Quantitative Easing and Financial Stability∗

The massive expansion of central-bank balance sheets in response to recent crises raises important questions about the effects of such “quantitative easing” policies, both their effects on financial conditions and on aggregate demand (the intended effects of the policies), and their possible collateral effects on financial stability. The present paper compares three alternative dimensions of centralbank policy — conventional interest-rate policy, increases in the central bank’s supply of safe (monetary) liabilities, and macroprudential policy (possibly implemented through discretionary changes in reserve requirements) — showing in the context of a simple intertemporal general-equilibrium model why they are logically independent dimensions of variation in policy, and how they jointly determine financial conditions, aggregate demand, and the severity of the risks associated with a funding crisis in the banking sector. In the proposed model, each of the three dimensions of policy can be used independently to influence aggregate demand, and in each case a more stimulative policy also increases financial stability risk. However, the policies are not equivalent, and in particular the relative magnitudes of the two kinds of effects are not the same. Quantitative easing policies increase financial stability risk (in the absence of an offsetting tightening of macroprudential policy), but they actually increase such risk less than either of the other two policies, relative to the magnitude of aggregate demand stimulus; and a combination of expansion of the cental bank’s balance sheet with a suitable tightening of macroprudential policy can have a net expansionary effect on aggregate demand with no increased risk to financial stability. This suggests that quantitative easing policies may be useful as an approach to aggregate demand management not only when the zero lower bound precludes further use of conventional interest-rate policy, but also when it is not desirable to further reduce interest rates because of financial stability concerns. ∗I would like to thank Vasco Cúrdia, Emmanuel Farhi, Robin Greenwood, Ricardo Reis, Hélène Rey, and Lars Svensson for helpful comments, Chengcheng Jia and Dmitriy Sergeyev for excellent research assistance, and the National Science Foundation for supporting this research. Since the global financial crisis of 2008-09, many of the leading central banks have dramatically increased the size of their balance sheets, and also have shifted the composition of the assets that they hold, toward greater holdings of longer-term securities (as well as toward assets that are riskier in other respects). While many have hailed these policies as contributing significantly to contain the degree of damage to both the countries’ financial systems and real economies resulting from the collapse of confidence in certain types of risky assets, the policies have also been and remain quite controversial. One of the concerns raised by skeptics has been the suggestion that such “quantitative easing” by central banks may have been supporting countries’ banking systems and aggregate demand only by encouraging risk-taking by ultimate borrowers and by financial intermediaries of a kind that increases the risk of precisely the sort of destructive financial crisis that had led these policies to be introduced. The most basic argument for suspecting that such policies create risks to financial stability is simply that, according to proponents of these policies in the central banks (e.g., Bernanke, 2012), they represent alternative means of achieving the same kind of relaxation of financial conditions that would under more ordinary circumstances be achieved by lowering the central bank’s operating target for short-term interest rates — but a means that continues to be available even when short-term nominal interest rates have already reached their effective lower bound, and so cannot be lowered to provide further stimulus. If one believes that cuts in short-term interest rates have as a collateral effect — or perhaps even as the main channel through which they affect aggregate demand, as argued by Adrian and Shin (2010) — an increase in the degree to which intermediaries take more highly leveraged positions in risky assets, increasing the likelihood of and/or severity of a potential financial crisis, then one might suppose that to the extent that quantitative easing policies are effective in relaxing financial conditions in order to stimulate aggregate demand, they should similarly increase risks to financial stability. One might go further and argue that such policies relax financial conditions by increasing the supply of central-bank reserves, and one might suppose that such an increase in the availability of reserves matters for financial conditions precisely because it relaxes a constraint on the extent to which private financial intermediaries The term “quantitative easing,” originally introduced by the Bank of Japan to describe the policy that it adopted in 2001 in attempt to stem the deflationary slump that Japan had suffered in the aftermath of the collapse of an asset bubble in the early 1990s, refers precisely to the intention to increase the monetary base (and hence, it was hoped, the money supply more broadly) by increasing the supply of reserves.