Scott Sumner  

Krugman's dangerous idea (It worked for me too!)

Greece and tax sadist tourism... On Fringe Benefits I Blew the ...

In a recent post I argued that Paul Krugman had started out as a critic of intellectuals who ignored economic models---the sort of person who argued that the laws of supply and demand do not apply to the minimum wage debate. Of course now Krugman himself argues the laws of supply and demand do not apply to the minimum wage debate. I suggested that his new stance had an appeal to non-economist intellectuals that was roughly comparable to the appeal of John Kenneth Galbraith in the 1950s and 1960s.

Commenter Roger pointed me to a fascinating essay written by Krugman in the 1990s, called "Ricardo's Dangerous Difficult Idea." This passage reminded him of my post:

I once had a very unpleasant, but ultimately useful, conversation with the editor of one of America's leading intellectual magazines. He was in the process of refusing to print a piece I had written at his request, and his dissatisfaction with what I had written was the main subject at hand. But along the way I somehow mentioned the need to represent economic ideas with carefully thought-out models, and he responded with a mixture of bafflement and asperity. Clearly the idea that economic ideas could benefit from being modeled was new to him, even though his journal frequently publishes articles on economic affairs; and he suggested to me that in future I would do well to explain why models are sometimes useful and why they usually are not.

At the time I was fairly flabbergasted: to question the usefulness of economic models at this late date seemed rather strange. But the economist's idea that economic theory for the most part consists of models has by no means been accepted by intellectuals outside our field. In fact, if one looks at the favorite economic writers of the non-economist intellectual -- Robert Reich, Lester Thurow, John Kenneth Galbraith -- one realizes that they have in common an aversion to or ignorance of modeling. There are model-oriented economists, like Alan Blinder, who also write for a broader audience, and they don't put their equations in their books and articles; but the skeleton of the models that structure their thought is visible under the surface to those who know how to look. By contrast, in the writings of Reich or Galbraith what you read is what you get -- there is no hidden mathematical structure to the argument, no diagram one might draw on a blackboard or simulation one might run on a computer to clarify the point.

Just to be clear, even today Krugman's essays are much more informed by mathematical models than Galbraith's ever were. But he has certainly moved in a direction where he is much more appealing to the sort of intellectuals who like Galbraith.

Rereading Krugman's essay after many years I was stunned by just how good it is. I could not recommend it more highly; indeed I don't recall ever reading anything better in this genre. Toward the end of the essay he has 4 recommendations for public intellectuals. Here's one that caught my eye:

(ii) Adopt the stance of rebel: There is nothing that plays worse in our culture than seeming to be the stodgy defender of old ideas, no matter how true those ideas may be. Luckily, at this point the orthodoxy of the academic economists is very much a minority position among intellectuals in general; one can seem to be a courageous maverick, boldly challenging the powers that be, by reciting the contents of a standard textbook. It has worked for me!
Hmmm, "contents of a standard textbook." So let's see how that would work. Let's take three key concepts that Frederic Mishkin emphasized (in 2008) at the end of his best-selling textbook on Money and Banking:
1. It is dangerous always to associate the easing or the tightening of monetary policy with a fall or a rise in short-term nominal interest rates.

2. Other asset prices besides those on short-term debt instruments contain important information about the stance of monetary policy because they are important elements in various monetary policy transmission mechanisms.

3. Monetary policy can be highly effective in reviving a weak economy even if short term rates are already near zero.

So the fact that interest rates were low and falling in 2008 does not imply that money was easy. OK, but it also doesn't mean it was tight. But all the other asset markets were also suggesting---no not suggesting, they were all screaming---that money was very tight. And point #3 suggests that even when interest rates hit zero at the end of 2008 the Fed still had the ability to engage in monetary stimulus. Hence money was tight during 2008-09 and it was the Fed's fault. Or at least that's what the number one textbook implies. The Fed caused NGDP to fall at the sharpest rate since the 1930s, and this caused the Great Recession.

Now suppose that back in early 2009 some completely unknown economist at an unknown college took these boring, conventional textbook ideas and boldly presented them as if they were daring, dangerous, rebellious ideas, smashing the conventional wisdom on the Great Recession. I wonder how far he'd get before they arrested him?

Again, I can't emphasize enough how good Krugman's essay is. If you haven't done so you should read it carefully. Don't be put off by the fact that you might find Krugman's recent stuff to be irritating. It's a masterpiece.

Comments and Sharing

COMMENTS (19 to date)
David R. Henderson writes:

I agree with you about Krugman’s Ricardo’s Difficult Idea essay. I’ve pushed it a number of times on Econlog.

Daniel Kuehn writes:

Krugman seems more like a Blinder than a Galbraith, which is why I didn't like your earlier post.

On this recent minimum wage column the obvious subtext is search and matching models with wage bargaining, which doesn't have supply and demand curves. Wasn't that the underlying model we as economists were supposed to read into that article (to say nothing of the post Card-Krueger/Neumark-Wascher empirical literature he alluded to?).

Daniel Kuehn writes:

I think this is very clear with Piketty too, fwiw. Non-economists are a bit mesmerized by the two "fundamental laws" and r > g. It's just enough to be accessible to a wide audience. Economists recognize what's just under the surface and see the implicit and quite standard growth theory and dynamic inefficiency issues.

Don Boudreaux writes:


Right you are about Krugman's superb essay "Ricardo's Difficult Idea." Krugman (as you know) has lots of excellent essays - alas, almost none of which have been written during this millennium. Another of my very favorite essays - not only by Krugman, but by any economist ever - is Krugman's May 1993 essay in the American Economic Review entitled "What Do Undergraduates Need to Know About Trade?"

Here's a slice (found on page 123 of Krugman's excellent 1996 collection, Pop Internationalism, in which this 1993 essay is reprinted): "Trade policy should be debated in terms of its impact on efficiency, not in terms of phony numbers about jobs created or lost." This correct and wise statement strikes me as being seriously at odds with some of what Krugman has written recently about trade policy - for example, his June 2014 argument in support of reauthorizing the U.S. Export-Import Bank on the grounds that the allegedly increased trade that Ex-Im subsidies allegedly promote are allegedly good for job growth in an economy with lots of unemployment.

Scott Sumner writes:

Thanks David.

Daniel, I think that Krugman has moved somewhat in Galbraith's direction, but I certainly agree that he's still thinking in terms of underlying mathematical models much more than Galbraith did. But I do think that the appeal of Krugman to non-economists is similar to the appeal of Galbraith. Both rejected the conclusions of the "dismal science" in a number of key policy areas:

1. Krugman has shifted on the question of whether unemployment comp. leads to more unemployment.
2. He's shifted on the danger of deficits.
3. He's shifted on tariffs on Chinese goods.
4. He's shifted on the minimum wage.
5. He's shifted on the utility of fiscal stimulus in Japan.

Then there are the changes in tone, from seeming to approve of neoliberalism to seeming to disapprove.

All of those shifts are in the direction that non-economists on the left would prefer.

I would also refer you to the quotes I provided in my previous post on this, where in the earlier post Krugman is explicitly criticizing exactly the sort of thinking that he himself uses in the later post.

Also note that this post was not "anti-Krugman", as I was not being at all sarcastic when I described his essay as "brilliant. It is.

Don, Good point.

Daniel Kuehn writes:

I know you said you liked the essay several times. I do too. That's not what I take issue with, it's the Galbraith comparison.

As I can only dream of being compared to EITHER Galbraith or Blinder, I hardly consider that anti-Krugman. But I do think it's a misread of his recent stuff.

David R. Henderson writes:

I second what Don says. Pop Internationalism is great. I’m glad Don reminded me of the piece for undergrads. I may put it on my next syllabus.

Jacob A. Geller writes:

In his essay Krugman deftly handles the claim, made by Michael Lind in the early 90s, that real wages had fallen 13% from 1977 - 1992. The claim was, in short, a "totally untrue...statistical quirk."

Why doesn't Krugman, and other progressives, now think just as critically about another, almost identical claim made today -- that real median household incomes have stagnated or fallen over the last 3 to 4 decades?

As if divorce rates, tax law, immigration, female labor force participation, college attendance, population aging, entitlements, etc. etc., have no (purely statistical) effect on median household income worth adjusting for...

foosion writes:

As if divorce rates, ... have no (purely statistical) effect on median household income worth adjusting for

I frequently see similar statements in response to claims that real median household incomes have stagnated or fallen over the last 3 to 4 decades

I have yet to see anyone actually do convincing analysis regarding HH size. My back of the envelope look at census data indicates that changes in household size are not nearly enough to counteract the claim.

foosion writes:

now Krugman himself argues the laws of supply and demand do not apply to the minimum wage debate

I thought his current position was that of course supply and demand apply, but as an empirical matter the effects are minimal, at most, and that other factors apply which make increasing the minimum wage a good idea. I don't want to argue whether these reasons are true or valid, just that if they are true, you'd have a good reason to favor a reasonable increase in the minimum wage.

Or take trade, the direct subject of Ricardo's Difficult Idea. That we agree with the essay does not mean we can't oppose, for example, the Trans Pacific Partnership (so far as its terms are currently public). You might believe distribution issues overcome the advantages. Or you might believe the agreement is more about protecting intellectual property (a form of protectionism) or changing the legal system to favor corporate claims than about increasing trade.

It is a brilliant essay.

mico writes:

What makes me most suspicious of minimum wage advocates is that they don't compare the policy to alternatives that do essentially the same thing but without deadweight loss.

Any country with a welfare system in effect has a minimum wage: the smallest amount of welfare to which you can be entitled. A negative income tax provides for an explicit such minimum wage with strong safeguards against disemployment.

The empirical argument for the minimum wage seems to be that the effects are so small, because the minimum wage is so low, that any harms are smaller than the measurement uncertainty. That may be true. But if the benefits are also smaller than the measurement uncertainty, which does this risky policy recommend itself as against a negative income tax?

Ultimately support for the minimum wage seems to be a combination of status quo bias and social approval seeking from people who are not even aware of the economic debate.

E. Harding writes:

@Jacob A. Geller
Old Krugman:
"More specifically: the 30 percent productivity increase he cites was achieved only in the manufacturing sector; in the business sector as a whole the increase was only 13 percent. The 13 percent decline in real wages was true only for production workers, and ignores the increase in their benefits: total compensation of the average worker actually rose 2 percent. And even that remaining gap turns out to be a statistical quirk: it is entirely due to a difference in the price indexes used to deflate business output and consumption (probably reflecting overstatement of both productivity growth and consumer price inflation). When the same price index is used, the increases in productivity and compensation have been almost exactly equal. But then how could it be otherwise? Any difference in the rates of growth of productivity and compensation would necessarily show up as a fall in labor's share of national income -- and as everyone who is even slightly familiar with the numbers knows, the share of compensation in U.S. national income has been quite stable in recent decades, and actually rose slightly over the period Lind describes."
New Krugman:
Not only did he forget his own arguments, he lied about doing so!
Note: There has been some noticeable deterioration in the labor share since the Great Recession. But that doesn't negate the fact that New Krugman lied!

Scott Sumner writes:

Daniel, I respect that argument. I feel much more confident in saying that his appeal is now similar to Galbriath's appeal, then in saying his approach is similar. I think he's moved somewhat in the Galbraith direction in terms of approach, but I'm not sure how far. I certainly agree that his thinking is more rigorously grounded in models.

David, I agree.

Jacob, I noticed that and was thinking of commenting on the point.

Foosion, No, he says the laws of supply and demand do not apply.

Mico, I like the wage subsidy idea even better, so I agree with your point.

Lorenzo from Oz writes:

I like Krugman's essay a lot, but it is still just a little economic-jargon heavy to really work for a lay audience.

Brett writes:


Ultimately support for the minimum wage seems to be a combination of status quo bias and social approval seeking from people who are not even aware of the economic debate.

It's more a mixture of practical policy-making and old norms about "rewarding work" and "corporate responsibility". The minimum wage at least appears to be pro-work, since it encourages people to compete for and work more.

@Scott Sumner

Seems like you could actually make your wage subsidies counter-cyclical if you wanted to, pegging them to unemployment (or some other metric). During periods of high unemployment they'd spike upwards, driving down the costs of hiring new employees to much lower across the board. Vice versa during periods of low unemployment.

Tim Worstall writes:

"Again, I can't emphasize enough how good Krugman's essay is. If you haven't done so you should read it carefully. Don't be put off by the fact that you might find Krugman's recent stuff to be irritating. It's a masterpiece."

I'm afraid that essay drives me into incoherent foaming rage.

How dare the man be so damn good at two things? Both academic economics and also pop essay writing?

It's like finding out that the most beautiful woman in the world helped invent wi-fi (Hedy Lamar...patent on spectrum skipping) or finding out that an Oscar winning actress not only has a Bacon number but an Erdos one too (Natalie Portman).

Completely enraging. What are us normal poeple supposed to do?

Scott Sumner writes:

Lorenzo, That's probably true, I wouldn't notice that.

Brett, Yes, that could be done. I seem to recall the Germans did something similar in the recent recession.

Tim, That's high praise from someone who writes as well as you.

E. Harding writes:

[Comment removed for irrelevance. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

maynardGkeynes writes:

It's worth noting that Galbraith was one of the finest essayists, rhetoricians and all around "wits" of the 20th century, up there with Mencken, Edmund Wilson, and E.B White. I think his appeal and influence grew as much if not more from that than from his technical work as an economist. He's still so much fun to read, even today, when his ideas seem quaint (Affluence ha!).

Comments for this entry have been closed
Return to top