The Nature of Secular Stagnation

Tyler Cowen links to Scott Sumner’s skeptical notes on secular stagnation. As frequent readers know, I’ve been very skeptical in the past myself, but I think the argument has a lot more merit than Sumner gives credit for – i.e. that this is an open question that economists can more decisively accept or reject in ten years.

Sumner:

It seems to me that the Krugman/Summers view has three big problems:

1. The standard textbook model says demand shocks have cyclical effects, and that after wages and prices adjust the economy self-corrects back to the natural rate after a few years. Even if it takes 10 years, it would not explain the longer-term stagnation that they believe is occurring.

2. Krugman might respond to the first point by saying we should dump the new Keynesian model and go back to the old Keynesian unemployment equilibrium model. But even that won’t work, as the old Keynesian model used unemployment as the mechanism for the transmission of demand shocks to low output. If you showed Keynes the US unemployment data since 2009, with the unemployment rate dropping from 10% to 6.1%, he would have assumed that we had had fast growth. If you then told him RGDP growth had averaged just over 2%, he would have had no explanation. That’s a supply-side problem. And it’s even worse in Britain, where job growth has been stronger than in the US, and RGDP growth has been weaker. The eurozone also suffers from this problem.

The truth is that we have three problems:

1. A demand-side (unemployment) problem that was severe in 2009, and (in the US) has been gradually improving since.

2. Slow growth in the working-age population.

3. Supply-side problems ranging from increasing worker disability to slower productivity growth

Only the last two can explain the slowing long run trend rate of RGDP growth, as well as the low real interest rates on 30 year T-bonds.

Both (1) and (2) don’t get to the heart of what Krugman/Summers believe and are implicitly embedded with false assumptions. The “standard textbook model” is a nebulous concept. Even within many of the workhorse, New Keynesian, models the economy will not return to full employment without some external push if the natural rate to which it must correct is below 0%. This is very different from the old Keynesian permanent slump, which suggested that if wages are flexible, downward adjustment would further increase the real interest rate and tempt a vicious, deflationary cycle.

Krugman has definitely agreed before that to the extent that falling prices are expected to be temporary, they can also be expansionary.

More importantly, disregarding the fact that a fall in unemployment to 6.1% doesn’t adequately capture the whole labor market, if you showed Keynes a chart of US unemployment since 2007, he wouldn’t have expected rapid growth if you also told him inflation has consistently been well-below its long-term average every year thereof. Sumner’s suggestion that productivity growth is to blame (while perhaps true) does nothing to change the fact that, if anything, declining supply-side fundamentals should have increased the price level.

In fact, that’s specifically the conundrum that secular stagnation attempts to address. It’s not a comment on the financial crisis and recession as much as a meditation on the decade that preceded it. How could a rapid increase in wealth, credit creation, construction, government spending, and accommodative monetary policy not result in above-average inflation?

John Taylor loves pointing out that the housing bubble was caused by discretionary policy that deviated from his eponymous rule. Not only does that ignore the fact that the target FFR pretty much followed the implied Taylor Rule if one looked at the Fed’s inflation forecast instead of current inflation (which is both noisy and laggy), but also assumes that the equilibrium rate of interest was constant over this period. That deviation from Taylor’s version of the Taylor Rule begat a housing bubble , and not any inflationary pressures suggests this is false.

So if you told Keynes that the unemployment rate had been cut in half without any associated increase in inflation, he’d probably ask you to look at a different metric.

More peculiarly, Sumner seems to be missing the Krugman/Summers point entirely by prescribing policies that they would both advocate:

I mentioned that there was a third problem with the Krugman/Summers view. They favor big government Keynesian demand-side remedies for what they see as a sort of permanent liquidity trap. This fits with the newly fashionable anti-neoliberal views on the left. Thomas Piketty’s new book made the wildly implausible claim that neoliberal reforms had not helped countries like Britain. However the countries least likely to be mentioned in discussion of “The Great Stagnation” are precisely those countries that have pretty good supply-side fundamentals, and/or relatively small government. Here’s the Heritage Foundation’s list of the top 10 countries for Economic Freedom […]

Now I don’t want to oversell this list. Many of the top 6 countries have fast population growth. It’s hard for any country to completely overcome the slowdown in the rate of global productivity growth. But I think any fair observer would note that (with the exception of Ireland) the “usual suspects” in the stagnation discussion (Japan, the US, Britain, the 18 eurozone members, etc) are conspicuously missing from that list. And while Ireland undoubtedly was hammered by a big demand shock, their RGDP rose 7.7% over the past 12 months, a rate the US could only dream about. So while the top ten countries are not perfect (Denmark’s performance has been mediocre) they’ve clearly done better than most developed countries. That doesn’t provide much support for the progressives’ claim that the eurozone is doing really poorly because while they have the biggest governments on Earth, their governments need to be even bigger to overcome the Great Stagnation.

There’s pretty much nothing exclusive in those two paragraphs with what Summers and Krugman believe, except perhaps for any credibility assigned to the Heritage Foundation. If you read Summers’ introduction to the definitive ebook on the subject, he notes the following as some of the most promising avenues out of secular stagnation:

• “Removing barriers for labor mobility between firms by trimming down employment protection legislation.”
• “Increasing incentives for low-skilled workers to participate on the labor market.”
• “Simplifying procedures for starting up businesses.”
• “Applying anti-monopoly policies to reduce the profit margins in new IT industries”.

In fact, the central thrust of the secular stagnation argument is that falling productivity growth, a chief determinant of the prevailing natural interest rate, has fallen to a point where there is a vicious cycle between lower potential growth and lower actual growth.

The comments on the Eurozone are also tough to confirm. While Germany was definitely the success story of the decade, it worked largely because the rest of the world was able and willing to absorb its excess savings, thereby masking instability in the periphery. Instead, the real question is why did the periphery grow so slowly, despite a rapid inflow of capital from Germany and subsequent fall in borrowing rates.

The German model, by the unfortunate but inevitable requirement that international assets and international liabilities must sum to zero, could not have been applied everywhere to similar effect.

This post mischaracterizes what progressives actually believe. The conversation on secular stagnation most certainly does not, at least by itself, advocate “anti-neoliberal” viewpoints. If anything, this is a much stronger case for supply-side reforms than the Reaganites were ever able to produce in the 80s.