Back-of-the-envelope low-quality musings: #takes

Longform stuff I've put time and thought into is at #effortposts.

To hear from academics and policymakers, #interviews.

Compiled scribbles on what I read can be found in #notes.

Topic-related tags include #economics, #coronavirus and #miscellaneous.

# Why Bob Lucas Was Wrong

15 Nov 2020

I spend a lot of time talking about monetary theory and policy - but should I? More specifically, how large are the potential welfare gains from an optimal monetary policy regime compared to where we are right now? And how does this compare with optimising other sorts of policies, whether that is in growth, development or the many other areas which people have ranked as more impactful? I suspect it will be difficult to give a comprehensive and comparative account regarding the relative importance of different policy issues - but I will endeavour to at least shed some light on the first question i.e. the plausible gains from monetary policy being conducted better.

As perhaps the most influential macroeconomist of the late 20th century, Nobel Laureate Bob Lucas has made two famous statements which may shed some light on this. The first is his 1988 remark that “once one starts thinking about [questions of growth], it is hard to think about anything else”. The second is his comment, 5 years before the Great Recession, that the “central problem of depression-prevention has been solved, for all practical purposes”. Consequently, he saw the potential for welfare gains in improving monetary policy to be limited.

And intuitively, it is easy to see why this makes sense - growth is an exponential process, and so even small improvements in growth rates compound to enormous gains over time. By contrast, monetary policy is simply making sure the economy doesn’t undershoot or overshoot some rate of growth, with its benefits limited to reducing inefficient fluctuations around the rate of growth. Indeed, the long-run neutrality of money means that there is a reasonably level of consensus that the equilibrium growth rate is determined by real variables - as such, to the extent to which monetary can increase average growth, it comes from maintaining low and stable inflation. Consequently, Lucas suggested that tweaking monetary policy beyond that produced gains that were fairly negligible - that is, most of the important work in monetary policy has already been done. There are three reasons to doubt Lucas’s claims hold true today.

The first relates to the premise that monetary policy is only able to affect the variance but not the mean of the growth rate. For one, the fact that Lucas’s figures don’t accord with the way asset markets price volatility suggests that the gains from reducing macroeconomic volatility are larger than previously estimated. But more importantly, we assumed that long-run monetary neutrality meant an inability to change the average rate of growth. This is because we were implicitly operating with the natural rate theory, where there is some equilibrium trend rate of growth that the economy operates around, with fluctuations being symmetric. Here, the size of the boom predicts the size of the bust, since the economy must average the natural rate of output. However, one opposing view is that of Milton Friedman’s 1964 plucking model - rather than having some growth rate the economy can exceed or subceed, he saw a ceiling above which the economy could not produce. Fluctuations were not around the equilibrium rate, but rather downward drops - and much like plucking the string of a violin downwards, the size of the downwards movement (recession) predicts the size of the upwards return (recovery). This gives us testable predictions, and Friedman himself confirmed these predictions in a 1993 paper. This is particularly nice, in that there is no danger of overfitting the model, since Friedman would not have been able to know what would happen in the 30 years between his initial paper and the one afterwards. Since then, we have had more research confirming the plucking model as a more plausible characterisation, including a microfounded model by the inimitable Emi Nakamura. What her and her co-authors find is that the welfare gains increase by an order of magnitude compared to traditional models, because monetary policy can affect the average level of output and employment.

The second criticism is based on the changing state of the world. Lucas was writing in the time of the Great Moderation, where it seemed that the calamitous events of the Great Depression were firmly behind us, and where inflationary spirals had been controlled by Paul Volcker. To the extent to which business cycles existed, these were small and brief. However, the Great Recession has firmly punctured a hole in the idea that we wouldn’t face these sorts of depression-level risks - and although some may have believed that our experience with the 1930s and the 1970s meant we were ready for any future events, this is by no means guaranteed.

To understand how scarring these events can be even when we are operating with the policy wisdoms gathered across the last century or so, it is worth looking back to the costs of the Great Recession. The direct costs in terms of the output foregone, have conservative estimates ranging from 40% to 90% of a single year’s output. This is only made worse by hysteresis effects, where high unemployment can lead to skill atrophy, causing the level of potential output to fall. And to add to all of this, there are significant nonpecuniary costs - for example, the psychological burdens on wellbeing associated with mass unemployment meant that nonfarm payrolls falling by 6.3% did not only impact income. These large humanitarian costs of economic downswings are only exacerbated by the modern macroeconomic environment - whether this is financialisation leading to Minsky moments being more likely or the globalisation making global business cycles more concomitant, the modern world economy is one that is deeply fragile. And amongst all of this, we have seen the decline of $r*$, constraining the traditional mechanism of changing interest rates. So the idea that we have it all figured out is deeply implausible, especially if you are persuaded by the critiques leveled by the likes of the neo-Fisherites and Market Monetarists, who each posit something systematically wrong with the way we conduct monetary policy.

And finally, the third reason to care more about swings in the business cycle relates to the area of political economy. When we are thinking about the rise of populism and the harnessing of that towards malicious ends across the last 12 years, the Great Recession undoubtedly played a role. It did so by reducing the level of public trust and altering the perception of economic opportunity towards one that was more pessimistic. The result has been and continues to be clear - not only does this directly lead to increased hostility towards perceived outgroups, it has ushered in an age of epistemic crisis in journalism and fake news, as well as empowering the rise of amoral and opportunistic political actors who have put forward problematic policies.

All of this is not to say that monetary policy is more important than increasing growth and overcoming the sluggish productivity we face, or indeed helping out the most vulnerable in the world via changes in immigration or development policy. The thrust of Bob Lucas’s argument stands. But it is to suggest that he may have grossly underestimated the magnitudes, with economic crises remaining some of the most salient and pernicious things people can live through - consequently, we would do well not to forget about the importance of monetary policy in stabilising economies and avoiding these cataclysmic shocks.