Thursday, 26 February 2015

Lawrence Summers’s “secular stagnation” is complete nonsense.

Summary.    Secular stagnation is the idea that even at zero interest rates, it’s possible an economy does not achieve full employment. Ergo negative interest rates are needed, but implementing the latter is difficult. Ergo it’s conceivable there is no escape from excess unemployment.

The truth is that as Keynes pointed out and as MMTers keep repeating, it doesn’t matter how reluctant businesses are to invest or how reluctant households are to spend, if the state simply increases and carries on increasing the amount of money spent and fed into household pockets, the point must eventually come where households react by  spending enough to bring full employment. And until household spending rises far enough, there is no theoretical limit to the latter public spending.

As leading MMTer Warren Mosler put it in his “Mosler’s law” which appears at the top of his blog: “There is no financial crisis so deep that a sufficiently large tax cut or increase in public spending cannot deal with it.”


Summers first proposed his secular stagnation idea in a speech at an IMF conference in 2013. Mostly it’s incoherent nonsense far as I can see, but if Summers is saying anything at all, I go along with the summary of his speech set out by Gavyn Davies in the Financial Times. As Davies puts it in his 2nd and 3rd paragraphs, the theory is that demand can decline to such an extent that even a zero interest rate won’t solve the problem, thus a NEGATIVE rate is needed, and allegedly because cutting interest rates “has been the only means available to boost demand”.

Now the first flaw there is that cutting interest rates is most certainly not the “only means available to boost demand”. That is, if a zero rate doesn’t bring full employment, the state can simply print money and spend it, and/or cut taxes. The effect of that is to boost household cash balances, and (if the increased public spending option is taken) to increase employment in all the usual public sector areas: education, health, infrastructure repair, law and order, defence and so on.

And if the latter policy is implemented in robust enough form and for long enough, then household cash balances must at some point induce households to spend enough to bring full employment.

Indeed, the latter is exactly what we’ve done over the last three years or so. That is we’ve implemented fiscal stimulus (i.e. have government borrow and spend (and/or cut taxes)), then we’ve had central banks print money and buy government bonds: that’s called “Quantitative Easing”. And that comes to the same thing as having government and central bank, i.e. “the state”, print money and spend it and/or cut taxes.

Perhaps Summers hasn’t heard of QE. Or if he has, it seems he doesn’t understand the basic central bank book keeping entries involved when central banks do QE.

Two years later: 2015.

Having briefly sumarized Summers’s ideas as of 2013, we’re now in 2015 and he seems to have learned nothing in the meantime. In this speech given a few days and entitled “Reflections on Secular Stagnation” he says:

“Go back to basic Keynesian economics, and imagine that the point where the IS curve coincides with full employment involves a nominal interest rate that is lower than the attainable nominal interest rate. In that case, the creation, the printing of more money will be unavailing in generating economic growth.”

What on Earth is he talking about? Robert Mugabe didn’t find the “IS curve” any problem when he was printing and churning out ludicrously large amounts of money. It would be nice if Harvard economists had the same grasp of this subject as Robert Mugabe, wouldn’t it? (As I pointed out here some time ago).

And later in his speech of a few days ago he says:

“Secular stagnation is the phenomenon that the equilibrium level that savings are chronically in excess of investment, at reasonable interest rates.”

Well obviously it’s possible there is a decline in the amount that firms want to invest, and obviously its also possible there is a rise in the desire by households to save – in particular save MONEY rather than save in the sense of acquiring bigger houses, newer cars, etc. But the solution is easy: GIVE PEOPLE MORE MONEY!

Of course if the latter process goes too far, then excess inflation ensues. But until that point is reached or looks as though it’s about to be reached, there’s nothing wrong with simply printing money and expanding public spending and/or cutting taxes.


  1. Hi Ralph,

    I'm probably driving you crazy with less-than-positive comments here but here I go again...

    First, I agree that in this crisis, more federal borrowing/spending/tax-cutting would have been a very good thing, and the political opposition to it was either just wrong-headed or possibly very nefarious. Bad.

    But I also don't agree with the strong statements you've made here. Do you believe there is no limit to the amount of debt the US government can take on? Both Krugman and DeLong have recent posts about how public debt can become a problem, even for a government borrowing its own currency:

    As Krugman says, "too much government debt is a strange thing to be worrying about now." He's probably right about that, but he is at least implying that it would be a very reasonable thing to worry about if circumstances were a little different. You have made a (much) stronger claim than Krugman, by claiming that there is no limit to the amount of borrowing/spending a government could/should do to plug an output gap. I think Krugman is probably right here.

    DeLong is more straight forward than Krugman about the limitations of government borrowing/spending in the long run. He points out how bad excessive government debt can become in the long run, strongly suggesting that more borrowing/spending has limits as a way of dealing with an output gap.

    So I suspect two of the world's leading Keynesians would dispute the strong form of your claim, or "Mosler's Law".

    I wish your post emphasized how much better monetary solutions to AD shortfall are than fiscal solutions. I think the Keynesians agree with that as well, but regard monetary policy to be "pushing on a string" at the ZLB, hence fiscal is the only choice left. Both the negative-nominal-rates people and the MM'ers dispute that, arguing that better monetary policy would retrain traction at the ZLB. Let's put chains on our tires and stop spinning our monetary wheels for no good reason.


    Kenneth Duda
    Menlo Park, CA

    1. Kenneth,

      Don’t worry about “driving me crazy”. I went crazy long ago...:-)

      Re Krugman, the first article of his that you quote pretty much agrees with me, far as I can see. I.e. he is saying that rising debt is nothing to worry about. The second article (by DeLong) is MUCH LONGER and I don’t have time to read it. However, I’m not suggesting there is NEVER a potential problem with a large debt or monetary base (the two are almost the same thing as Martin Wolf pointed out in the FT recently).

      The potential problem is that the population suddenly tries to DISSAVE or spend away its stock of debt / base money. And the effect of that could be hyperinflation. But the same potential problem exists even in a LOW debt / base money scenario. That is, it’s always possible that every other household suddenly decides to spend as much as their credit card will allow them and spend to contents of their bank accounts. And if they did that, the result would be hyperinflation. But the reality seems to be that even in high debt / base money countries, e.g. Japan, household behavior does not gyrate all that violently.

      So what do we do? There’s a dilemma there. Personally I don’t favour enduring excess unemployment, which is a DEFINITE problem just because the alternative (a high debt / base money scenario) involves a POSSIBLE problem.

  2. Interest rates have an effect on the private economy but little effect on the spending of a government that can print money.

    With little effect of interest rates on government, we can get an economy that depends upon government deficit spending, has constant employment, and continually expands the level of government debt.

    This debt driven economy would have increasing divergence between zero earners and high income participants. Exactly like we see in today's industrial economies.

    Are we already at the point where we would PAY small businesses to borrow? We would call it a 'subsidy'.

    1. Roger, Relying on deficits does not necessarily mean increased debt: the deficit can accumulate as EITHER debt or more base money. Milton Friedman actually favored a zero debt regime, i.e. having deficits accumulate JUST AS more base money. I think he was right, though I’m not 100% sure.

    2. I think we have an accounting issue here. Usually debt is a simple record of money issue, such as, when banks lend or government borrows from either banks or non-bank private sources.

      Central Banks, or even a government treasury can issue money in exchange for equities, Mortgage Back Securities, or other assets but this exchange should result in a balance between value purchased and value of money issued. We could then determine how much money was issued by looking at the value of the assets held.

      How would the CB know how much money was issued if there was no matching debt record?


Post a comment.