Scott Sumner  

Russ Roberts vs. Simon Wren-Lewis

Henderson, Meckling, and Fried... The Democrats move left...

I have repeatedly pointed out that the US government did a massive amount of austerity in 2013, and yet GDP growth was higher than during 2012. One would think that Keynesians would have a response to my claim, a response that addresses the specific points that I made. Unfortunately, that has not been the case.

Russ Roberts recently did a post pointing to this problem:

When there was an increase in the payroll tax in January 2013 and the sequester reduced spending beginning in March 2013, Krugman wrote that monetary policy, which was expansionary, would not be able to offset the impact of fiscal policy: Mike Konczal points out, we are in effect getting a test of the market monetarist view right now, with the Fed having adopted more expansionary policies even as fiscal policy tightens...

Sorry, guys, but as a practical matter the Fed - while it should be doing more - can't make up for contractionary fiscal policy in the face of a depressed economy.

When the economy grew more quickly in 2013 than in 2012, the test was no longer a test. Here is Krugman in January of 2014:

Incidentally, these other factors are why I don't take seriously the claims of market monetarists that the failure of growth to collapse in 2013 somehow showed that fiscal policy doesn't matter. US austerity, although a really bad thing, wasn't nearly as intense as what happened in southern Europe; it was small enough that it could be, and I'd argue was, more or less offset by other stuff over the course of a single year.
Strange answer. As Scott Sumner pointed out, it wasn't that growth didn't collapse. The economy grew faster, particularly in the 3rd and 4th quarter when the sequester took effect. But the key is the phrase in the last sentence-the reference to "other stuff." Krugman's answer is that if it weren't for "other stuff" the economy would have collapsed.

There is always "other stuff."

When it's convenient, Krugman ignores the "other stuff." (We all do.) So according to Krugman, the economy of Kansas is struggling because the governor cut taxes. One variable explains everything. We can ignore the other stuff. Or Iceland is doing better than Ireland because Iceland didn't listen to those foolish austerians and rejected the fiscal austerity that Ireland fell prey to. One variable, fiscal policy, explains everything. When it's convenient, when it confirms our worldview, the other stuff can be ignored. More troubling for Krugman's claims is that it appears that Iceland actually embraced austerity big time--raising taxes and cutting spending. How will Krugman explain Iceland's success? Must be other stuff. But there is always other stuff. The world is a complicated place. I would suggest that our job as economists is to always remember the existence of other stuff, all the time and not just when it's convenient.

Roberts is right, and indeed it's even worse than he suggests. The government did not just increase payroll taxes at the beginning of 2013, income taxes also rose.

This seems like a pretty convincing argument, and thus you'd expect a serious response from Keynesians economists. Instead, here's what Simon Wren-Lewis said:

To illustrate their belief that Keynesians ignore awkward facts both the authors above use the example of US growth following the 2013 sequester. (In my experience anti-Keynesians tend to shy away from data series, and especially econometrics, and prefer evidence of the 'they said this, and it didn't happen' kind - particularly if 'they' happens to be Paul Krugman.) The problem is that this episode actually illustrates the opposite: that anti-Keynesians are so keen to grasp anything that appears to conflict with Keynesian ideas that they fail to do simple analysis and ignore others that do.

In this post I just looked at the data and did some simple arithmetic to show that this episode was quite consistent with Keynesian fiscal policy analysis. I'm sure others have done the same. But such analysis just gets ignored: they have a superficially good story, and that is all that matters. (Read this post to see how Scott Sumner in response to my work dug himself an even deeper hole.)

So I checked the post that supposedly refuted my views, and found that Wren-Lewis was wrong on both issues he raises. First, he addresses my claim that when examining the effects of the austerity that began on January 1st, you need to use Q4 over Q4 growth rates, not year-over-year growth rates. I doubt if there's a single serious macroeconomist who would disagree with me. And yet all he does is respond with snark, he doesn't even address my complaint. I think it's fair to say he doesn't have any argument against my claim, because there are no good arguments against my claim.

His second claim is to deny that austerity occurred in 2013. This claim boggles the mind, as the budget deficit fell by $500 billion in calendar 2013, from a bit over $1 trillion to a bit over $500 billion. (You need to use calendar year data because the austerity began January 1st, and the fiscal year begins October 1st.) Of course there's also that letter signed by 350 Keynesians warning that the 2013 austerity could lead to recession.

So Wren-Lewis must have some sort of evidence for the claim of no austerity. Yes he does, but the evidence is of no relevance. He cites data on government output. But of course that's not the variable used to measure austerity in the Keynesian model, as it excludes both tax changes and transfer changes, which are the majority of fiscal policy in the modern world. So Wren-Lewis argues I am wrong by citing data that even Keynesian economists would admit has no bearing on the argument. Or does he want to argue those deep pension reductions being forced on Greece are not "austerity"?

I also find his claim that my side doesn't do econometrics to be rather breathtaking in its audacity. We've done econometric studies that refute Keynesian econometric studies, and done many posts discussing those results. But as far as I know, no Keynesian blogger has ever responded to any of our studies. Over at TheMoneyIllusion I will do a post later today with recent empirical work by Mark Sadowski. I expect Keynesians to ignore this one too.

Simon Wren-Lewis seems to call for civil discourse in blogosphere debates. He starts off his recent post quoting Russ Roberts:

"The evidence for the Keynesian worldview is very mixed. Most economists come down in favor or against it because of their prior ideological beliefs. Krugman is a Keynesian because he wants bigger government. I'm an anti-Keynesian because I want smaller government."
And then responding:
Statements like this tell us rather a lot about those who make them.
But here's the title of Wren-Lewis's post criticizing mine:
"Faith based macroeconomics."
I guess that's why (left) liberal economists like Jeffrey Sachs think fiscal stimulus has failed---blind faith in laissez-faire economics.

Comments and Sharing

COMMENTS (32 to date)
Rajat writes:

That was the post that sealed it for me - I've unsubscribed to SWL's blog. He has been doing post after post railing against British austerity and against Obsorne's new plans all based on his same argument that monetary policy is only unpredictably effective at the ZLB. But this post jumped the shark.

Adam writes:

I just read Simon's "Faith-based Macro" post. It seems to me that the motivation for his post is about use of english and civility. He complains that you used the words "wrong" and "he claims" in reference to his quoting of year-on-year GDP figures. I can understand his perspective. He did not get the raw statistics wrong, as your post implied. You could have said he chose an inappropriate time-period instead.

You sometimes (rightly) highlight the negative tone of Krugman's posts. We should all aspire to be civil and constructive, so perhaps in this case, your original post could have been a little more precise to avoid needlessly prickling Simon.


Nick writes:

Fair enough. But it would be more 'civil' of the Keynesians to admit that people have actually done some research which makes their claims look *a little* silly. They are acting like 2013 never happened, and people are crazy to think it did! It's not so hard to say that, yes, the data from that year did not fit their model or their personal predictions. These things happen!
If the GS warriors had lost the NBA finals, it would have gone against lots of people's models of the sport. Those people would still hold to their models, just like the Keynesians, but they would at least admit that the games actually occurred!

I always thought that from a Keynesian viewpoint, deficit spending was supposed to be stimulative. If you decrease deficit spending from one year to the next but still run a deficit, is that supposed to be contractionary?

What if you go from running a massive surplus to running a small surplus? Is that supposed to be stimulative?

I think it is interesting that countries can run massive deficits and have it considered to be "austerity." Economists definitely use the term differently from the general population!

Mike Konczal writes:

"I think it's fair to say he doesn't have any argument against my claim, because there are no good arguments against my claim."

What happens if you move your data point forward one-step, and do Q1 over Q1 2014?

"because the austerity began January 1st"

No. You can see the military spending fall in Q4 2012 GDP, which matters because it depresses the end of 2012 and throws off your starting point. And the sequestration comes later in March.

From my POV, what gets lost is that the original debate was whether or not the Federal Reserve was going to hit a 2% inflation target after doing the Evans Rule and QE3 together. It did not, which I read as a problem for the omnipotent Fed crowd.

A writes:

Mike Konczal,

Tim Duy argued from, at least, 2012 that the Fed was signaling an asymmetric response to hitting the inflation target from above vs. from below. This is an example:

The most recent Fed presentation predicted falling short of the "target" through 2017, yet the Fed still forecasts a rate increase later this year. How are these actions consistent with a central bank failing to hit its target?

A writes:

Mike Konczal,

Tim Duy argued from, at least, 2012 that the Fed was signaling an asymmetric response to hitting the inflation target from above vs. from below. This is an example:

The most recent Fed presentation predicted falling short of the "target" through 2017, yet the Fed still forecasts a rate increase later this year. These actions seem less like a failure to achieve a target, and more like an achievement of an implicit goal.

Mike Konczal writes:

Google "Evans Rule", which was explicitly engineered to address the concerns of Duy et al.

Mike Konczal writes:

Didn't mean for that "google this" to sound rude. Just wanted to flag the significant conversation at the time about how the Evans Rule avoided “Delphic” guidance and instead used “Odyssean” guidance instead. Didn't seem to matter.

Kevin Erdmann writes:

Here's the picture.

So, do I understand this right? If the Federal government hadn't screwed up and cut the deficit in half in 2013, we would have seen a surge in NGDP to 8% or something? That would just assume a multiplier of 1, though. If the multiplier is higher, did we just screw up a chance at 10% NGDP growth?

Scott Sumner writes:

Rajat, I was also underwhelmed.

Adam, Perhaps, but it seems to me that it was pretty obvious that when I said "wrong" I meant he was using the wrong sort of data. In context, I think my meaning was pretty clear. It seems to me that he plays word games without addressing the substantive point. That's not my style. I don't think the term 'wrong' is particularly insulting relative to all the name-calling on the internet, but I certainly respect your view that the debate should be kept civil.

Mark, I wondered about that to, but that seems to be the Keynesian assumption. Because I believe their entire model is wrong, it's not clear to me which definition is right.

Mike, I'm perfectly happy using Q1 to Q1 data, and if we do then market monetarism still passes Krugman's test, just not as easily. Or you could average Q4 and Q1 to get a point estimate of year end GDP.

Arguable the jobs numbers are more accurate, and they are at monthly intervals. They show a very small acceleration of growth in 2013, but certainly no slowdown.

A, Very good point. If the Fed really was out of ammo back around 2013 then they are being extremely irresponsible talking about a rate increase this year. Policy works with long leads. Expectations of this rate increase would have depressed demand back in 2013.

Kevin Erdmann writes:

Here's the same graph with a long time frame and just the YOY GDP lines, to clean it up. This shows what a rare natural experiment this was regarding a sharp change in Federal spending outside of recessions.

It's a shame that such a huge positive exogenous shock suddenly hit the GDP numbers at exactly the right time to screw up the experiment.

Kevin Erdmann writes:

Sorry, here are hyperlinks for the two graphs.



Tom West writes:

When it's convenient, Krugman ignores the "other stuff." (We all do.)

One of my favourites from years ago was when I was debating with both a free-marketer and a Keynesian. I had a little fun at the Keynesian's expense by poking gentle fun at the growth during a period of austerity, and the free-marketer jumped in about being freed from government waste.

I then pointed out that taxes had gone substantially up during that time.

He got pretty huffy. That was *completely* different. Everyone knew higher taxes killed growth, so that was *definitely* "other stuff".

I'd hoped to get them laughing a little at themselves, but actually I just ended up getting them both irritated at me for not understanding the "other stuff".

Mark A. Sadowski writes:

Tom Brown writes:

@Tom West, that's a funny story. I invented a scenario in my head like you describe after reading Scott's post here earlier today, so it's amusing to see that you actually experienced it.

D. F. Linton writes:

I would understand Russ Robert's point in light of his ongoing position that the facts of economic history simply will not serve to settle arguments over economic theory. There is always a number just-so stories that fit an observed set of historical facts.

The real economy is unknowably complex, non-stationary, and asynchronous.

Even the best "natural experiments" in economics are surrounded with random froth which econometricians hope, but can not prove, are not too relevant.

This is pretty much a "see you and raise you" from "never argue from a price change".

embrace squalor writes:

Here's how I see it. Lets say the economy is a 57 chevy with a dead battery. The car is 50 feet away from cliff. In case you don't understand metapors, the cliff represents financial collapse.
In order to get the car started you have to talk your friends into helping you push the car up to a speed where you can pop the clutch and get going before you go over the cliff. Your friends in this case represent a financial stimulus of sorts. There are four girls in the Chevy. Girls in this scenario represent 16 trillion dollars worth of debt. The girls won't get out of the car because its raining and they just got their hair done. Do we the people underestimate how much harder it is to push start a 57 chevy with 17 trillion dollars of debt versus push starting a 57 chevy with much lighter girls or fewer girls.
At some point you have more girls (debt) than you can ever have friends (financial stimulus) and your friends naturally know that as you are pushing you are getting closer to pushing your 57 chevy full of girls over a cliff and they naturaly stop pushing.
where am I going wrong here?

embrace squalor writes:

To make the scenario even more realistic, the tires on the chevy are getting increasing muddy as you push (The more you push (stimulate)
the more drag you get from the mud (increasing debt being a by product of the stimulus), and it is beginning to feel like maybe the girls were a
a little heavier than what you thought because you were maybe a little blinded by the possibility of finding " paradise by the dashboard light". And then you realize that nobody really knows where the cliff is because it is dark and rainy and finally you realize that your Wisemen friends from Harvard aren't really as strong as they think they are when it comes to pushing cars.

ThomasH writes:

A definite win in blogger ping pong.

Nothing much was really learned, however, because we do not know the specifications of the "Keynesian" model Krugman or the MM model that you were using to make your predictions so we can't go back and use actual values instead of expected values to see whose model performed better.

I take it you are saying either that Krugman was mistaken about his expectation of some monetary policy variable or that he is mistaken about the parameter of that variable as it apples to other variables of interest like income.

You got the right answer but was it because you made a better estimate of what policy would be or because your model links policy to outcomes better?

embrace squalor writes:

Mr. Linton writes"The real economy is unknowably complex, non-stationary, and asynchronous."
Sounds like you are describing "God"
Does this mean that "economics" is not the dismal science, maybe not a science at all. Maybe more of
religion or a secular belief system upon which those in the field project their neuroses.
I am a biologist. I would never describe the field of biology as "unknowable"

E. Harding writes:

@ Kevin
This would be a better graph:

Kevin Erdmann writes:

E. Harding,

I thought the idea was that a change in the deficit changes the rate of growth in GDP. So, I graphed the deficit and GDP, over time. Wouldn't you need to change the GDP numbers to the second derivative in your graph?

Scott Sumner writes:

Thanks Kevin, Very nice graphs.

Tom, That's a good story.

Mark, Excellent post by Marcus.

Embrace, I'm afraid you lost me there somewhere.

Jose Romeu Robazzi writes:

SWL once completely deleted a comment I made. I never went back to his blog. I am not an economist, I just want to learn something. As such, I should be allowed to ask very silly questions once in a while. Also, I am allowed to ask "common sense" questions that sometimes will be completely wrong. I am not obligated to go there and always agree with the blogger.

Deleting comments from readers (provided they are not outright offensive) also says a lot about the blogger.

E. Harding writes:

"I thought the idea was that a change in the deficit changes the rate of growth in GDP."
-Me, too.
"So, I graphed the deficit and GDP, over time."
-And I changed that to the change in the deficit.
"Wouldn't you need to change the GDP numbers to the second derivative in your graph?"
-No. Why?

Kevin Erdmann writes:

E. Harding,

If you like your graph, that's fine. If you moved the GDP numbers to the second derivative, then if changing deficits affected GDP growth, the lines would move up and down together. But, it could be that I have something wrong about how fiscal consolidation is supposed to affect GDP growth.

Scott Freelander writes:


It saddens me to see a fair number of these Keynesian economists making such fools of themselves. Simon Wren-Lewis is a smart and nice guy, who like you, takes a great deal of time writing for both economists and non-economists, and yet he's very blind on the this debate. It isn't even fair to call it a debate. And I'm sorry to say, that leftist ideology might be part of the problem, though who knows?

One thing that reading your posts has done for me as a liberal is reveal that liberals have just as many economic biases as conservatives. But, at least the liberal biases usually come without the racism and nativism.

D. F. Linton writes:

Biology is a science. But can you tell me how the bugs under a rock in my garden will do this winter or even whether the US average bugs-under-rocks average will be up by 2% this summer?

Scott Sumner writes:

Jose, I see that a lot from Keynesian bloggers.

Scott, Yes, I also see lots of bias on both sides. Perhaps you are right that liberals are less nativist and racist (I generally don't like these sorts of generalization.) But based on interactions with many liberals over the years, I can assure you that many of them have their own forms of bigotry. They have a disturbing tendency to call other people "racist", which is sort of like conservatives calling people "communist" back in the 1950s. Fine if the label fits, but sometimes used to hurt the reputation of innocent people.

Mike Sax writes:

"I doubt if there's a single serious macroeconomist who would disagree with me"

Maybe you got to give us your list of serious macroeconomists as you seem to be forgetting Wren-Lewis himself is regarded much more seriously by most than you. To speak of you and Krugman in the same sentence is just a joke.

Is Russ Roberts your whole list? Or maybe just all conservative economists are on this serious list.

Scott, you're the one who doesn't respond to awkward points.

I've mentioned this before and you just either don't answer or engage in snark which I guess shows how serious an macroeconomist you are. I'll say this you are far more snarky than any serious macroeconomist on the Internet which again underscores the fact that: maybe you just aren't all that serious.

Again, show us the counterfactual that if you had the exact same monetary policy in 2013 but without the austerity we would have had lower growth or even the same amount of growth as we did with the austerity.

That's' the real test and you never address this and I'm betting you won't here either. You'll probably come up with a one word insult that proves nothing but that you can't and never will be able to.

The argument is that QE3+fiscal contraction yields slower growth than QE3+no fiscal contraction. Unless you can show this you've proven nothing.

Mike Sax writes:

And while you're not responding to that point maybe you can also fail to respond to this one.

Iceland's so-called austerity came after the recession was over and the finance minister declared victory.

Great odds you won't touch this one as it would cut you and Russ Roberts off at the knees.

[Your broken url has been fixed. Please watch for duplicated pastes of https in your own url. Maybe you were hastily in a bad mood? Just a thought.--Econlib Ed.]

Comments for this entry have been closed
Return to top