Scott Sumner  

Bastiat just rolled over in his grave

Ballparking the Marital Return... My Cato Address to Students fo...

The Administration has released a report defending the ARRA program, and it's every bit as embarrassing as you might expect. Where to begin . . .

1. The report seems to ignore the fact that fiscal stimulus failed to produce the effects the administration claimed it would produce. Now I'm willing to cut them some slack on that point, as ceteris is never paribus. They claimed that the recession turned out to be more severe than they expected, although to be honest it was very clear by early 2009 when Obama took office that the recession was extremely severe. But let's give them the benefit of the doubt.

2. In 2013 the Federal government switched to a policy of fiscal austerity, which should have slowed growth. Instead growth in 2013 accelerated. This time there was no "the recession of 2009 was deeper than we thought" excuse. The fiscal multiplier model simply failed.

3. Instead of comparing actual to predicted results, they have lots of graphs showing the success of fiscal stimulus based on models of the economy that simply assume the policy was successful. How is that supposed to convince anyone?

4. These models are defended with a very inadequate discussion of the theoretical and empirical research that underpins the multiplier model. Let's start with by far the biggest problem, monetary offset. They correctly point out that there are new Keynesian models that show a fiscal multiplier can work when monetary policy is constrained by the zero bound. But those models assume the Fed does not engage in unconventional stimulus such as QE and forward guidance. In fact, the Fed does both, which makes those NK models completely inappropriate for evaluating fiscal stimulus in the US. They also ignore international evidence of the failure of fiscal stimulus at the zero bound, such as the Japanese fiscal stimulus of the 1990s and 2000s

5. They cite studies of fiscal stimulus using wartime spending, which is of little applicability to this crisis. These models are based on the notion that people work harder when military spending depresses consumption. Military spending during WWII did boost RGDP, but also depressed consumption and living standards (recall the Great Depression was basically over by early December 1941, before Pearl Harbor.) In contrast, only about 18% of ARRA plus subsequent stimulus was in the form of public investment, it was mostly tax cuts and transfers. Military spending declined. Given monetary offset, there's no reason to assume that tax cuts had much demand-side effect. And they failed to do the sorts of tax cuts that might have boosted the supply side, such as the employer-side payroll tax cut that Christina Romer recommended

6. If it's really true (as they claim) that fiscal stimulus was needed in 2009 because monetary policy was constrained by the zero bound, why didn't Obama appoint some people to the Fed in 2009 who would adopt Paul Krugman-style unconventional stimulus, such as the forward guidance that was eventually adopted in late 2012? After all, the Dems had a filibuster-proof majority in 2009. I think we all know the answer. Larry Summers says he prefers fiscal to monetary stimulus, even where monetary stimulus is possible. Indeed Summers is so pro-fiscal that progressive blogger Matt Yglesias once characterized his views as "socialist." (Yes, I'm sure Matt was half joking, but there has to be some truth in any joke.) The report suggests fiscal stimulus was needed to save the economy, whereas it was actually adopted for ideological reasons. Obama has not adopted a far left set of public policies. But his actual views (as far as I can tell) are left wing on almost every single economic issue. He's smart enough to know he rules over one of the most conservative developed economies, and hence is cautious in his policy initiatives. But make no mistake about it, he'll use any and all opportunities to expand government, and he'll never try to scale it back in any important way. He's not a Clinton.

7. The report also discusses cross-sectional studies of the effect of fiscal stimulus, despite the fact that cross sectional studies tell us nothing about the aggregate fiscal multiplier. Even if it were zero at the national level, building a billion dollar air force base in Fargo would boost Fargo's GDP. Everyone agrees on that. While the report does acknowledge the monetary offset issue at the state level, they go on to discuss data from eurozone countries that seems to ignore the monetary offset issue with cross-sectional studies within a single currency. Mark Sadowski re-estimated the effect for countries with independent monetary policies, and found little impact. The report seems to have been written by people living within the liberal Keynesian echo chamber, who have little knowledge of important criticisms of the fiscal multiplier model.

8. The report brushes aside any concern about the effect of UI insurance on the unemployment rate. Admittedly we don't have conclusive evidence in this area, but most of the studies I've seen (including studies cited by Brad DeLong) point to effects on the order of 1/2%. Not huge, but not trivial either.

9. There are widely implausible estimates of the impact of ARRA on GDP:

Despite the differences in the models, these private-sector forecasters all estimated that the Recovery Act would raise GDP substantially from 2009 to 2011, including a boost to GDP of between 2.0 and 3.4 percent in 2010.
Any private sector forecaster who thinks the Fed would have allowed 2010 GDP to be about 2.7% less than it actually turned out to be needs to have their forecasting license revoked, that estimate is an insult to Ben Bernanke. He was not about to preside over another depression. Bernanke would have instituted price level targeting long before that outcome materialized. People working in this area need to have common sense about plausible outcomes, otherwise nothing they say will be taken seriously. Again, look at some of the absurd NK predictions as to how the 2013 austerity would impact 2013 GDP growth. Here's what the report says about 2013:
Second, the economy has encountered a long list of additional headwinds in recent years. . . . It includes fiscal austerity at the State and local level that intensified as the Recovery Act began to phase out and has cost hundreds of thousands of additional job losses even during the expansion period. And it includes measures like the sequester which CBO estimated took 0.6 percentage point off growth in 2013, the 16-day government shutdown which the Bureau of Economic Analysis (BEA) estimated directly subtracted 0.3 percentage point from growth in the fourth quarter, and dangerous brinksmanship around the debt limit in 2011 and 2013.
They forgot to mention to big payroll tax increase, or the income tax increases, both occurred at the beginning of 2013. Quite a fiscal drag (the Economist magazine estimated the effect at 1.9% of GDP, even without the shutdown), and yet RGDP growth accelerates from 1.95% in 2012 to 2.7% in 2013. Nor is there a discussion of the fact that fiscal austerity in the U.S. was slightly greater than in the eurozone in recent years. Yet the US economy did far better, solely due to the greater willingness of the Fed to do monetary offset.

10. Elsewhere they suggest that ARRA might have actually reduced the national debt. I guess "voodoo economics" is OK if done by liberals. And note that the standard supply-side model does allow for Laffer-curve type effects, whereas the standard Keynesian model does not. That's right, even the absurd hydraulic 1960s Keynesian model does not allow for fiscal stimulus to reduce the debt. To get that result you must add on even more absurd assumptions.

I do understand that governments feel a need to defend whatever they have done. Bush still thinks the Iraq War was a success. So I expected a whitewash. But I thought they might at least be able to come up with some plausible arguments to counteract conservative criticism of the program. Unfortunately they've failed at even that modest goal. They don't even engage with their critics.

This is by far the weakest economic recovery in my lifetime.

PS. They even cite Bastiat in defense of the fiscal stimulus. Page 33; read it and weep.

HT Marcus Nunes

Comments and Sharing

COMMENTS (30 to date)
Joel Aaron Freeman writes:

Sumner, you are the voice of one crying out in the wilderness.

Lol about Bastiat

Here's another (call it The Broken Window Fallacy with Reverse English) from a rich kid who took $45,000 of his family's money and helped finance the start-up of Jeff Bezos;

... while Hanauer says he’s “incredibly proud” of what Amazon has accomplished, “it’s not an unalloyed good.”

“Amazon didn’t create any jobs. Amazon probably destroyed a million jobs in our economy,” he said, pointing to reduced employment at brick-and-mortar stores displaced by the Internet giant.

“We have to find a way as a capitalist democracy to account for that,” Hanauer said.

And that's not even the silliest belief he espouses in the Seattle Times piece.

Jon Murphy writes:

Holy frejole...that was to be the worst interpretation of Bastiat I have ever read.

I see I forgot to provide a link to the Seattle Times piece on Nick Hanauer.

James D. writes:

How convenient that the Council of Economic Advisers missed the part of Bastiat's essay concerning "Public Works":

1.87 - The sophism that I am attacking in this essay is all the more dangerous when applied to public works, since it serves to justify the most foolishly prodigal enterprises. When a railroad or a bridge has real utility, it suffices to rely on this fact in arguing in its favor. But if one cannot do this, what does one do? One has recourse to this mumbo jumbo: "We must create jobs for the workers."

1.88 - This means that the terraces of the Champ-de-Mars*12 are ordered first to be built up and then to be torn down. The great Napoleon, it is said, thought he was doing philanthropic work when he had ditches dug and then filled in. He also said: "What difference does the result make? All we need is to see wealth spread among the laboring classes."

1.89 - Let us get to the bottom of things. Money creates an illusion for us. To ask for co-operation, in the form of money, from all the citizens in a common enterprise is, in reality, to ask of them actual physical co-operation, for each one of them procures for himself by his labor the amount he is taxed. Now, if we were to gather together all the citizens and exact their services from them in order to have a piece of work performed that is useful to all, this would be understandable; their recompense would consist in the results of the work itself. But if, after being brought together, they were forced to build roads on which no one would travel, or palaces that no one would live in, all under the pretext of providing work for them, it would seem absurd, and they would certainly be justified in objecting: We will have none of that kind of work. We would rather work for ourselves.

1.90 - Having the citizens contribute money, and not labor, changes nothing in the general results. But if labor were contributed, the loss would be shared by everyone. Where money is contributed, those whom the state keeps busy escape their share of the loss, while adding much more to that which their compatriots already have to suffer.

Of course, I also think they missed Bastiat's section on "Taxes" as well.
Bob Murphy writes:

Whether it was you or Marcus Nunes, great catch on Bastiat!

john writes:

"They don't even engage with their critics."
to be fair they probably consider the republican party to be their only important critics and current elected repubs don't believe in monetary policy. they put a gold standard in the official party platform in 2012. i tend to agree with your ideas but they are not found anywhere in mainstream american politics. the criticism of this report that will be offered by the repub party will look nothing like what you list here and will be as unfounded if not more than this report

[edited with commenter's permission--Econlib Ed.]

Scott Sumner writes:

Thanks everyone.

Ebsim writes:

I feel like Keynesians use the fiscal multiplier for all arguments and then stress that they are the only "numbers" school. Like they think theoretically that all they need to do is raise the minimum wage and no one will be fired because everyone makes more money. If thats central to their argument then no wonder they have been wrong about the recovery till now. They are usually only right if they transform into monetarists or neoliberal for a short period.

WC writes:

Scott, what do you think of this poll of economists by the IGM at U of Chicago showing overwhelming support for the stimulus?

All of your points make sense and I have never been given convincing explanations for any of them by stimulus supporters. Yet the belief lives on!

Andrew_FL writes:
Military spending during WWII did boost RGDP

I have to object to this statement. In the first place, I don't know how you reliably measure real GDP when your price index is not going to show inflation due to extensive price controls.

But secondly, I think it's extremely misleading. If you were to subtract the government spending back out, you'd see that the rest of GDP was down, not up. Investment was way down, and this is much more apparent than any decrease in consumption. What seems to have actually happened is that the government completely displaced private investment.

Mark Brophy writes:

It's absurd to claim, "In 2013 the Federal government switched to a policy of fiscal austerity." The government budget has skyrocketed during the past 14 years and is practically unlimited. None of this profligacy has been reversed.

Bill Conerly writes:

I was once advised to ask myself, how would you know if you were wrong? What evidence would disprove your theory? The ARRA proponents would have had to admit that no quantitative evidence could overturn their theory. (sort of like Austrians.)

Mark writes:

Ironic that the Administration goes "mark-to-model" on ARRA, having spent, with allies, the past 5 years calling that approach a fundamental regulatory failure in the financial sector.

Hazel Meade writes:

For a bunch of people who are dedicated emphiricists when it comes to the minimum wage, they are strangely non-emphirical when it comes to the multiplier effect.

James writes:

If you had worked at the CEA, you would be kinder. Your criticisms are mostly fair, but CEA has neither the time nor the resources to do an academic-level study of ARRA. They are tasked to do this same report every 3 months, along with hundreds of other tasks. People work around the clock. Plugging in spending numbers to a multiplier model is the best they can do given the circumstances. I can guarantee you they wouldn't be doing these reports if not congressionally mandated.

Travis Allison writes:

I think of the govt as a big (usually monopolistic) corporation within the economy, where revenue is gained sometimes by fees but mostly by taxes. If the economy is below potential, then govt spending by borrowing should push accounting GDP (AGDP) closer to potential, just as if a bunch of private corporations were motivated to spend borrowed money, that would push AGDP closer to potential.

The Fed could engage in monetary offset against that govt spending, just as it can engage in monetary offset against private spending.

The special aspect of govt spending is that it is mostly paid for with taxes, which can result in a deadweight loss, whereas voluntary fees don't. So if govt created the exact same good as the private sector but the govt paid for the good with taxes, it still would be better to have the private sector provide the good. If the govt can provide the good more efficiently than the private sector to offset the deadweight loss (maybe health care?), then govt should provide the good.

Suppose for some reason, relative wages get out of whack. Govt activity alone *probably* can't change relative wages enough to make the private economy grow at full potential by itself once the govt activity stops. In the same way, if the corporations forming the SP500 all banned together to all borrow money and spend (for the good of the country, for example), once that activity stopped, the economy would still grow below potential.

Fundamentally, what's needed is for the Fed to adjust relative wages prices through manipulating the medium of account so that individual corporations are motivated to hire people. Once that happens, the economy will grow at full potential.

Given this framework, certain things are clear. Debt financed empty govt activity (digging and refilling ditches) does nothing but increase the future deadweight loss of taxes to pay off debt. Transfer payments aren't going to grow the economy (though they may provide an important function of ameliorating economic misery due to an incompetent Fed).

Thus all of this talk about multipliers is misguided. The economy will only grow naturally once relative wages have adjusted, regardless of the source of economic activity (public or private). Only the Fed can do that or a lot of time must pass, as wages will probably adjust *eventually*. ('Eventually' might be 10 years.)

Does anyone see any flaws in my reasoning?

Colin writes:

Thank you for being so forward and honest with your analysis.

Peter Hund writes:

"Don't cry for me ...... " Argentina?!

We live in a crazy world if 2013 was considered "fiscal austerity" (FY 2013 had a deficit of about 6% of GDP and spending of 22.7% of GDP, which was about 15% higher than FY 2007).

Scott Sumner writes:

WC, Most of them have probably never studied the issue, or even thought much about it.

Andrew, I did say that living standards went down in WWII.

Mark Brophy, The word 'austerity' doesn't mean smaller Federal government. And even if it did, what happened over the past 14 years would have no bearing on my assertion.

James, That's a good point. I don't mean to personally criticize the people who were forced to write the report. They were simply doing what they were told. The problem is the system that mandated the production of this sort of report in the first place.

Mark, I believe the deficit is now down to about 4% of GDP, from 10% a few years ago. It is getting smaller.

Ray Lopez writes:

Sumner says: "They correctly point out that there are new Keynesian models that show a fiscal multiplier can work when monetary policy is constrained by the zero bound. But those models assume the Fed does not engage in unconventional stimulus such as QE and forward guidance."

So QE and forward guidance (jawboning?) are the hand waving arguments that negate the new Keynesian models, which themselves are suspect? Economics is a science? Not.

David Khoo writes:

Your arguments rely on monetary policy having been loose throughout the recession. I would actually argue that monetary policy has been too tight until recently, regardless of the efforts of the Fed. The Fed has only regained traction since 2013, which accounts for the growth that year despite fiscal retrenchment. The US finally has monetary policy again! In the face of overly tight monetary policy and the inability of the Fed to act effectively before then, fiscal stimulus via the ARRA was effective.

As a market monetarist, it is natural for you to assign infinite, mystical powers to the Fed at all times and in all circumstances. Not everyone agrees. Not even "Krugman-style unconventional stimulus" grants the Fed traction under all circumstances, as Krugman would be the first to say. Just because the Fed was churning the oceans with enormous QE exercises did not mean it was actually succeeding in fighting the current. -- there were many indicators that your "monetary offset" just did not exist until 2013.

Ray Lopez writes:

Sumner says: "They correctly point out that there are new Keynesian models that show a fiscal multiplier can work when monetary policy is constrained by the zero bound. But those models assume the Fed does not engage in unconventional stimulus such as QE and forward guidance."

So QE and forward guidance (jawboning?) are the hand waving arguments that negate the new Keynesian models, which themselves are suspect? Economics is a science? Not.

PeterP writes:

[Comment removed pending confirmation of email address. Email the to request restoring this comment. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]


See Table 1.2—Summary of Receipts, Outlays, and
Surpluses or Deficits (-) as Percentages of GDP:

It reports a deficit of 6% of GDP for FY 2013 (ending
09/30/13) and a deficit of 4.4% of GDP for FY 2014 (ending
09/30/14). Also, we still spent 36% more than we took in for
FY 2013, and spending as a percentage of GDP was much higher
than before the economic downturn.

It's also interesting to note that a deficit of 6% of
GDP is higher than any deficits during the New Deal.

While we were marginally less fiscally irresponsible than in
the prior year, I would hardly call this

Steve Roth writes:

Predicted difference between ARRA and no-ARRA was incorrect.

Three possibilities:

1. ARRA prediction was incorrect.

2. No-ARRA prediction was incorrect.

3. Both.

Nothing in this post proves which of these occurred.

Steve Roth writes:

Great minds...


"The striking thing here isn’t what they thought the Recovery Act would achieve, it’s how optimistic they were about how the economy would perform even without stimulus."

Brian W writes:

The biggest question I have always had about the keyensian multiplier arguement is if you temporary boost spending to fix the economy what happens later on when you must cut spending to pay for the spending of the past? Wouldn't using their theory cause the same economic problems they insist the spending fixed just at a different time? Would love to hear from Scott or any of the commenters about how they somehow explain this away

TallDave writes:

Instead of comparing actual to predicted results, they have lots of graphs showing the success of fiscal stimulus based on models of the economy that simply assume the policy was successful.

Texas Sharpshooter Fallacy rides again. Seemingly endemic.

Comments for this entry have been closed
Return to top