Friday, June 06, 2014

Secular Stagnation Part 1 - Paul Krugman's Bicycling Problem

"What’s really happening fast is the demographic transition, with Europe very quickly turning Japanese."
Paul Krugman - For Bonds, This Time is Different



Ever since Larry Summers gave his game-changing speech at last autumn's IMF research conference the back-and-forth flow of arguments about secular stagnation has been almost non-stop (indeed Larry himself now has a webpage dedicated to the topic). First to dive into the swimming pool after the sounding of the starting pistol was Paul Krugman, with a series of blogposts and NYT articles (here, here, here, here, here, here, and here).  These were followed/accompanied by a series of commentaries (both for and against), and then finally we got to one of the potential end points of the argument, the "negative natural interest rates for ever and ever" model produced by Gauti Eggertsson and Neil Mehrotra (here) - summarised by Krugman in his memorable "Stagnation Without End, Amen" post.   A fuller bibliography of major milestones in the secular stagnation issue can be found at the foot of this post, but I'm going to eat up most of the column space here looking at just one Krugmans arguments - the one contained in Demography and the Bicycle Effect - since in many ways in goes right to the nub of the issue.

The background to the argument (as originally, even if prematurely, explored by Alvin Hansen, Günar Myrdal and even John Maynard Keynes himself in the 1930s) is that the population dynamics set in motion by the industrial revolution of the late 18th century have reached a historic turning point. In the two centuries or so that have elapsed since what many term the modern growth era got going three related but distinct processes co-existed in time:

1/ positive trend population growth
2/ positive trend economic growth 
3/ steadily accelerating technical change

You could call these the "stylized facts" which characterize the modern growth era, but the secular tendency in two of them  is about to undergo a seismic shift. At some stage during the 21st century global population will peak, and then gradually start to decline, probably forever more. In fact in some countries (principally in Eastern Europe, but also Japan, Germany, Spain, Portugal, and Greece) population is already falling. All of Europe will likely head in this direction sometime in the 2020s, although there is considerable uncertainty still about the actual path dynamics of this process since in addition to birth rates immigration rates also play a part. At the present time some countries in Europe (the UK, Germany, Switzerland) are in receipt of large numbers of migrants annually, while others are losing working age population precisely to the aforementioned trio.

Hence the significance of the fact that Paul Krugman uses the expression "demographic transition" (for the first time to my knowledge) in the quote which opens this post. We are not talking here about some one-off problem (although often observers have spoken about Japan in just these terms), but a generalized phenomenon, a transition, something which eventually will affect all countries on the planet and our entire species. Previously people have tended to use the expression "demographic transition" to refer to the increase in the proportion of working age population and total population that accompanies the drop in fertility from high levels in less developed economies. The expression "demographic dividend" has often been used to describe the boost to economic activity this shift entails. Normally people assumed that this process would come to a halt around the 2.1tfr replacement level, but in one developed economy after another this hasn't happened, and those in which it has have been more the exception than the rule. So now reality pushes us towards a broadening in the definition of that transition towards acceptance of a later phase wherein populations age, and ultimately decline, a process which is greatly accelerated in those countries which have experienced long term very low fertility.

What Alvin Hansen and others started to think about in the 1930s was what the consequence would be for the second of the secular process which have characterized the modern growth era  (positive economic growth). What happens if populations (or better put working age populations) start to shrink? The kernel of their argument is summed up (here) by Paul Krugman as follows:
"To have more or less full employment, we need sufficient spending to make use of the economy’s potential. But one important component of spending, investment, is subject to the accelerator effect: the demand for new capital depends on the economy’s rate of growth, rather than the current level of output. So if growth slows due to a falloff in population growth, investment demand falls — potentially pushing the economy into a semi-permanent slump."
Pretty simple really, but isn't this just what Keynes says at the start of the General Theory, sometimes the simplest things are the hardest to see. Especially if our natural intellectual disposition leads us towards assuming the opposite. Apart from the theoretical simplicity, the idea is backed by plenty of empirical evidence. It has become clear in one country after another that there is a steady falling off of economic growth after the rate of working age population growth peaks and then starts to decline.


The Japan case (as shown above) is clear enough, but the start of this process is already observable in China, where working age population is currently peaking, and where growth rates have now fallen from the earlier double digit levels to ones in the 6%/7% range. People are even alarming themselves by saying we'd better get used to the idea of 5% growth, but in fact this easing in growth rates has little to do with a housing slowdown. Rather it is structural and long term, and Chinese growth rates will eventually fall to the level of Japanese ones (see my 2008 "Has China's Economic Growth Now Passed It's Peak"). Or does anyone seriously think China can keep going on the basis of attracting immigration?

What we are seeing then is that as working age population growth drops towards zero, so GDP growth weakens. The question is, as it turns negative will GDP growth (as opposed to GDP per capita growth) also turn negative. The answer is it depends. A simple approximation to a growth accounting model of the type used by IMF, OECD etc to calculate trend growth in an economy would be the following:

GDP growth = growth in working age population + TFP (total factor productivity) growth

Now, if working age population growth turns negative, then GDP growth can only be positive if productivity grows faster. One conclusion is very obvious here, handling the later stages of the demographic transition is all about managing the rate of working age population decline. This can be achieved to some extent via lengthening the working life, raising the participation rate, or having immigration. Even so, we may arrive at a point were GDP trend growth rates drop below zero. This situation is very near to being the case in Japan, Italy and Portugal, among others.



So we could be moving to a world were eventually both population and GDP decline. What about technology? Well this will need to be the subject of a separate post, but it is highly likely that the rate of technical  change will continue to accelerate, so we could have the peculiar situation that while GDP notional falls, and keeps falling, materially we feel a lot better off.

Well, after this excursus, let's go back to Paul Krugman's bicycling issue. Basically Paul has started to run into a problem that Claus Vistessen and I ran into on our Demography Matters blog. If the world is overpopulated, and there is a constant pressure on natural resources, then why are you economist types worried about falling populations? Wouldn't it be a positive for the planet?
"whenever I raise these points, I get questions from people who ask why I don’t regard slowing population growth as a good thing. After all, it means less pressure on resources, less environmental damage, and so on".
Now naturally at the moment in the US we are talking about slower population growth, but in Japan and parts of Europe - and eventually as we have seen probably everywhere else - population and working age population are already falling. But still, if slowing population growth is going to be a problem for an economy, then sure as hell falling population could be.
"What’s important to realize, then, is that slower population growth indeed could and should be a good thing — but that what passes for sound economic policy is all too likely to turn this potentially good development into a major problem. Why? Because under the current rules of the game, there’s a strong bicycle aspect to our economies: unless they’re moving forward sufficiently rapidly, they tend to fall over."
So what's the issue here? What he calls the current rules of the game (is it in our power to change them?)? Or the bicycle effect, which implies that we are perpetually "condemned to grow", since as Paul says, if our economies don't move forward fast enough they tend to fall over. This is what I intend to examine in the posts which will follow.

To close this introduction to my series on Secular Stagnation I would like to cite the end of the Keynes speech I mention above (here):

"A too rapidly declining population  would obviously involve many severe problems, and there are strong reasons......why in that event, or in the threat of that event, measures ought to be taken to prevent it. But a stationary, or slowly declining population may, if we exercise the necessary strength and wisdom, enable us to raise the standard of life to what it should be, whilst retaining those parts of our traditional life which we value the more now we see what happens to those who lose them"

"In the final summing up, therefore....... I only wish to warn you, that the chaining up of one devil may, if we are careless, only serve to loose another still fiercer and more intractable."

Conclusion

In this introduction we have seen that population decline may now well be inevitable, as may a turning negative of economic growth. This need not make us poorer, depending on how we:

a) develop technology
b) manage the process

As Paul Krugman warns, the current economic and financial "set up" implies the bicycle needs to constantly move forward, something which may prove more and more difficult to achieve. We therefore need as Keynes said, to exercise the necessary strength and wisdom. That is to say we need to adapt the current rules of the game to the new reality, and learn to manage the process. Is simply blowing bubbles the best way to handle this sort of epochal change? That is the topic we will turn to next.

Bibliography

 E. Cary Brown: Alvin H. Hansen's Contributions To Business Cycle Analyisis, 1989 (here)

Willem H Buiter: The Simple Analytics of Helicopter Money, Why It Always Works, Cepr 2014 (here)
Willem H Buiter, Ebrahim Rahbari & Joe Seydl: Secular Stagnation: Only If We Ask For It, Citi Global Economics View January 2014

Gauti Eggertsson & Neil Mehrotra: A Model of Secular Stagnation, 2014 (here)

Alvin Hansen (1939), “Economic progress and declining population growth,” American Economic Review.

John Maynard Keynes: Some Economic Consequences of a Declining Population, Galton Lecture, 1937 (here)

Paul Krugman: Inflation Targets Reconsidered, Sintra May 2014 (here)

Paul Krugman: For Bonds, This Time is Different, NYT 2 June 2014 (here)
Paul Krugman: Demography and the Bicycle Effect, NYT 19 May 2014 (here)
Paul Krugman: Secular Stagnation in the Euro Area, NYT 17 May 2014 (here)
Paul Krugman: Stagnation Without End, Amen (Wonkish), NYT 9 April 2014 (here)
Paul Krugman: Monetary and Fiscal Implications of Secular Stagnation, NYT 19 November 2013 (here)
Paul Krugman: A Permanent Slump?, NYT 17 November 2013 (here)
Paul Krugman: Secular Stagnation, Coalmines, Bubbles, and Larry Summers, NYT 16 November 2013 (here)


Paul Krugman: Monetary Policy In A Liquidity Trap NYT 11 April 2013 (here)
Paul Krugman: The Japan Story, NYT 5 Feb 2013 (here)
Paul Krugman: Japanese Relative Performance, NYT 9 February 2013 (here)

Paul Krugman: Japan: What Went Wrong? June 1998 (available in Japan section here)
Paul Krugman: Further Notes On Japan's Liquidity Trap, June 1998 (available in Japan section here)
Paul Krugman: It's Baaack: Japan's Slump and the Return of the Liquidity Trap, Brookings Papers on Economic Activity, 2:1998 (here)

Gunnar Myrdal: The Effects of Population Decline. Godkin Lectures, Lecture VI, 1938 (here)


Masaaki Shirakawa (2014):Is inflation (or deflation) “always and everywhere” a monetary phenomenon?, BIS (here)
Masaaki Shirakawa (2012): Demographic Changes and Macroeconomic Performance, Bank of Japan  (here)

Larry Summers (2014):On secular stagnation (here).

Timothy Taylor: Secular Stagnation: Back To Alvin Hansen, 2013 (here)

3 comments:

Abu Daoud said...

Thanks, and I'm looking forward to the next part of the article! Please don't make us wait too long...

Anonymous said...

Interesting but you have a silly premise: you take Paul Krugman seriously. Anybody with basic knowledge of economics knows that Krugman is at best a clown and possibly a fraud.

Janet Baker said...

Regarding this statement:


"And one last closing point. I have long favored (and continue to favor) sustainable immigration as one way to manage population aging (see my Message To Central Bankers Target Median Ages!: 2006, or my reply to Marty Feldstein "The Effects of the Ageing European Population on Economic Growth" of the same year). But for such a policy to work as intended we need to stop trying to reflate economies which for demographic reasons are trying to disinflate, and develop policies which are appropriate to the times we live in. "

But Mr. Hugh HAS gone further, at least once, I remember well his earlier statement (somewhere! But it made me smile is how I remember it) that 'it wouldn't kill the Spanish to have some babies.'

It breaks down that simply. Just as those popes who warned us about the consequences of contraception said it would.