Earlier last month the IMF’s Chief Economist, Olivier Blanchard, and two colleagues published a paper reflecting on some of the possible lessons for economic policy-makers flowing from the recent financial crisis and the ensuing global recession.
One of the issues which they raised was the fact that, in most of the major advanced economies, central banks couldn’t cut interest rates as much as they might have liked to because it’s not possible to reduce interest rates to below zero. Had inflation been higher before the onset of the crisis, they argue, interest rates would also have been higher, which “would have made it possible to cut interest rates by more”, thereby probably lessening the depth of the subsequent recession and the ensuing deterioration in governments’ fiscal positions. This leads them to wonder whether “policymakers should aim for a higher target inflation rate in normal times in order to increase the room for monetary policy to react” to large adverse shocks - not only from the financial sector, but from a pandemic, or from a terrorist attack.
Specifically, Blanchard and his colleagues pose the question, “are the net costs of inflation much higher at, say, 4 per cent than at 2 per cent?” They don’t actually answer that question. They acknowledge that higher inflation will have costs; but they ask whether those costs might be outweighed by the potential benefits of being able to reduce interest rates by more in the face of a shock if inflation had previously been maintained at 4 per cent rather than at 2 per cent. They don’t (and probably can’t) answer that question either.
Nonetheless, their paper has been widely interpreted as a recommendation on the part of the IMF (long regarded as a bastion of economic orthodoxy) that central banks should lift their inflation targets, a recommendation which was promptly endorsed by the Nobel Prize-winning economist Paul Krugman. And it assumes particular importance in the current context because many people are concerned that governments will see higher rates of inflation as offering a relatively painless way of reducing their outstanding debts to more sustainable proportions of their national incomes (as governments have often done in the past when confronted with very high levels of public debt).
There are at least two reasons why a considered answer to the questions raised by Blanchard and his colleagues should be in the negative - that is, that it would be a bad idea for central banks to target a higher inflation rate merely in order to give them more room to cut interest rates in the event of a major shock.
The first is that it is likely to be very difficult in practice to maintain a stable inflation rate materially higher than the “2 to 3 per cent on average over the course of the cycle” targeted by Australia’s Reserve Bank, or the slightly lower inflation targets pursued (implicitly or explicitly) by most other central banks in the industrialised world - or, in the language of the IMF paper, to “anchor” inflation expectations at a higher rate.
These inflation targets were chosen because, when inflation is around “2 point something”, people tend not to notice it. And when they don’t notice it, they tend not to do things to protect themselves against it which are likely to lead eventually to prices rising at a faster rate.
By contrast, when inflation is, say, 4 per cent or higher, experience amply demonstrates that people do tend to notice it - and they start to do things to protect themselves against its adverse consequences, such as seeking higher wages, or (in the case of businesses) putting up their own prices in anticipation of faster increases in their costs.
The inevitable result is that, sooner or later, inflation starts rising at a faster rate than 4 per cent: and the central bank is eventually obliged to raise interest rates to slow the economy sufficiently in order to bring inflation back down to 4 per cent again. But when it has done so (at some cost in terms of unemployment), people then start doing the same things to protect themselves against the effects on them of 4 per cent inflation that they were doing before.
In other words, a 4 per cent (or higher) inflation rate is unlikely to be sustainable in the way that a 2 or 3 per cent inflation rate has been. Instead, it is likely to result not only in inflation being more volatile, but also in economic activity being more volatile and (probably) slower on average.
The second reason why the question posed by Blanchard and his colleagues should be answered in the negative is that it isn’t necessary to have higher inflation in order to be able to bring interest rates down by more when confronted by a crisis.
Australia’s experience before and during the recent financial crisis bears that out. Australian interest rates were significantly higher than in most other industrialised economies prior to the onset of the financial crisis, even though the Reserve Bank’s inflation target isn’t significantly higher than the inflation targets pursued by central banks in other industrialised economies.
That was of course partly because growth in Australia’s economy had been fuelled, in the years leading up to the onset of the crisis, by the resources boom, in a way that hadn’t occurred in most other industrialised countries. But it also reflects the fact that Australia’s Reserve Bank had been more willing to tighten monetary policy in the years before the crisis than most of its peers elsewhere in the industrialised world. The US Federal Reserve kept its cash rate negative in real terms (that is, below the “core” inflation rate) from October 2002 (long after the mild recession which followed the “tech wreck” had ended) until October 2005 (by which time the US housing boom was only nine months away from its peak). Had the Fed (and central banks in other industrialised countries) not kept interest rates too low for too long in the first half of the past decade, the “bubbles” which preceded the financial crisis might not have got quite so out of hand; and the Fed and other central banks would have been able to bring interest rates down by more when those bubbles eventually burst.
The IMF paper actually contains many other useful thought-starters on the scope for using other instruments to promote financial system stability and reduce the risks of asset price bubbles. It also makes some very sensible suggestions about the use of fiscal policy. But the idea that central banks can, let alone should, target a higher rate of inflation, and that doing so would give them more room to deal effectively with another major shock, is not one that should be contemplated seriously.