observations:

  1. It might be that the drastic fall and the subsequent rise in the interest rate were the result of a positive and then a negative monetary policy shock that affected commercial (shadow) banks pro-cyclically (counter-cyclically).
  2. An alternative narrative is that the decline in the interest rate was an endogenous response to falling demand during the early 2000s, that monetary policy did not tighten fast enough following the sharp recovery, and that the rise in the interest rate was therefore an effective ‘loosening’ of monetary policy to which shadow banking responded pro-cyclically.
  3. Another explanation is that both the fall and rise in the interest rate were completely systematic responses of policy to other shocks such as financial innovation and regulation, which had an asymmetric impact on financial balance sheet dynamics, and monetary policy shocks have not been a driving force.

The impact of monetary policy surprises: Is there a ‘waterbed’ effect?

To provide a causal analysis of the relationship suggested by Figure 1, in Nelson et al. (2014) we use a Bayesian vector autoregression (VAR) framework to provide evidence pertaining to the effects of monetary policy surprises on the balance sheet growth of financial intermediaries, distinguishing their effects on commercial banks from those on entities in the shadow banking sector. Using VAR models we find that monetary surprises made only modest contributions to the asset growth of commercial and shadow banks during the 2000s. In line with the conventional view, monetary contractions tended to reduce commercial bank asset growth. But in contrast to the conventional view, we find that unexpected monetary contractions tended to expand shadow bank asset growth, rather than reduce it.

Figure 2. The impact of a 100 basis point monetary policy shock on commercial and shadow banks

Notes: The figure is based on the authors’ calculations. The vertical axes are in percentage points. The identification is with Choleski ordering. The sample period is 1966Q3–2007Q2. The impulse responses in the left column show the pointwise median, 32nd–68th, and 16th–84th percentiles of the posterior distribution. The historical contribution of monetary policy shocks in the right column is calculated based on the median estimates.

The left column of Figure 2 shows the impact of an unexpected 100 basis point increase in the federal funds rate on the size of the balance sheets of the two types of intermediaries. The impact on commercial bank asset growth is persistently negative and peaks at around −0.1% after one year. The policy shock has an immediate +0.2% effect on shadow bank asset growth.

The second column of Figure 2 shows historical contributions of monetary policy shocks to the variables of interest. The results suggest that monetary policy shocks were more important in the first half of the sample than during the Great Moderation. More specifically, monetary policy shocks had the largest impact in the late 1970s and the early 1980s. During the low interest rate environment in 2001–2005, policy shocks contributed positively to commercial bank asset growth, but shadow banking activity that expanded rapidly due to increasing securitisation seems to have been curbed by expansionary monetary policy shocks.

This result is robust across a number of model specifications and assumptions regarding the identification of monetary policy shocks. Our findings suggest that using monetary policy to lean against financial sector activity would be problematic. Its tendency to generate leakages through securitisation activity – a ‘waterbed’ effect – would make deciding on the correct sign of the appropriate monetary response non-trivial. Our paper also provides empirical evidence indicating that securitisation activity (proxied by the size of the balance sheets of asset-backed securities issuers, as well as government-sponsored enterprises and agency- and GSE-backed mortgage pools) tends to rise following unexpected monetary contractions.

However, given that the contribution of monetary policy surprises to the growth rates in this period ranged between −0.3 and 0.3 percentage points (see Figure 2), and considering that quarterly growth of real assets of commercial and shadow banks during the 2001Q3–2007Q2 period averaged about 1.5 and 2.3 percentage points, respectively, we argue that the overall importance of unexpectedly loose policy in the pre-crisis build-up was small relative to other contributing factors.

To build a theory to explain these empirical results, we extend the monetary dynamic stochastic general equilibrium (DSGE) model of Gertler and Karadi (2011) to include a shadow banking sector, as in Meeks et al. (2014). In the model, commercial banks originate loans to the private non-financial sector, before selling a portion of these securities to shadow banks. In turn, shadow banks bundle loans together, using them as collateral against which to issue asset-backed securities (ABS). These securities are held by commercial banks, whose creditors regard ABS as better collateral than individual loans. In this context, a monetary contraction tightens funding conditions for commercial banks, leading them to contract their balance sheets. But to mitigate that contraction, and therefore to maintain their profitability, commercial banks seek out pledgeable collateral in the form of ABS, in exchange for relatively illiquid loans.

Hence the model makes predictions regarding the effects of monetary policy shocks that are consistent with the evidence, lending force to the idea that the sign of monetary response needed to lean against financial sector leverage varies with the component of the financial sector in question.

Policy implications and concluding remarks

Our findings have important implications for the debate on the role of monetary policy in addressing financial stability concerns. This is relevant to both the academic debate on ‘leaning against the wind’ (Woodford 2010, Svensson 2013, Gambacorta and Signoretti 2014, Gali 2014) and the surrounding policy discussion (Bean 2014 and Stein 2013). One line of argument is that monetary policy is a powerful tool for tackling financial excess because it “gets in all of the cracks” (Stein 2013).

Our results are consistent with this claim – monetary policy shocks do seem to affect the balance sheets of both commercial banks and their unregulated counterparts in the shadow banking sector, albeit modestly. But our results point to an important caveat to that conclusion – a monetary contraction aimed at reducing the asset growth of commercial banks would tend to cause a migration of activity beyond the regulatory perimeter to the shadow banking sector. The monetary response needed to lean against shadow bank asset growth is of opposite sign to that needed to lean against commercial bank asset growth. That casts doubt on the ease with which monetary policy could be used in this way.

That reinforces the case made elsewhere (eg Hanson et al. 2011) for the development of macroprudential tools that address the build-up of leverage in the regulated sector more directly than monetary policy, and which extend oversight to the parts of the shadow banking sector that are most prone to excessive risk-taking (see Financial Stability Board (FSB 2013)). That would leave monetary policy to retain its relative focus on addressing the consequences of nominal rigidities in goods and labour markets (Svensson 2013).

Disclaimer: The views expressed here are those of the authors and do not necessarily reflect those of the Bank of England.

References

Adrian, T, E Moench, and H S Shin (2010), “Financial intermediation, asset prices, and macroeconomic dynamics”, Federal Reserve Bank of New York Staff Report 422.

Bean, C (2014), “The future of monetary policy”, Speech at the London School of Economics, 20 May.

FSB (2013), “Strengthening Oversight and Regulation of Shadow Banking”, Report, Financial Stability Board, 29 August.

Gali, J (2014), “Monetary Policy and Rational Asset Price Bubbles”, The American Economic Review 104(3): 721–752.

Gambacorta, L and F M Signoretti (2014), “Should monetary policy lean against the wind?: An analysis based on a DSGE model with banking”, Journal of Economic Dynamics and Control 43(6): 146–174.

Gertler, M and P Karadi (2011), “A model of unconventional monetary policy”, Journal of Monetary Economics 58(1): 17–34.

Hanson, G S, A K Kashyap, and J C Stein (2011), “A Macroprudential Approach to Financial Regulation”, Journal of Economic Perspectives 25(1): 3–28.

Meeks, R, B Nelson, and P Alessandri (2014), “Shadow banks and macroeconomic instability”, Bank of England Working Paper 487.

Nelson, B, G Pinter, and T Konstantinos (2015), “Do Contractionary Monetary Policy Shocks Expand Shadow Banking?”, Bank of England Working Paper 521.

Stein, J C (2013), “Overheating in Credit Markets: Origins, Measurement, and Policy Responses”, Speech at the “Restoring Household Financial Stability after the Great Recession: Why Household Balance Sheets Matter” research symposium sponsored by the Federal Reserve Bank of St Louis, St Louis, Missouri, 7 February.

Svensson, L O (2013), “Some Lessons from Six Years of Practical Inflation Targeting”, Stockholm School of Economics.

The Economist (2014), “The lure of shadow banking”, 10 May.

Woodford, M (2010), “Financial Intermediation and Macroeconomic Analysis”, Journal of Economic Perspectives 24(4): 21–44.

2,310 Reads