Professor Krugman and Crude Keynesianism, by Jeff Sachs: I recently published a Washington Post op-ed called "Deficits do Matter" that was co-authored with talk-show host, commentator and former Congressman Joe Scarborough. The piece argues that high and rising levels of public debt are a real concern. It also makes the case that the stimulus packages that began in 2009 --which have consisted mainly of temporary tax cuts and transfer payments -- have significantly raised the public debt while doing very little to solve the nation's long-term employment and growth problems.
Let's stop there. As many people have pointed out, the basic premise that Krugman thinks deficits don't matter is just wrong (see some of his earlier writing when we weren't in a liquidity trap as well, he is very clear on this point). Here's Krugman:
Right now, deficits don’t matter — a point borne out by all the evidence. But there’s a school of thought — the modern monetary theory people — who say that deficits never matter, as long as you have your own currency. I wish I could agree with that view — and it’s not a fight I especially want, since the clear and present policy danger is from the deficit peacocks of the right. But for the record, it’s just not right.
Back to Sachs:
... I have argued against short-term stimulus packages. Krugman has supported them, and indeed argued that they should have been even larger. I have been against temporary tax cuts and temporary spending programs, believing that instead we need a consistent, planned, decade-long boost in public investments in people, technology, and infrastructure. Such a sustained rise in public investment should have been paid for by ending the Bush-era tax cuts in 2010, or by adopting a comparable boost in revenues. Instead Obama and Congress have now made almost all of those tax cuts permanent, putting us into a deeper fiscal bind.
On Twitter, he goes a bit further:
He can think whatever he wants, but the evidence says otherwise. In a severe recession government spending multipliers are much larger than in normal times. That's why this statement is so puzzling:
...First off, here is what I mean when I say that Krugman is a crude Keynesian... There are four elements of crude Keynesianism and, indeed, of Krugman's position:
(1) The belief that multipliers on tax cuts and transfers are stable, predictable and large; (2) The belief that America's employment and growth problems are overwhelmingly cyclical, not structural, and therefore remediable by short-term aggregate demand management; (3) The belief that a growing debt burden is a minor nuisance as long as the economy is in recession; (4) The belief that for practical purposes, the most urgent need is to raise aggregate demand rather than to focus on the quality and type of public spending.
I believe that all of these positions are misguided.
Again, he can believe whatever he wants, but the evidence is against him. On (1), as already noted, Krugman has NOT said that multipliers are "stable, predictable, and large." He is saying what both the theorists and empiricists looking at this are saying, that multipliers are larger in a liquidity trap. It's a result that comes directly out of modern DSGE models. Second, asserting that our problems are mostly cyclical is fine, but go to the SF Fed who has done a lot of work on this, talk to Narayana Kocherlakota at the Minnesota Fed sho was persuaded to change his mind on this -- go to paper after paper and they all say the same thing, there is a very large cyclical component to our problem. I know he wants to use this argument to push his favorite green policies, etc., but it needs to be grounded in theory and the empirical evidence. On (3), where's the evidence that financial markets are worried? It's not there. And (4) -- the charge that Keynesians/Krugman don't worry about the quality of spending -- is again wrong.
More specifics from Sachs on these arguments:
First, Krugman believes that fiscal multipliers are predictable and large. Thus, a $1 rise in government spending of any kind, according to Krugman, predictably leads to something like $1.50 in higher GDP. Similarly, a $1 cut in payroll taxes leads to something like a $1.30 rise in GDP.
Don't miss the work "might":
The belief in stable, predictable, and large multipliers is belied by both theory and evidence. Households and local governments might simply use a temporary tax cut or temporary transfer, for example, to pay down debts rather than to increase spending, especially because the tax cut or transfer is seen to be temporary. Businesses, concerned about the buildup of public debt, might hold back on business investment in the face of large deficits, anticipating higher taxes in the future.
The original stimulus legislation was overwhelmingly of the form of temporary tax cuts and temporary transfer payments, the kind of deficit spending especially likely to have little effect on aggregate demand. Only $88 billion of the $787 billion stimulus-package was in direct purchases of goods and services by the federal government. The rest was temporary transfers and tax cuts.
(This was not an accident. A critical and predictable weakness of the 2009 stimulus is that House Democrats and the White House negotiated it in just a few weeks. In the unnecessary haste, there was no serious consideration given to long-term needs in infrastructure, for example. With presidential leadership we could -- and should -- have forged a decade-long strategy. ...
He got what he could from Republicans, and it had to include tax cuts (nearly 40 percent). I wasn't happy either, but it is a lot more complicated than just simply asserting "poor leadership." But the weird thing is the assertion about no infrastructure spending. Let me turn it over to Ryan Cooper:
Jeff Sachs ... [has] one of the most bizarre pieces of economic analysis I’ve seen, arguing among other things that 1) the stimulus was too focused on short-term stuff like tax cuts which 2) aren’t effective stimulus anyway (huh?) and 3) should have had much more long-term investment. Listen to him complain...
Nutty aggregate demand issues aside, this is simply ignoring the plain facts. The stimulus did have money for renewable energy ($90 billion in fact), upgrading our rail network, and highway maintenance. A book has been written about this very topic, large sections of which is devoted to lamenting the fact that lazy, irresponsible pundits and reporters then and now keep complaining that the stimulus didn’t have things it in fact had.
Peggy Noonan got hammered for complaining Obama hadn’t done infrastructure spending he had in fact done, but no one was surprised because it’s Peggy Noonan and she’s basically your crazy aunt. But Jeff Sachs is supposed to be a professional economist. ... I would have thought he could at least bother to read a summary of such a huge bill before holding forth. ...
Back to Sachs:
... Krugman argues that the stimulus worked just as advertised, with the multipliers that were predicted, but that other factors held up recovery. ... My own view is that the predicted multipliers were too high in the first place, especially for such a poorly designed tax-and-transfer program, and that recovery is impeded by structural factors. These structural components are not susceptible to a Keynesian diagnosis or to a Keynesian remedy. They require a long-term public investment response that has not been forthcoming.
Again, he talks about my view rather than the actual evidence on these points. Sure, you can find estimates of multipliers that support his position, but the majority of the evidence is on Krugman's side. Also, Krugman said the package wasn't large enough (and could have been constructed better), so there's no real disagreement on that point, this is just another way for Sachs to claim multipliers are small contrary to much of the evidence.
Second, Krugman seriously and repeatedly downplays these structural changes occurring in the U.S. economy. He repeatedly emphasizes that we suffer a demand shortfall, pure and simple, one easily remedied by more stimulus. Yet it's increasingly hard to reconcile many features of the U.S. economy with this view. ...
What are some of the structural problems? These include large-scale offshoring of jobs, large-scale automation of jobs, decline in demand for low-skilled workers, skill mismatches, broken infrastructure, and rising global energy and food prices. These require various kinds of targeted public investment spending, not simply aggregate demand.
This is silly, most of those factors were already present prior to the recession. The recession added a large cyclical component on top of whatever pre-existing structural issues were in place.
To summarize so far, despite considerable evidence to the contrary, Sachs position is that short-run multipliers are small, and even if they aren't there's little cyclical unemployment. So we need to do the things I was pushing prior to the recession instead.
Anyway, trudging on:
In view of these challenges, would a different kind of spending program have worked better? If the spending had been concentrated on long-term infrastructure and job skills, with investments carried out not for two years but over the course of a full decade (as in the 1950s-60s national highway program), the answer is probably yes. But such projects were not "shovel ready."
Such long-term investment programs are very different from quick-and-dirty Keynesian "stimulus" packages such as temporary tax cuts. Long-term investment programs require thinking and planning of the kind that has never happened with Obama's stimulus packages. ...
The Administration should indeed have taken several months in 2009 to design and advocate for long-term investment programs for renewable energy, fast intercity rail, large-scale highway upgrading, large-scale skill and job training, and so forth, rather than rushing to pass a stimulus package of hundreds of billions of dollars of shortsighted and largely ineffective temporary tax cuts and transfer programs. The budget should have paid for such new long-term investments by allowing the temporary Bush-era tax cuts to expire on schedule in 2010 (or by negotiating equivalent revenues of 2-3 percent of GDP per year as the price for maintaining the Bush-era tax cuts).
Again, this is mainly a disagreement about short-run multipliers. Take your time, no need to hurry (other than the millions of people who are unemployed and struggling to make ends meet). Krugman and others (like me) say that yes, of course, shovel-ready infrastructure is the first choice, and of course we need to make progress on our long-run issues. But while we are getting that done, why the hell wouldn't we want to do whatever it takes to alleviate the suffering in the shorter-run?
Back to Sachs:
One of the Obama arguments at the time was that the rush in the stimulus program was needed to avoid a Great Depression. This was and is highly doubtful (though, yes, it is widely accepted). The US economic emergency in late 2008 and early 2009 wasn't really an aggregate demand crisis but a financial crisis. The chaotic failure of Lehman Brothers had led to an intense panic and credit squeeze. The Fed therefore needed to flood the markets with liquidity, which it rightly did, in order to unwind the panic. The Fed's action was the real difference with 1933 (when the Fed allowed the banks to fail). It was the Fed, not the fiscal stimulus, which prevented a fall into depression.
But, again, so what if it wasn't the key to avoiding another "Great Depression" (though he seriously underplays the role of automatic stabilizers in helping along these lines), if we can end suffering -- help the unemployed -- we should do so.
Now the deficit hawkery:
Third, crude Keynesians like Krugman believe that we don't have to worry about the rising public debt for many years to come, perhaps well into the next decade. This is remarkably shortsighted. The public debt has already soared, from around 41 percent of GDP when Obama came into office to around 76 percent of GDP today (and with no lasting benefit to show for it). If Krugman had his way, and deficits were not restrained, the debt-GDP ratio would already be above 80 percent by now and would be rising rapidly towards 90 percent and above (as shown in the recent CBO alternative scenario).
No lasting benefit? Is he serious? The automatic stabilizers alone had a huge benefit, and the discretionary policy helped as well (despite his insistence in every possible way that short-run multipliers are near zero).
Not sure I can go on, but here's more (reacting to my first read, and don't plan to proof what I've written as usual for typos, etc. as I don't feel like reading this again):
... It's true that we've not paid heavily so far for this rising debt burden because interest rates are historically low. Yet interest rates are likely to return to normal levels later this decade, and if and when that happens, debt service would then rise steeply, increasing by around 2 percent of GDP compared with 2012. Many people seem to believe that we can worry about rising interest rates when that happens, not now, but that is unsound advice. The build-up of debt will leave the budget and the economy highly vulnerable to the rise in interest rates when it occurs. The debt will be in place, and it will be too late to do much about it then.
Yes, because of problems that might happen "later this decade" we should let people suffer now (and besides, it's all structural and multipliers are zero).
Let's move to his fourth point:
Fourth, crude Keynesians believe that for all intents and purposes, "spending is spending." ...
No he doesn't. Not at all. Again, this is just an argument that short-run multipliers are zero, blah, blah, blah. Krugman has made it clear that quality of spending matters, but if we can't get infrastructure spending in place in time we should help people in other ways. Sorry if that interferes with Sach's pet issues.
This approach is disastrous both politically and economically. Progressives like myself believe strongly in the potential role of public investments to address society's needs... Spending is not spending. The U.S. needs productive public investments, not wasteful spending. ...
Yes. it's very wasteful to help the unemployed through non-infrastructure spending. Wonder how he feels about food stamps? Are those okay?
Let's go to Sachs' conclusion:
... Mr. Krugman, I believe that you are a crude Keynesian at a time when we need subtler, surer, longer-term policies.
That subtler set of policies should include:
(1) Decade-long public investment programs in renewable energy, upgraded public infrastructure, fast rail, job training and the like; (2) Adequate fiscal revenues (including tolls on infrastructure) to pay for these investments over the course of a decade, including a downward path of the debt-GDP ratio; (3) Increased revenues through taxation on high net worth, financial transactions, high incomes, capital gains and carried interest, offshore corporate earnings, and carbon emissions, and a stiff crackdown on tax havens and phony transfer pricing.
All of this would have been much easier if Obama had started down this long-term path in 2009, and had never conceded the permanence of the Bush-era tax cuts for almost all households. Instead, he followed a populist and shortsighted policy of "stimulus" and tax cuts. ...
We all agree on the need to address the long-run issues, and I have called for infrastructure spending again, and again, and again as a way to help the economy is both the short and longer runs. But that doesn't mean we should ignore other policies -- money spent on things other than infrastructure -- that might help people in the short-run. Short-run multipliers are sufficiently large, there is substantial cyclical unemployment, and out debt problems are not immediate. I hope Jeff Sachs does manage to get his favorite projects in place -- I support them -- but not at the expense of people who have already spend far too long struggling to get by as they look for work.