Tuesday, November 09, 2010

The Conscience of a Cassandra

It's hard not to give considerable weight to Paul Krugman's endlessly repeated warnings -- and complaints once the warnings went unheeded -- that the Feb. 2009 stimulus was too small. In this narrative, the cautious trimming that led Obama to put forward a proposal too weighted to tax cuts and appropriating less than 2/3 the total originally recommended by Christina Romer failed to jump-start true recovery and thus destroyed Obama's political  capital.

Can being right too often be hazardous to your intellectual health? Krugman warned us from Feb 2000 forward about the voodoo math behind the Bush tax cuts; he warned in 2005 about the housing bubble; he warned from February 2009 forward that the proposed stimulus was too small. [Update: He was also right about the the Iraq war and the Euro.] Being right and unheeded creates a Cassandra syndrome: you prophesy, and you feel in your bones that the leadership won't listen. Hence Krugman slips in Glenn Greenwald mode --  concocting a simplified counterfactual in which, under his policies, everything would have been okay, whereas current policy is leading straight to disaster.

As a symptom, take this sideswipe at the stimulus in Krugman's latest crie de coeur about the Fed's refusal to raise its inflation target:
...fearing opposition in Congress, the Obama administration offered an inadequate plan, only to see the plan weakened further in the Senate. In the end, the small rise in federal spending was effectively offset by cuts at the state and local level, so that there was no real stimulus to the economy.
Wasn't one of the core purposes of the stimulus to offset spending cuts by state and local governments? That's what stimulus does -- offset drops in demand. Of course, Krugman believes that there was too little state aid, along with too little of everything, in the Recovery Act, and naturally a stimulus should do more than offset other government spending cuts. But still, offsetting those cuts is a "real stimulus to the economy." The Recovery Act unquestionably preserved hundreds of thousands of state and local government jobs, and GDP would have been lower if this hadn't been done.

This may seem a semantic quibble. But it's a symptom, I think, of an oversimplified narrative. Krugman has implied elsewhere that the right-sized stimulus would have set the economy roaring back to life (as has Martin Wolf). To what degree is that credible? If Obama had asked for $1.2 trillion and got $950 billion, what's the math on the counterfactual? Unemployment at 8.3%?  And the effect of such a drop on the electorate? I confess it could be substantial. But there's an awful lot of what-ifs there, beginning with the premise that a substantially larger stimulus could have got through the Senate. And the narrative leaves out some externalities, such as the Euro sovereign debt crisis, which seemed to stop a decent-looking recovery in its tracks. Not to mention the credible possibility, forecast now by a growing number of economists and business leaders (e.g., here and here), that a substantial recovery may be on the horizon now.

Krugman's critique of the Obama administration has real bite. But neither he nor anyone else knows what would have been, what will be, what the full or long-term effect of the Recovery Act will be, or what the effect of a Krugman-designed stimulus (passed through the Congressional wringer) would have been.

UPDATE: With respect to Obama, Krugman came around in 2012


  1. I agree with your points here. Krugman is often right on economics, though not as infallible as claimed by Brad DeLong, moves into a rant category on politics. He did advocate taking over the nation's largest banks using Sweden's example. Those banks would have been crippled if we followed his advice and the stress tests combined with TARP money were a better solution.
    btw, keep up the good work with the blog, regular reader

  2. I’d add one other important contributing factor to Krugman’s most recent Cassandra inclinations. His strident support for Hillary Clinton and his corresponding invidiousness toward her chief Democratic primary opponent helped to inscribe a well-worn groove of anti-Obamaism in his rhetoric in the run-up to the November 2008 election. And before Krugman turned against Obama again in the aftermath of what Krugman has deemed to be a disastrously inadequate stimulus, my recollection is that there was a certain concern-troll disingenuousness to the support Krugman grudgingly extended toward Obama against Republican assault.

  3. Can't can't agree with this. The fact is a 1% lower unemployment rate means hundred's of thousands of people, who are not working now, would have been able to find employment. They will be materially better off & so will their families. There would be fewer long-term unemployed slowly becoming unemployable. The politics may not have been better at 8.5% vs. 9.5% unemployment, but there will be real, long-term benefits to millions of people who had fallen into hardship.

    The point is Obama got elected on he back of a deep economic crisis. His first order task is to fix the economy, and he clearly didn't do all he could to do so. The Summers gang always felt that all that needed to be done was to stop the bleeding, and with luck, the economy will improve on its own. Well, they weren't that lucky. The stimulus may have stopped the bleeding, but it wasn't enough to to give the economy a strong rebound. Now, a stimulus twice the size may not be enough either, but Obama didn't even try. And then for months afterwards he pretended everything was just dandy, even as he realized the economy was worse than expected.

    As a final point, since Krugman is always right, that he is so worried about a Japanese-style deflationary spiral should make us all worried. Now, as you say, the counterfactuals are not there. Perhaps we are doomed to Japanese style economic stagnation no matter what Obama did, but the point again is Obama didn't even try very hard to fight that possibility. And the consequence at this point is that there is nothing we can do about it except to hope we are very lucky indeed.

  4. It is a semantic quibble. You do not appear to disagree with Krugman's central contention that the stimulus was too small. Of course no one knows the what would have happened with a larger stimulus, but we do know what happened with the stimulus we got--9.6% unemployment and an electoral shellacking. The counterfactual case is important because the prevailing Republican narrative--that the stimulus just made things worse--needs to be debunked. Why? Because taking fiscal policy off the table as a tool to fight another economic crisis would be a huge error.

  5. Doesn't passing Krugman's ideas "through the Congressional wringer" kind of miss the point? The impression I get from Krugman is that he considers the Congressional wringer a critical point of failure responsible for the stimulus being as inadequate as it was; he's arguing for larger stimulus as final system output, not merely a bigger opening bid as input to Congress.

  6. Maintaining state and local jobs many of which are funded by government with all the perks..ie healthcare -unions-pension etc while many in the rest of the population are losing theirs deosnt look like stimulus if your an out of work taxpayer who isnt lucky enough to work for the government,it looks like a bribe.Thats a problem.
    And since Obama didnt go big with the stimulus he went just enough to keep our nose above the water it still left people feeling panicked and insecure,especially since so many lost alot of equity the last few years.Obama is no FDR and even he seems to get the lack of real reform anf help from his economic team has hurt him ie his "snark" about "heck of a job Summers".

  7. I understand your point but don't agree. The point of stimulus is not just to make up for lower state and municipal demand. The point is to substitute gov't demand for falling demand as a whole. It is my understanding the Krugman used one or more models to estimate the amount of stimulus needed. The condition of states continued to worsen raising the question whether his original analysis was sufficient. In any case, the stimulus did not adequately replace private demand and the recession deepened. On the bank issue, there remains a real question whether we wouldn't have been better following Krugman and Barry Ritholz for example. Large banks are not lending and may be insolvent. They are facing lawsuits, as well, over misrepresentations about the content of CDOs. Arguably, a FDIC process, if the gov't had the skills to do it, would have been more effective. It's worked for hundreds of banks in the last two years. Generally speaking, I have to agree with Krugman's critique. It's the same as other Nobel prize winners (including this year's winner) and Roubini, among many others. Substantive analysis versus politics. Substantive analysis will generally be correct every time.

  8. “Krugman has ‘implied’ elsewhere that the right-sized stimulus would have set the economy ‘roaring back to life’”

    What you have there are weasel words, and a straw man to boot. It’s almost overwhelming and somewhat breathtaking ;). Where’s the attribution? Apparently you’ve missed Krugman’s consistent thesis through multiple pieces on the stimulus, which has been to address both its political and practical elements. See for example http://www.nytimes.com/2010/10/11/opinion/11krugman.html?ref=paulkrugman

    Krugman recognizes that the ultimate impact of a larger stimulus is an unknown. What he’s claiming, with some very clear support, is that having failed to at least advocate a larger stimulus, Obama has put himself into the tough political position of defending a policy that he stated would be adequate for goals that have not been met. Obama, and by extension the Democrats, have paid a steep price for this politically.

    As an economist, Krugman optimistically suggests that a larger stimulus would have had a greater impact on the economy. As a pundit, Krugman astutely observes that a strong case for a weaker stimulus has had political consequences for Obama and the Democrats.


  9. I have to confess that I agree with most of these comments. The bone I picked with the Krugman passage cited is indeed semantic; my argument is that the semantic sleight-of-hand is a gateway to overdrawing the necessarily hypothetical contrast between the stimulus we might have got and the one we did get. I do accept that what we got was too small, though. As to the last anon comment, I plead laziness -- or rather, lack of workday time -- in not locating the Krugman column (there are so many on this subject!) in which he envisioned a virtuous cycle arising from a large-enough stimulus. But he did definitely do that. I'll see if I can find it later.

  10. You can't know what would have happened, but you can have a pretty good idea. Basically, thPis election was two close Senate races -- Illinois and Pennsylvania -- from being declared a 1982-like wash. That there is a stepping back from the 2006-08 cycle in the House can't be denied. But if this had come out as four Senate seats, three in deep red states, while Dems pick up Govs in Minn., Calif. and Conn. (darn Florida) then the politics are radically different. People would have said, win some lose some, and the House seats would have been seen more in the context of giving back all the unholdable seats from the last two elections, plus a normal off-year effect. That's very different. And it's not at all hard to see how lower unemployment would have tipped Ill., Pa., Ohio governor and maybe even Florida governor (a seat that has been red since 1998).

    It's also true that the hiring outlook seems to be improving palpably. I happen to be out of work and I can see it clearly in my field. I won't be out long.

    Dems like me just have to count on Eric Cantor et al -- plus the 1984-like effect of a cyclical recovery that might be pretty sharp -- to win us the next election. Fortunately Eric is up to the task.

  11. I agree with others that this is a bit of a strawman analysis. I've read a lot of Krugman's writing as well and don't think you're accurately characterizing his positions by saying that he thought "everything would be okay" and the economy would come "roaring back to life."

    My general takeaway was and is that a too-small stimulus would be bad macroeconomics and bad politics. It would be bad macroeconomics because, based on his analysis, which generally matched up with what Romer did, a $1.2 trillion package would stand a chance of restoring historical levels of normal growth and unemployment, assuming that the projections for the worst-case scenario turned out to be correct. Of course, they didn't; the worst case turned out be 10% unemployment, rather than 9%, and even larger drops in GDP growth. So, even $1.2 trillion probably wouldn't have been big enough, but it would have been closer, reducing the suffering caused by the recession.

    His argument that it was bad politics was that, even if the stimulus proposed was successful, it would still leave unemployment at a level that Republicans could call it a failure, which they did. As a corollary, he suggested that proposing something larger and having Republicans cut it back would force them to share in responsibility for its shortcomings, while also energizing the base.

    I'm not sure what parts of those positions are overreaching counterfactuals.

  12. I agree with the post above that reminds that one of Krugman's central jeremiads is that good politics is bad economics, with Congress Exhibit A.

    Krugman's only slip here was not to add "and no/little more" in between "level," and "so".

    With the linguistics match up with what I think is quite obviously his argument, namely, that no *net* stimulus occurred. So yes, it's a valid semantic quibble, and no more than that.

    Bear in mind Krugman wants stimulus to offset the slump in private, as well as public, spending.

    I don't think its incumbent on him when making the case for this to soothsay that X amount in stimulus equals Y amount in GDP improvement/ unemployment rate/whatever. In fact I can't think when he has ever made a calculation of this sort.

    What he says is the principle of fiscal stimulus at the zero bound is sound, when done big and done right, with reference to the amount needed previously (i.e. a lot).

    Krugman compares this with the ample evidence that the current approach of doing it softly softly catchee monkey doesn't work.

    So I agree with the other commenters here there is some well-intentioned straw man-ing going on here.

    If you want specious counterfactuals (or as I call them, after Kipling, 'Not So Stories' - or perhaps 'So What Stories'), Niall Ferguson's your man.