article

Secular stagnation? Be patient, Larry

In a recent speech to the IMF, Larry Summers kicked-off a collective outpouring of pessimism among the economics profession. The only thing reassuring is that when this many economists agree about a prognosis, it is usually wrong. Not because their analysis is flawed, but often because things happen which they don’t expect, or the data gets revised. It’s difficult enough to know what’s going on in a complex system such as the global economy. Even more so when the numbers we currently use for productivity, capital expenditure and even employment are subject to very significant retrospective adjustments.

That said, Summers’s argument is well-made, if not original (it’s sounds very like the arguments Stephen Roach was making a decade ago, or Albert Edwards, for what seems like several decades).

Summers’ starting point is a valid observation: there may have been a housing bubble, but there was no real evidence of overheating at the peak of the cycle. Despite a booming construction industry, and rapid growth in credit (and an oil price shock), underlying inflation didn’t pick up significantly and neither did wages (Fig 1). In a conventional inflation-targeting/output gap framework, interest rates were not “too low”.

Secular Stagnation chart Feb 14

Summers concludes with the observation that generating demand, not supply, seems to be the economic problem of the day. Bubbles are a symptom of this deeper problem, which the post-crisis consensus does not address. Limiting fiscal expansion, terminating QE, and restricting finance, all make sense in their own right, but what are the implications for demand? If it was already too weak, won’t the problem be greater going forward? Hence secular stagnation.

There are a number of gaps in Summers’ thesis. But they are easy enough to fill. Most obvious is: why? I am sympathetic to his “permanent demand shortfall” hypothesis, but he provides no cause. There are a number of plausible culprits. Demographic trends in the developed economies (particularly Japan and Europe) and Asia’s post-crisis balance of payments policies (effectively raising the national savings rate so as to accumulate FX reserves as an insurance policy) should have caused a major decline in global real interest rates. The secular increase in desired savings rates causes a trend decline in real interest rates, which undermines the efficacy of monetary policies based solely on changes in policy rates. Even in a conventional model, this shift in global savings preferences may have pushed equilibrium real interest rates into negative territory.

The next gap in Summers’s thesis concerns what to do about this. He has little to offer here. We will return to this in another blog, as there are many options that have not been tried and which don’t involve over-optimism about the government’s ability to build infrastructure, recurrent asset bubbles, or silly attempts to change “inflation expectations”.

But could Summers’s fear be fundamentally misplaced? I think so. I am not all convinced by current bearishness over corporate investment, real wage growth, and weak productivity growth post-financial crisis. There are very good reasons to doubt most of the data. Vast amounts of corporate capex is in fact occurring, but it is expensed. Much of this has little to do with animal spirits. For example, retailers today have to reinvent themselves (through investing), or face extinction. And technology is attacking an ever-broadening array of industries. Inflation is almost certainly lower than measured, which implies productivity, growth and real wages are all being underestimated. Contrary to the Luddite argument that the statisticians are cheating with quality-adjustments and re-weightings, the opposite is far more likely. That is what most of the academic literature shows, and so does common sense. Is a visit to the doctor today of equivalent “quality” to a visit 30-years ago? It may seem more rushed and less personal, but the chances of effective treatment are infinitely superior. What has happened to the price of movies or music, or education, when they are now freely available to anyone with an internet connection?

But the real counter-argument to all the calls for further aggressive cyclical activism is simpler: be patient, and be careful what you wish for. The global and US economies could grow more rapidly, but they are doing pretty well. The financial sector was in a colossal mess five years ago, and is now in rude health, as measured by capital and liquidity ratios and pricing power (at least in the English-speaking world).  The Eurozone is still completing its financial infrastructure, with perhaps inevitable short-term cyclical costs. More of the world is embracing free-market policies with tangible benefits. And technology is dramatically increasing competition, which will ultimately favour labour over capital, and increasing productivity in ways that are largely going unmeasured because the output lies outside the formal economy (think media – youtube, facebook, and hundreds of other content distribution processes – and perhaps most importantly, education).  Furthermore, more rapid growth may not be the blessing it seems. A prolonged period of slightly lower growth may be preferable to policy-induced over-optimism. China and India’s current economic challenges are a reminder of the perils of too much “success” in cyclical demand management.

History tells us the sure way to risk growth is monetary tightening. A prolonged phase of zero interest rates caused by central banks’ caution in the face of “modest” growth may well be the best way to ensure multi-year expansion.


The value of investments will fluctuate, which will cause prices to fall as well as rise and you may not get back the original amount you invested. Past performance is not a guide to future performance.