Quantitative easing and unconventional monetary policy

Viv Davies: Hello and welcome to Vox Talks, a series of audio interviews with leading economists from around the world. I’m Viv Davies from the Centre for Economic Policy Research, it’s 21 November 2012 and I’m speaking with Professor David Miles, member of the Bank of England’s Monetary Policy Committee. We discuss Professor Miles’ recent research on quantitative easing and unconventional monetary policy. I began the interview by asking David to provide us with a brief summary of the paper and its conclusions.

David Miles: It’s a paper that really tried to step back and look at some of the recent empirical and theoretical literature on what the impact of quantitative easing might be, and try to see where we stand. I think that the main conclusion of the paper was that if one looks at the now quite large literature on what the impact of large-scale asset purchases is, I think there is a consensus emerging that those asset purchases by the Bank of England – the Fed in the US has done things on a similar scale – have had a positive on economies. They brought down a range of interest rates and had a positive impact on demand.

Different studies will give you different numbers for quite how powerful they are; the studies I’m most familiar with are those which look at the impact of purchases by the Bank of England. I think the average estimate would put the impact of the large-scale asset purchases done here in the UK at maybe the equivalent of boosting GDP by as much as a couple of per cent, and being the equivalent in terms of its impact of a very large cut in short-term nominal interest rates. Obviously those policies were embarked on by the Fed and the Bank of England, and also the ECB in its own way, once the interest rates had reached their lower bound, but you can nonetheless ask the question: what equivalent impact of cutting interest rates in ordinary times have asset purchases had? And the answer that most studies have come up with is that they’ve had an impact which is equivalent to a very large reduction in short-term interest rates, many percentage points.

VD: So it seems that there are a number of different types on quantitative easing or asset-purchasing programmes. Could you give us some insight in to what these are and why and how these programmes differ between central banks?

DM: One way in which they differ is in the nature of the assets that are purchased. Here in the UK, the Bank of England has overwhelmingly bought government bonds. In the US it’s been a mix of the Fed buying government debt but also buying securities. So one way to distinguish between different forms of asset purchases is whether the assets bought are issued by the private sector or by governments. In some ways the differences between what asset purchases happened in the US and the UK is not quite as great as appears at first sight, because nearly all the assets bought, nearly all the mortgage-backed securities bought by the Fed are actually insured one way or the other by quasi-government agencies. So they’re not quite pure private-sector assets.

The ECB has embarked on a slightly different strategy: its outright asset purchases are limited relative to the scale of asset purchases by the Fed and by the Bank of England, but they have nonetheless undertaken an enormous increase in the balance sheet, largely through so-called ‘repos’, that is long-term loans made directly to the banking sector in exchange for collateral. So they’re not asset purchases, but they have an impact which is in some ways comparable to asset purchases. You could think of a repo transaction, a long-term repo, as a decision by the central bank to buy some assets off the banking sector, but agree in advance the date and price at which you would resell them later. That’s a bit different from what the Fed and the Bank of England have done, where we’ve bought assets outright without a fixed date and certainly no fixed price at which we would resell them.

VD: Your paper was one of a number of papers on unconventional monetary policy that were published recently in a special edition of the Royal Economic Society’s Economic Journal. How important would you say the contribution of the research community is to developing our understanding of the mechanisms and impact of QE-type programmes?

DM: I think it’s very important. I know from my own experience here at the Bank of England, when we meet as a monetary policy committee each month, we have been updating and reassessing the evidence on the impact of asset purchases. In many ways it was a step in to the unknown for us in the Bank of England, and the same would be true for the ECB and the Fed, because we hadn’t undertaken asset purchases on this scale in recent history. There was a significant degree of uncertainty at the outset of quite how powerful a tool this was, and so the evidence that’s been amassed in the last three or four years on the impact of this has been something that here in the Bank of England we’ve been monitoring very closely. It’s been helpful for us on the Monetary Policy Committee in working out quite what the right scale of asset purchases should be.

VD: Is there a deeper existing literature on this, or has the current research been born out of the severity of the financial crisis?

DM: I think the scale of asset purchases we’ve seen from central banks in the last few years is unusual, and there are very few historical episodes to draw upon for assessing what impact those asset purchases might have. There is a large, perhaps theoretical literature on what the impact of changes in central banks’ balance sheets might be. Some of that literature gives you very clear-cut answers, but at the expense of making a set of assumptions that are not really applicable in the current environment. So for example there’s an extremely clear and interesting paper written seven or eight years ago by Michael Woodford and Eggertsson which shows that large-scale asset purchases by a central bank, either of government bonds or indeed of other assets, under certain conditions has no impact whatsoever. That’s an absolutely watertight theoretical result, but it only holds under extreme assumptions that the private sector can, in a sense, unwind the actions of a central bank by operating in perfectly efficient financial markets.

It’s a useful theoretical benchmark, but I think even under ordinary circumstances it’s of rather limited relevance. Certainly in the circumstances of the last three or four years, in the aftermath of a major financial crisis, the assumption that financial markets are operating efficiently and people are lending and borrowing freely in the private sector is an assumption that’s so far off the mark that that theoretical result is of very limited practical relevance. Though I think the very recent literature on the impact of asset purchases has been crucial for central banks assessing what the scale of the operation should be.

VD: So clearly there are risks associated with QE, not least the potential threat of uncontrollable inflation. What do you see as the main risks to QE, and how confident are you that those risks can be mitigated?

DM: The risk that large-scale asset purchases will generate very substantially higher rates of inflation is a risk that’s often overemphasised, I believe. The reason I think it’s overemphasised is that the reason why central banks and the reason why the Bank of England embarked on asset purchasing was precisely to try and hit an inflation target in an environment where the economic downturn was so severe that it looked likely that, without a more expansionary monetary policy, ultimately inflation might be driven beneath the target level and sit beneath it. Just as it was with an eye to longer-term inflation pressures that the Bank of England – and the same is true for the Fed – undertook the asset purchases, it will be the underlying inflation pressures in the future that will determine the rate at which the asset purchases are unwound or reversed. I don’t see any great technical difficulty in reversing quantitative easing.

When the time comes at the Bank of England we’ll sell the assets in much the same way as we bought them: by holding regular auctions. The only difference will be that instead of auctions in which the bank is buying assets they’ll be more conventional ones in the sense that we’ll be trying to sell something which is the gilts. The timing of the asset sales and the rate of asset sales will depend very much on the inflation outlook. I certainly don’t believe for one moment that the fact that the asset purchases were financed by the creation of money, bank reserves, that inevitably must mean that ultimately we will generate inflation. I think that ignores the fact that those transactions can and indeed will be reversed.

VD: Finally, David, how far do you think we can go with quantitative easing before there’s a danger that it morphs in to fiscal policy by another name? Is there a sense now that the boundaries are becoming rather blurred?

DM: I don’t think they are blurred, and it really is because the decision to undertake asset purchases has been made with a view to the broader economic environment and, in particular, the inflation pressures. Because the outlook for inflation was the main determinant for the scale of asset purchases, provided that remains the case I think that really makes the risk of what you may call permanent money financing by the central bank, and undermining of its independence, and a mixing of fiscal and monetary policy, the risk that that happens and that it inevitably means that inflation will end up being considerably higher is not a very substantial one. What’s crucial is that the central bank has in mind, as the main driver of its decision about asset purchases, what the underlying inflation pressures in the economy are.

In terms of the other aspect of your question – are we getting close to the feasible limits of quantitative easing – here in the UK it remains the case that the stock of outstanding government bonds that are available in principle for the Bank of England to buy, remains extremely large even after the Bank of England having bought £375bn of gilts. The stock of outstanding bonds available to buy is larger now than it was when we embarked on asset purchases back in 2009. That’s because the government has issued substantial new debt in the intervening period.

VD: David Miles, thanks very much for taking the time to talk to us today.