There is one very bad, possibly dishonest, argument for austerity today, and that is the idea that cutting government spending will have no impact on the economy, even though we are in a liquidity trap. I normally attempt to debunk this argument by going through basic macroeconomic theory to show how, in a recession with unchanged interest rates, cutting government spending is bound to reduce aggregate demand and output. (See Wren-Lewis, S (2011), Lessons from failure: fiscal policy, indulgence and ideology, National Institute Economic Review vol. 217 no. 1 R31-R46. For those who like their theory state of the art, see Woodford, M. (2011), ‘Simple analytics of the government expenditure multiplier’, American Economic Journal of Macroeconomics, 3, pp. 1–35.) The data backs up the theory when the empirics are properly done, as this nice survey by Christina Romer shows.
However I have in the past had more respect for the following argument, which we might call the case for ‘precautionary austerity’. Although there are good reasons why a rapid reduction in government debt is unnecessary, financial markets do not always behave rationally. There is a chance that markets might suddenly panic, and stop buying government debt, forcing up interest rates. As the cost of such an outcome would be very high, macroeconomic policy should do everything it can to avoid it – including reducing debt rapidly - even if that meant deepening the recession.
That is the argument. I was never convinced by it, partly because I think this risk is pretty small. It is certainly wrong to use the Eurozone crisis as evidence otherwise, as I’ve argued here. I also agree with a recent post by Jonathan Portes, that the musings of the credit rating agencies tell us nothing about market sentiment. However Jonathan’s post also made me wonder whether the precautionary austerity argument was simply wrong.
Jonathan and others make the argument that there is no way the UK (or US) government will ever default, because they will always prefer to make their central bank buy debt through printing money. While this seemed logical, I had worried about what would happen before this point. It is unlikely that the markets would stop buying debt at any price overnight. Instead, demand for UK government debt would be positive but less than supply, and so interest rates would begin to rise. If interest rates got quite high before the central bank was forced to print money, damage could still be done.
Quantitative Easing (QE) is the process by which, in the UK, the Bank of England buys government debt in an effort to reduce longer term interest rates. At the moment, every month the Monetary Policy Committee reviews not just the level of short term interest rates, but also the scale of QE. Suppose, therefore, that interest rates on government debt did start rising because the market started to panic, and yet the economy remained depressed and the outlook for inflation was benign. In these circumstances the Bank would buy government debt as part of its inflation targeting strategy. There would be no need for the government to instruct it to do so. In principle the same logic would apply to the US. If the central bank acted quickly in this way, any significant increase in interest rates would be counteracted by aggressive action by the central bank. Of course this action would involve printing money on a massive scale, but it would be temporary, and so it would not be inflationary (see here).
We do not need to worry about a market panic because of Quantitative Easing. In a sense it is the counterpart to the argument that we have a Eurozone crisis because the ECB is so reluctant to act as a lender of last resort to governments. We can go further. Even when markets are behaving rationally, they don’t just worry about the fundamentals (in this case the chances of default), but also what other market participants think. If you think there are enough people in the market that might panic, it is rational for you not to buy. But if you also know that there is a residual buyer who will never panic (the central bank), you can just focus on the fundamentals, which in this case includes how long QE will exist as an option. That in turn depends on the outlook for the economy. So QE is not only the fire engine that will put out the fire, it also reduces the chances of a fire occurring in the first place.
The fire engine analogy is often extended to include moral hazard: the existence of the fire engine makes house owners less cautious, so fires are more likely. But Quantitative Easing will disappear once the recovery is secure. The central bank will sell back all the debt it now owns, as Japan did in 2006. For this reason it does not remove the need for governments to control debt in the longer term. Instead it just allows them to be more flexible in the short term, while we are at the bottom of a severe recession.There have been a number of studies trying to assess how effective QE has been in keeping long term interest rates low. However perhaps it has a more important precautionary role, which is to eliminate the possibility of a self-fulfilling panic in the government debt market. But if QE does this, why do we need deep austerity now to placate the markets?