Recent Posts

Sunday, October 26, 2014

Book Review: Secular Stagnation: Facts, Causes and Cures

The Centre for Economic Policy Research (CEPR) has recently published the free eBook "Secular Stagnation: Facts, Causes and Cures" (edited by Coen Teulings and Richard Baldwin), which is a series of 13 articles by prominent economists on the topic of secular stagnation, along with a summary by the editors. (It is also available at Amazon for the Kindle, but it is not free in that format.) This review will act as a brief summary of some of the articles, along with some high level comments. I am currently writing a sequence of articles on "slow growth", and I will address some of the topics within the eBook at a later date in more detail.


The articles within the eBook are divided by theme, and are as follows.
  • Introduction Coen Teulings and Richard Baldwin
Opening the debate
  • Reflections on the ‘New Secular Stagnation Hypothesis’ Laurence H Summers
Three issues: Potential growth, effective demand, and sclerosis
  • Secular stagnation: A review of the issues Barry Eichengreen
  • The turtle’s progress: Secular stagnation meets the headwinds Robert J Gordon
  • Four observations on secular stagnation Paul Krugman
  • Secular joblessness Edward L Glaeser
Further on potential growth
  • Secular stagnation? Not in your life Joel Mokyr
  • Secular stagnation: US hypochondria, European disease? Nicholas Crafts
Further on effective demand
  • A prolonged period of low real interest rates? Olivier Blanchard, Davide Furceri and Andrea Pescatori
  • On the role of safe asset shortages in secular stagnation Ricardo J Caballero and Emmanuel Farhi
  • A model of secular stagnation Gauti B. Eggertsson and Neil Mehrotra
  • Balance sheet recession is the reason for secular stagnation Richard C Koo
  • Monetary policy cannot solve secular stagnation alone Guntram B Wolff
Further on sclerosis
  • Secular stagnation: A view from the Eurozone Juan F. Jimeno, Frank Smets and Jonathan Yiangou

Secular Stagnation: Background

The term secular stagnation was associated with the economist Alvin Hansen, which gained prominence in his 1938 Presidential speech to the American Economics Association - "Economic Progress and Declining Population Growth". The strong economic growth that started in conjunction with the Second World and in the following decades led to stagnation theories dropping out of sight. Larry Summers brought the idea back in a 2013 speech to the IMF forum.

The use of the term "secular stagnation" was marketing genius on Larry Summers' part. People had been writing about how this cycle was sluggish for some time; secular stagnation acts as a catchy summary phrase. But, as Barry Eichengreen observes,
But while the term ‘secular stagnation’ was widely repeated, it was not widely understood. Secular stagnation, we have learned, is an economist’s Rorschach Test. It means different things to different people.
Additionally, it is not particularly new, as I argued in an earlier essay,  It turns out that Larry Summers (amongst others) shares that assessment:
Imagine that US credit standards had been maintained, that housing had not turned into a bubble, and that fiscal and monetary policy had not been simulative. In all likelihood, output growth would have been manifestly inadequate because of an insufficiency of demand. Prior to 2003, the economy was in the throes of the 2001 downturn, and prior to that it was being driven by the internet and stock market bubbles of the late 1990s. So it has been close to 20 years since the American economy grew at a healthy pace supported by sustainable finance.
Given that many market commentators and participants tend to only look a few months ahead at a time, a great many people will be ready to proclaim "secular stagnation is dead" if the United States manages to string together a few quarters of strong GDP growth. Although the secular trend for growth is weak, the business cycle has not been abolished. As a result, such a burst of growth would not invalidate this interpretation of "secular stagnation" (although it would cast doubt on some of the formulations, such as those that attach importance to the zero bound of interest rates).

Interest Rates

From the perspective of a fixed income analyst, the most important implication of secular stagnation is the effect on the level of interest rates. The generally mainstream analysis presented within the book argues that secular stagnation implies that the real rate of interest needed to return the economy to full employment is negative, which appears impossible to achieve given the nominal rigidities within the economy (resistance to falling wages). The implication is that we will spending a lot of time with nominal rates at the zero lower bound.

Chart: 5-Year Treasury Rates, 5 Years Forward

If correct, this would explain the drop in forward rates shown above. In other words, the rally in bonds is not a "bubble", rather it is a belated catching up to reality by the bond market. Although I am sympathetic to this interpretation, I am cautious about Treasury valuation. Term premia could very easily reset higher in the face of Fed rate hikes.

I will discuss here two of the essays that focus heavily on interest rates.

In Larry Summers' article, he refers to the article "Measuring The Natural Rate of Interest" by Thomas Laubach and John Williams of the San Francisco Fed (link to letter which discusses the concept of the Natural Rate of interest, which also links to the article). The "natural rate of interest" is measured by their methodology by using a Kalman filter to estimate simultaneously the natural rate of interest, potential output, and the trend growth rate of output. The embedded assumption within this methodology is that deviations of the actual real Fed Funds rate from the natural rate will result in an acceleration of growth. This estimated natural rate of interest fell from 4.42% in 1961 to -0.07% in 2014 (based on updates published in a spreadsheet at the FRBSF).

The Krugman paper gives a less rigourous way of getting to the same general conclusion; he looked at average real Fed Funds rates across the cycle. This has tended lower since the 1980s. Once again, the embedded assumption is that the actual real Fed Funds rate cannot markedly deviate from the "natural" real rate for an extended period of time, without causing economic acceleration and/or inflation which we did not observe.

This is most troubling part of "secular stagnation" for the economic mainstream, which had become obsessed with monetary policy (and which assumed fiscal policy out of existence). And it certainly matters for bond investing. However, a non-mainstream viewpoint casts some doubt upon this assessment.

If the real economy is not particularly sensitive to the real rate of interest, then the asymmetric nature of central bank reaction functions implies that nominal rates will eventually crash to zero during a period of secular stagnation. What I mean by asymmetry is that the central bank will panic and cut rates quickly during a financial crisis (and there is invariably a financial crisis during a "modern" recession), but the slow growth during an expansion means that the rate cuts are not reversed before the next recession hits. Tim Duy, in "The Methodical Fed"  discusses how this realisation has featured in recent discussions about the return to the zero bound when the next recession hits (assuming that rates are indeed hiked before then).

From this point of view, low real interest rates are purely an artefact of central bank reaction functions, and not an inherent property of "secular stagnation". Given the moribund state of modern mainstream macroeconomics, I do not see those reaction functions changing, but I would caution that it is always possible.

Financial Instability

It has been known for centuries that investors act in a loopy fashion when confronted with low nominal interest rates on government bonds. ("'John Bull can stand many things, but he cannot stand two per cent.'" from Lombard Street by Bagehot, 1873). This is a clear worry of some of the authors within the eBook.

Although it is clear that many investors have been deranged by the JGB market, but I have some doubts about the causality between financial instability and low interest rates. The 1950s featured low interest rates, yet the financial system was stable. Hyman Minsky in particular discussed the reasons for that stability. However, the 1950s was a special period, and it would not be easy to replicate those conditions once again. Correspondingly, the low interest rate environment will coincide with financial instability, but that may just reflect the unwillingness to do anything useful about that instability.

Supply Side Arguments

Robert Gordon looks at the "supply side" story - potential growth - and has a different viewpoint.
As the US unemployment rate declines toward the normal level consistent with steady non-accelerating inflation, by definition actual output catches up to potential output. I have provided (Gordon 2014b) a layman’s guide to the numbers that link the performance of real GDP and the unemployment rate and have concluded that US potential real GDP over the next few years will grow at only 1.4 to 1.6% per year, a much slower rate that is built into current US government economic and budget projections. My analysis suggests that the gap of actual performance below potential that concerns Summers is currently quite narrow and that the slow growth he observes is more a problem of slow potential growth than a remaining gap [emphasis added - BR].
His earlier analysis that pointed to lower potential growth rates was often misinterpreted as being a commentary about technology. He emphasises that his view implies a continuation of the trends that have been in place for about 40 years.
My forecast of growth over the 25 to 40 years is measured from 2007, not from now. The sources of slow growth do not involve technological change, which I assume will continue at a rate similar to that of the last four decades. Instead, the source of the growth slowdown is a set of four headwinds, already blowing their gale-force to slow economic progress to that of the turtle. These four barriers to growth are demographics, education, inequality, and government debt. These will reduce growth for real GDP per capita from the 2.0% per year that prevailed during 1891-2007 to 0.9% per year from 2007 to 2032. Growth in the real disposable income of the bottom 99% of the income distribution is projected at an even lower 0.2% per year. 
Potential growth rates are often tied to productivity and technology. He argues that the emphasis on the question why productivity growth slowed since the 1970s is incorrect; rather the question is why productivity was high from the 1930s-1970s? Productivity growth both before and after that period is similarly low.

Note that Joel Mokyr takes the opposite view towards potential growth within his contribution.

In the aftermath of the Great Recession, many economists are persuaded that slow growth is here to stay. This chapter argues that technological progress – particularly in areas such as computing, materials, and genetic engineering – will prove the pessimists wrong. The indirect effects of science on productivity through the tools it provides scientific research may dwarf the direct effects in the long run. Although technological advances may polarise labour markets, they also bring widespread benefits that are not accurately reflected in aggregate statistics.
I expect to return to this topic in later essays, so I will offer only a few brief points. As Robert Gordon notes, the decades from the 1930s-1970s were special for growth, but I do not think they were the result of "lucky breaks" (as Gordon puts it), rather the result of policies that emphasised growth at all costs, especially during World War II. In peacetime, priorities are different. Although productivity trends are not supposed to affected by the cycle, the long term is a series of short terms, and productivity is cyclic. (This cyclical nature of productivity makes its use in business cycle models problematic.)

Since the 1970s, policy makers have made many changes to the structure of the economy based on the theory that they would address supply side issues and so increase potential growth. Since growth is slower now than it was then, I remain skeptical that policy supposedly driven by supply side concerns will augment growth..

Policy Remedies

Given the number of authors involved, there is a large number of policy implications discussed within the eBook. I will list a few that were included in the summary by the editors.
  • Monetary policy is less effective.  Since rates will be stuck at the zero lower bound a good portion of the time, the 1990s consensus that monetary policy can cure all macro ills is due for a rethink. One possible way of reducing the time at the zero lower bound is to raise the target inflation rate to 4%. Such a policy shift would matter a lot for bond markets, but I do not view such a change as likely. Firstly, a good portion of the electorate is inflation-phobic, and I am unaware of any pro-inflation major political parties. Secondly, it would be very difficult for central banks to hit a 4% target. They can easily keep inflation close to a 2% inflation by keeping excess capacity in the economy. Nominal rigidities will keep the inflation rate slightly positive. Hitting a 4% target would require much stronger control of the economy.
  • Fiscal policy looks more important. Richard Koo's essay is the most emphatic on this topic, but other authors now see fiscal policy as being more important than they thought earlier.
  • Infrastructure investment. The 1960s mainstream "Keynesian" bias towards infrastructure investment is once again fashionable. I developed a skeptical attitude towards infrastructure investment as a result of reading Hyman Minsky. As the Qu├ębec government discovered the hard way this decade, pretty well all structures (other than the Pyramids) require continued costly maintenance. Launching a badly thought out infrastructure programme is not going to do future generations any favours. And in the short term, infrastructure investment is capital intensive, and hence it will create little employment and probably exacerbate inequality.  
My point of view on the policy response is straightforward. Why do we care about slow growth? If I am wearing my fixed income analyst hat, it matters for asset market valuation and for actuarial assumptions. This is only of interest to certain people. But for policy makers, slow growth is worrisome because of the resulting unemployment. This suggests the correct policy response: create jobs. This can be done via something like the Modern Monetary Theory "Job Guarantee" programme.

I would argue that fiscal policy is not an area for misdirection and subtlety. If you want to accomplish something, do it. If you need to win a war, produce a lot of munitions. If you need infrastructure, build it. Conversely, building infrastructure you do not really need nor intend to maintain properly just because there may be some collateral job creation along the way is not a sensible policy.

Other Topics

As noted earlier, I am in the process of writing a series of articles on slow growth. I expect that I will return to some of the concepts discussed within this eBook at a later point. But I will now briefly comment on a few miscellaneous topics.
  • Edward L Gleaser looks at the labour market, and argues that the current welfare state may discourage employment. This is an old complaint of Minsky, although updated to current circumstances. A policy of direct job creation would remove the problems he observes.
  • There are two articles discussing the particular problems of the euro zone. I have not spent a good time looking at the euro zone economies in recent years, and so I am not too well placed to discuss those essays. 
  • Richard Koo discusses "balance sheet recessions" and secular stagnation. Since it balance sheet oriented analysis, it parallels my two part article on stock-flow norms and secular stagnation (link to Part 1). If you do not get too deep into theoretical distinctions, there is a good deal of analysis that sounds similar, as they are all related to the idea of "underconsumption". 
  • Gauti Eggertsson and Neil Mehrotra discuss a overlapping generation (OLG) DSGE model that exhibits the properties of "secular stagnation". This was one of the more theoretical chapters in the book, but it is presented without mathematical detail.

Concluding Remarks

This eBook provides a readable introduction to the problem of slow growth that is facing the developed economies. Readers who want more technical details will need to look at the academic articles that are summarised therein. Since there are a great many contributors, there is a variety of views expressed. This gives the reader a wider view of the problem, at the cost of having a less unified narrative.

See Also

(c) Brian Romanchuk 2014


  1. 'Secular Stagnation' is just a fancy description of what Kalecki describes in his famous essay. Best summary is here

    Monetary policy is all about misdirection and indirection. It seems that it's sole purpose is to prove how really, real clever all these people doing fancy mathematics are, rather than actually solve the problem.

    1. I forgot about that Kalecki quote, but it was pretty prophetic. I am planning in writing a longer piece on slow growth, and I will have to work that in. Thanks.

  2. I think there are two aspects that are largely ignored in most of these economists arguments -- the globalization trend means that most of these closed economy models are useless! and the digitization of the economy - the exchange economy will contract as measured by recorded transactions. In a digitized world - human needs are satisfied using a computer vs a car. Imagine the substitution that is taking place. During the 80's one had to go out of the house to see entertainment - so use a car, use gas etc etc. Today, I can sit at home and use the computer for all my entertainment! What a diffierence

    1. I will admit to thinking a lot in terms of closed economy models, but I tend to view the external sector acting as yet another source of drag on the economy - a form of automatic stabiliser. But for the mainstream, it's hard to work in the external sector, as you then have to incorporate two sets of optimising households. So they have little choice in the matter.

      The tend towards digitisation is interesting, as it messes up the logic of "production models" which are based on analogies to manufacturing. Also, it does allow for economic growth to decouple from resource use (although the servers acting as the backbone of the internet consume a lot of electricity).

  3. Is it possible that the US Federal Reserve as an institution is slowly being made irrelevant - in today's global economy? The ultimate outcome of free market economics. Why do we need an institution that has little policy value? The ZIRP is a natural trend for the US economy - as it is only economy capable of equilibrating global inefficiencies imbalances. Remember, the US Fed was not responsible for historical declining interest rates - they are part of the free market evolution. Greenspan said he had no control, he was only reacting to global forces!

    1. Although some central bankers may think they are setting rates where the "markets" want them (based on rate expectations), the reality is that people in the markets set rate expectations on what they assume the committee will do. To what extent the short rate is "market determined", it is based on markets beyond the rates markets - consumer prices, etc. If you take such a wider view, then yes, it is out of the central bank's hands.

      I wrote about this here:

      How important the central bank is a point of debate. As you note, by backing themselves into a zero interest rate policy, the Fed may have wiped out a lot of its effectiveness. But as time passes, conditions will change. If the U.S. housing market returns to a reasonable equilibrium, it will provide a strong "transmission mechanism" for interest rates to the broader economy.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.