David R. Henderson  

Krugman's Kontradictions or Contradictions?

A Fortune Cookie for Ron Unz... Emporiophobia at Christmas...

Blogger and frequent Econlib Feature Article contributor Robert P. Murphy has coined the term "Krugman Kontradiction." I believe, and I think Bob agrees with me, that Krugman is very clever and crafty. (Or should that be Klever and Krafty?) And so when he appears to contradict himself, without ever admitting it, which he often does appear to do, he can usually get out of it because when you go and read him carefully, you find that he didn't really contradict himself but, instead, misled his audience into thinking that he said something that he didn't quite say. Thus Murphy's term "Krugman Kontradiction." It's not necessarily a contradiction. It just represents a Krafty and Klever mind at work creating certain misleading impressions.

I thought of that when reading his post "Unprecedented Austerity" this morning. I don't know whether his post evidences Krugman Kontradictions or simple contradictions. I lean to the latter. You be the judge.

Levels vs. Rates of Change

Krugman shows a graph titled "Three Year Changes in Real Government Spending" for the United States. What it shows is a dramatic drop in the change in real government spending. That is, it shows that the change in real government spending has fallen from a high positive number to a low positive number and then, finally, to a low negative number. Keep that in mind: he's talking about a drop in the change in government spending, not a drop in government spending. Real government spending through the whole period in the graph, up to the final year or so, actually rose. But the rate of increase fell.

But then he writes as if he has shown that real government spending fell. He writes:

You can see that there was a brief, modest spurt in spending associated with the Obama stimulus--but it has long since been outweighed and swamped by a collapse in spending without precedent in the past half century.

Wrong! There was a brief spurt in the rate of growth of government spending. But it was not "long since outweighed by a collapse in spending." The so-called collapse in spending happened only, going by his own chart, in the last two years.

Government Spending vs. G

Krugman is a Keynesian. When Keynesians look at government spending to judge whether there is austerity, they tend to look at government spending on goods and services, the "G" in the "C+I+G" in the Keynesian model. Yet Krugman, if he has labeled his graph correctly, is looking at all government spending, not just spending on goods and services.

Is Austerity Causing Slow Growth?

How has Krugman handled the fact that the decline in real government spending has not caused the slow growth or recession that he has so often feared? He tries to handle this issue up front in in the piece, writing:

Intellectually, the case for austerity has pretty much collapsed, having been reduced at this point to the Three Stooges Theory: we're supposed to consider austerity a success because it feels good when you stop, or at least let up. At the same time, however, austerity policies continue to be imposed, on both sides of the Atlantic.

Well, which is it, Paul? Have the austerity policies let up, as you seem to suggest in the first sentence above, in which case one could understand why real GDP in the third quarter grew at an annual rate of 3.6 percent? Or, have austerity policies continued on this side of the Atlantic, as you say in your second sentence? In that case, how do you account for this relatively healthy growth?

One final note. Krugman writes:

Taking it further back is tricky given data non-comparability, but as far as I can tell the recent austerity binge was bigger than the demobilization after the Korean War; you really have to go back to post-World-War-II demobilization to get anything similar.

Actually, the post-World-War-II demobilization was not similar. It was huge. I wrote:
In the four years from peak World War II spending in 1944 to 1948, the U.S. government cut spending by $72 billion--a 75-percent reduction. It brought federal spending down from a peak of 44 percent of gross national product (GNP) in 1944 to only 8.9 percent in 1948, a drop of over 35 percentage points of GNP.

Oh, and, by the way, what happened as government spending was falling like a stone? We had a boom. I've documented all that here.

Comments and Sharing

COMMENTS (15 to date)
MingoV writes:

Why do economists concern themselves with Krugman's recent writings? His writings often are blatantly biased towards the left. When they aren't blatantly biased, they're often wrong or misleading. I don't read anything he writes, which reduces the number of anger-triggering events in my life.

Mike writes:

Maybe economic reality is blatantly biased towards the left these days... Did you notice the distinction he made between stopping and letting up? It would appear not: "Have austerity policies let up. . . Or, have austerity policies continued on this side of the Atlantic?" It is possible for both to be true, and that's the case that he (clearly) makes (and what's actually happening). Maybe you should spend a little more time trying to understand what he's saying instead of reading him for the sole purpose of finding "Kontradictions," which like you say, aren't really contradictions, but rather nuances. I understand that nuances aren't really libertarians' strong suit (all or nothing, am I right?!).

I used to read your posts because I thought they provided an intellectually honest alternate perspective, but as of late, your posts have been nothing but disingenuous politicized dogma... I guess reality is making the divide between your libertarian ideology and intellectual integrity more difficult by the day. That's unfortunate, because a lot of your material used to be informative.

David R. Henderson writes:

Why do economists concern themselves with Krugman's recent writings?
I have a sense that you don’t really want me to answer.

Andrew_FL writes:

Hm, what *was* spending doing, precisely?



In *nominal* dollars, *total* US government spending was:

2007 $4945884*10^6
2008 $5357969*10^6
2009 $5971335*10^6
2010 $5962997*10^6
2011 $6157506*10^6
2012 $6188651*10^6

At least *nominally*, this represents an increase every year, except 2010, when the total (state, local, and federal) nominal spending actually shrank relative to the previous year.

However, there is, of course, the issue of inflation. So I download the GDP deflator data from the St Louis Fed website, index it to unity in 2012, and divide the above by it, to get constant 2012 dollars. This leads to:

2007 $5335462.39*10^6
2008 $5669288.331*10^6
2009 $6270012.04*10^6
2010 $6186362.42*10^6
2011 $6265116.263*10^6
2012 $6188651*10^6

Now, there might be better ways to account for inflation, but, given the above, it appears that total spending is slightly lower than in 2009, which did have the highest level of total spending. It's about 2% less. However, it is also 16% more than the spending level of 2007. That still represents a rather large increase. And moreover, while the government has managed to slam on the breaks recently, this has neither constituted anything close to serious Austerity nor has it had the negative consequences Krugman types doubtless would have expected it to have.

Steven Pinker writes:

Well, usually we talk about government expenditures in relation to the GDP. And the slowing rate of growth of gov. spending that is shown by Krugman may imply that the share of gov. expenditures in the GDP was falling. It depends on the changes in the GDP and I have not calculated it.

martin writes:
And the slowing rate of growth of gov. spending that is shown by Krugman may imply that the share of gov. expenditures in the GDP was falling.

In that case he could have just used a chart of government spending as a percentage of GDP.

Brian writes:

I don't think there's anything mysterious here. The Obama stimulus appears not to have done anything measureable and the so-called austerity, contrary to Krugman's fulminations, hasn't hurt the economy in the least. Since his worldview can't account for those observations, he's reduced to spin.

[Comment edited with commenter's approval.--Econlib Ed.]

eric falkenstein writes:

Pinker:Well, usually we talk about government expenditures in relation to the GDP.

I think that's a better way to think about the size of government in most contexts, but I doubt that's the way most journalists or economists talk about gov't, which is the simple change from prior year's expenditure.

What I find most frustrating is that when you give government $800B to spend quickly, they spend most on transfer payments or give to states that then simply shore up their pension deficits, because those are always shovel-ready. Then, when the stimulus fails, Keynesians say, it wasn't government 'purchases', so its failure doesn't count.

RPLong writes:

We are dealing with a moving target here.

If Krugman and his critics (Kritics?) began a conversation about austerity in which:

(1) The definition of "austerity" was agreed upon,
(2) All sides outlined what the world would look like if they were correct,
(3) All sides outlined what the world would look like if they were incorrect,
(4) All sides reached a consensus about what the starting and stopping points for the analysis period were,

maybe then we'd be able to assess the impact of economic policies. Instead, Krugman (and others, to be sure, of all political stripes) uses whatever number makes his case. Today it might be YOY growth rates. Tomorrow it might be % of GDP. The day after that, it might be a nominal time series. The next day it might be an inflation-adjusted time series.

It's hard for a layman like me to pay attention to economists who do not stick to one set of metrics, outline their predictions in advance, and then assess the robustness of their predictions ex post facto. It's not a rational analysis if it's not laid out rationally.

D. F. Linton writes:

He imagines himself as Hari Seldon, but he's really much more Ellsworth Monkton Toohey.

Jay writes:


Feel free to keep reading Krugman then if you want to be free from politics... Nice unsupported jab at nuance and libertarians though. If Krugman has a point to make he shouldn't appear contradictory and invite criticism and then bail out due to a technicality. He should come out and say what he means and if he's right he shouldn't have to mislead his readers as he does (we can stop arguing now if you truly believe people think about government spending as a rise/drop in the CHANGE in spending and not in nominal terms).

Rick writes:

I am truly asking this question sincerely: How did Krugman win the Nobel Prize for economics? Was it for a specific economic theory?

Now I am being a bit facetious: Krugman could not have won the Prize for honest, non-partisan reasoning of the reality of America's economy.

David R. Henderson writes:

The best way to get your question answered is to spend some time on the Nobel web site. They do a good job of laying out his work.

MingoV writes:
@MingoV, Why do economists concern themselves with Krugman's recent writings?

I have a sense that you don’t really want me to answer.

Actually, I do. One aspect of economics is the study of productivity. If part of your work involves reading current articles on economics, and one of the sources consistently puts out biased viewpoints or incorrect analyses, then why would you decrease your productivity by continuing to read that source? The only reason that makes sense to me is if you need to publish rebuttals.
Seth writes:

I am also interested in the answer to MingoV's question. It has often baffled me how much attention and respect Krugman draws.

Comments for this entry have been closed
Return to top