In the last few years, central banks have dramatically increased the size of their balance sheets and the maturity of their asset holdings through quantitative easing (QE) policies. These policies issue central bank reserves to buy long-term government bonds.
At first QE was justified as a response to the financial crisis of 2008-10 (Bernanke 2015). It was argued that arbitrage relations across government bonds of different maturities were disrupted so that purchases along the yield curve would have an effect on long-term interest rates that was beyond the usual effect of the overnight rate, controlled by the central rate. Moreover, with short-term interest rates at or close to zero, QE was a way to signal the central bank’s future intentions about the stance of monetary policy. Den Haan (2016a) covers these arguments in more detail.
In the US in 2016 interest rates are above zero, and rising. The financial crisis ended six years ago. If those were the only arguments for QE then they are no longer relevant. Conventional economic theory predicts that, outside of a financial crisis, QE should have no effect on real outcomes or inflation. Buying government bonds with central bank reserves merely shuffles the composition of overall government liabilities across maturities, and across their issuer. This theory suggests that any future changes in the size of the central bank balance sheet, up or down, should be of approximately no consequence. So, as some economists (e.g. Den Haan et al. 2016b, Ball et al. 2016) consider, might this be the end for QE?
Quantitative easing in a fiscal crisis
On the other hand, I my recent work I argue that there are two arguments for why QE will continue to have an effect on real outcomes, as well as on inflation (Reis 2016a). I use a model that merges new Keynesian costs of inflation, fragile banks, as in the research on credit frictions, a central bank balance sheet(from models of central bank insolvency), nominal debt that may be inflated away (from the fiscal theory of the price level) and sovereign default, as in the research on fiscal crises.
Importantly, the model does not have a binding zero lower bound on interest rates and the term structure hypothesis holds perfectly, so the arguments are separate, and therefore complementary, to previous research.
The crucial ingredient is a fiscal crisis. Following its recent experience, the ECB would attest that this is relevant in practice. But it seems odd to suggest that it is sensible monetary policy for the central bank to buy government bonds when the government is having trouble selling them. Surely, this will be monetary financing of the deficit. Also the debt offices already actively manage the maturity of government debt during a crisis, so the central bank’s intervention should not be necessary.
These two arguments against the use of QE in a fiscal crisis are orthodox, but wrong. They come from equating reserves with banknotes and short-term government bonds, respectively. Reserves are different. Their specialness makes QE a potentially effective tool in a fiscal crisis for a central bank mandated to control inflation and promote financial stability.
Quantitative easing and inflation
When a crisis hits, the public debt becomes too high to be paid with expected future fiscal surpluses. The government may refuse to default, insisting that, if the country has its own currency, inflation must debase the real value of the nominal debt.
With this fiscal dominance, the central bank is powerless to stop the overall increase in the price level. Monetary policy nevertheless has the tools to affect the time path of inflation. The sensitivity of the real value of nominal debt to inflation depends on both the maturity of the debt and the persistence of inflation. (Note that, in reality, Hilscher et al. 2015 developed a method to take these factors into account accurately, and reach the stark conclusion that it is very unlikely that the US would be able to inflate away its public debt.)
By using its control of the nominal interest rate, the central bank can control the path of expected inflation. By using QE to trade long-term bonds and short-term reserves, the central bank can control the effective maturity of government liabilities. Therefore, working together, interest-rate policy and QE can smooth out the path of inflation in a fiscal crisis. The central bank can guide the path of inflation over time. Because it is the time path of unexpected inflation that causes the relative-price distortions that are behind some of the most important costs of inflation, the central bank can discharge its valuable social responsibility of lowering the cost of inflation. QE makes this possible.
Quantitative easing and default
When governments default on public debt, this typically imposes a large shock to domestic financial systems. Reinhart and Rogoff (2009) and Reinhart and Kirkegaard (2012) note that even before default, fiscal crises come with large financial repression over the banking sector. Banks outside the US hold a large amount of public debt on their books. A sovereign default therefore triggers large bank losses. In turn, the resulting decline in bank capital typically means a cut in credit, which lowers real activity.
More generally, financial markets intensively use public debt as a safe asset that can be seized and pledged as collateral in financial contracts. During a fiscal crisis, the anticipation of default means that public debt’s value as a safe asset falls (Caballero and Farhi 2014). Some financial markets freeze because of the absence of collateral, and the credit flow to the real economy can become compromised.
Through QE, the central bank can help. When it buys government bonds from banks in exchange for reserves, the central bank is giving banks an asset that does not default. Reserves are the unit of account, so they cannot default in nominal terms. Only in very extreme cases would countries in a fiscal crisis renege on the liabilities of the central bank by implementing a currency reform. In most fiscal crises, reserves remain safe. By giving banks access to them, QE provides the banking sector with a shield against public default that can prevent a credit crunch. Because reserves remain safe, they may keep financial markets functioning. In a fiscal crisis, QE increases the supply of safe assets by absorbing risky government debt and providing safe bank reserves.
Reserves really are special
The power of QE comes from interest-paying reserves being a special financial asset with four distinct properties:
- Reserves are held exclusively by banks. Only they can hold these deposits at the central bank. Therefore, the central bank can make sure that the shield against default that reserves provides leads to an automatic recapitalisation of the banking sector after public default.
- Reserves are supplied exclusively by the central bank. Therefore, it can freely set the interest that must be paid on them. The joint control by the central bank of the quantity of reserves, and the remuneration of these reserves, give it the power to control the time path of inflation.
- Reserves are default-free. In nominal terms they always pay in full so, as long as the central bank remains solvent, they are always honoured. Therefore, they are a safe asset for the financial sector.
- Reserves are the unit of account in the economy. Therefore, their nominal value never changes, but their real value falls if there is sudden inflation.
As long as these properties hold, then using QE in a fiscal crisis is an effective policy. Moreover, these properties make clear that QE cannot be substituted by currency or government debt. QE is conceptually different from both monetary financing of the deficit and from Treasury management of the maturity profile of the public debt.
Monetary financing, or printing banknotes, creates inflation in itself and generates seignorage revenues. Reserves do neither, since in a market saturated with reserves the quantity of reserves outstanding has no direct link to inflation (Reis 2016b). And while banknotes are held by the whole of the private sector, reserves can only be held by banks.
Debt management authorities cannot replicate the QE actions of the central bank. They do not control the nominal interest rate, so they cannot control the path of inflation without the help of the central bank. In a fiscal crisis, public bonds cannot be made safe. Finally, their marginal holder is the private sector, not the banking sector.
Quantitative easing works in theory too
Ben Bernanke famously remarked that the problem with QE is that "it works in practice but it doesn’t work in theory". Quantitative easing has become an active policy tool for advanced central banks, but as opposed to the multiple theories proposed why nominal interest rates should affect the economy, there is still a lack of theoretical justifications for why QE may work. In a fiscal crisis, reserves and government bonds are far from being substitutes, but I have proposed two new channels through which may QE work in theory, too. In this case, QE can reduce the costs of a fiscal crisis by controlling inflation over time and reducing the distortions to the credit flow in the economy.
Ball, L, J Gagnon, P Honohan and S Krogstrup (2016), “What Else Can Central Banks Do?”, VoxEU.org, 2 September.
Bernanke, B (2015), The Federal Reserve and the Financial Crisis, Princeton University Press.
Caballero, R and E Farhi (2014), “On the role of safe asset shortages in secular stagnation”, VoxEU.org, 11 August.
Den Haan, W (2016a) “Quantitative Easing: A new VoxEU book”, VoxEU.org, 19 January.
Den Haan, W, M Ellison, E Ilzetzki, M McMahon and R Reis (2016b), “The future role of unconventional monetary policy: CFM Survey results”, Vox EU.org, 17 May.
Hilscher, J, A Raviv and R Reis (2014), “Will the US inflate away its public debt?”, VoxEU.org, 7 August.
Reinhart, C and J Kirkegard (2012), “Financial repression: Then and now”, VoxEU.org, 26 March.
Reinhart, C and K Rogoff (2009), This Time is Different, Princeton University Press.
Reis, R (2016a), “QE In the Future: the Central Bank’s Balance Sheet in a Fiscal Crisis” CEPR Discussion Paper No. 11381.
Reis, R (2016b). “Funding Quantitative Easing to Target Inflation” CEPR Discussion Paper No. 11505.