THE PROFESSORIATE FAILS US AGAIN: Harvard professor will travel for cash!

MONDAY, APRIL 29, 2013

Part 1—Much as we have told you: Early in 2010, a pair of Harvard professors authored a bungled study.

There’s certainly nothing new about that! Did you notice the horribly cherry-picked column Harvard assistant professor Jal Mehta recently published in the New York Times?

We did! For part 1, click here.

There’s nothing new about the failure of the ranking professoriate. And as noted, the paper we’re discussing today appeared in early 2010. According to our calculations, that was three years back!

Unfortunately, the bungled study turned out to be extremely influential. Its findings have been used to advance the idea that austerity measures are just the ticket in the case of our floundering economy.

The bungled study came from the desks of Reinhart and Rogoff, a pair of Harvard professors. And uh-oh! According to Paul Krugman, the bungled work of these professors “played right into the desire of many officials to ‘pivot’ from stimulus to austerity.”

The paper “became famous” overnight, Krugman wrote in this recent column. “It was, and is, surely the most influential economic analysis of recent years.”

In short, the bungled paper helped advance a policy agenda which powerful interests already hoped to pursue. For reasons we haven’t seen explained, three long, bungling years passed.

And then, finally! At some point in the past year, three graduate students at UMass requested access to Reinhart and Rogoff’s data. Three years after the paper appeared, these graduate students have now revealed the professors’ groaning errors.

In Krugman’s April 19 column, he offered a shorthand account of the kinds of mistakes involved in the professors' work. This was his account of the Harvard professors’ mistakes:

“First, they omitted some data,” Krugman wrote. “Second, they used unusual and highly questionable statistical procedures; and finally, yes, they made an Excel coding error.”

We’re not experts on this type of work, but that sounds like a lot of mistakes to us! That said, the coding error has received the most attention, perhaps because it seems so clownish. Somehow, the Harvard professors failed to include five major countries (out of just twenty) in one of their major computations.

When Australia, Austria, Belgium, Canada and Denmark were returned to life on earth, the findings of the professors’ study turned out to be substantially different, as the three graduate students showed. Omission of data is like that!

We’ll discuss the graduate students tomorrow. Just for today, can we talk?

For years, we’ve warned you about the rampant failures of our nation’s intellectual elites. We’ve often warned you about the errors of omission committed by the professoriate—about their endless failure to intervene in our bungled policy discussions.

We’ve warned you about the hapless performance of the nation’s “educational experts.” We’ve warned you about the unimpressive professors we sometimes get handed on The One True Channel.

This bungled study by Reinhart and Rogoff doesn’t involve a failute to speak when others are bungling badly. This is a case where a pair of Harvard professors engaged in affirmative bungling of a very serious type.

At any rate:

For reasons we don’t understand, three long, high-profile years went by before the professors’ bungles were spotted. And how strange! Even in the face of skepticism about their study’s conclusions, the professors themselves had never gone back and noticed that five countries were MIA from one part of their study!

After three long years, it took a trio of graduate students to reveal the professors’ mistakes—mistakes which drove “the most influential economic analysis of recent years.”

Before the week is done, we’re going to look at several reactions to this high-level bungling. We'll look at Ezra Klein's approach to this mess, and at Krugman's sardonic reaction to Klein.

For today, we’ll suggest that you look at something Matt Yglesias said.

Brother Yglesias went to Harvard! In the past, he has even gone so far as to refer to some Harvard professors as “smart.”

Despite these troubling manifestations, Yglesias has authored a tart insinuation about one of the profs now under review. His insinuation appears in a recent piece at Slate.

Who the heck is Professor Rogoff? For ourselves, we have no idea. But Yglesias caught our eye with the street-fighting passage we are happy to highlight:
YGLESIAS (4/26/13): [T]his isn't just some sad case of conservative politicians running around mischaracterizing a sober-minded study and then liberals overreacting in response. Ken Rogoff was writing op-eds drawing strong policy conclusions from this paper. He was delivering congressional testimony drawing strong policy conclusions from this paper. And it's not as if he's some political naif who stumbled down from the ivory tower into a partisan controversy he could never have predicted. He was research director at the International Monetary Fund and he knows how the game is played. He's signed up as a paid speaker for the Washington Speakers Bureau. His "fees vary based on event location" but they promise that in exchange for your money "Kenneth Rogoff reaches beyond the theoretical and delivers quantitative proof from his frequently cited research and best-selling book to explain why our financial history continues to repeat itself—and just where the US and global economies are heading."
Professor Rogoff has signed up with the Washington Speakers Bureau, where his fees “vary based on event location.” “In exchange for your money” (Yglesias’ term), he was willing to fly to the resort where you were holding your convention.

At this location, he would describe the deathless findings of his bungled research.

Just so you’ll know: The Washington Speakers Bureau tends to charge hefty fees. Austerity-friendly corporate groups are the types of folk who pay them. This seems to be the insinuation lurking in Yglesias’ profile.

To which, we say hurrah!

Why have our professors so constantly failed us? We’ve tried to get you to ask that question for many years.

We’ve asked you why our “educational experts” seem to churn so much bullshit and cant. We’ve asked you why our esteemed professors can’t seem to step up to the plate and challenge prevailing nonsense of various types.

(Was there really no professor of logic who noticed that Candidate Gore didn’t say he “invented the Internet?” With all the logicians in our employ, why didn’t one of them speak?)

We’ve asked you why our leading logicians don’t challenge our press corps’ relentless bungling. We’ve begged you to notice another fact: sometimes, our fiery progressive professors just aren’t all that sharp.

For years, we’ve asked you to note the way the professoriate continues to fail us. All week, we’ll look at the way these latest professors—from Harvard, no less!—failed in 2010.

We’ll also ask you why it took three years for someone to notice these latest howlers. We’ll ask why it took three graduate students to shoot down this latest crap.

Tomorrow: Where were all the other professors? Off in the south of France?

For extra credit, read ahead: After three long years, it took a trio of graduate students to reveal the professors’ errors.

The BBC has described their detective work. For extra credit, click here.

23 comments:

  1. Bob - I believe that the co-authors of the report by Thomas Herndon, Michael Ash and Robert Pollin, are both professors at U Mass, Amherst. While they once may have been graduate students, they have certainly advanced since then. Probably, it was a good idea for Herndon to get some more reputational heft behind his conclusions.

    ReplyDelete
  2. Is it true that no other professors recognized Reinhart's and Rogoff's bungling? Didn't Krugman and others note that the study seemed off? Didn't other professors repeated request Reinhart's and Rogoff's data sets, only to be rebuffed?

    Haven't there been over the years a number of professors who do speak out, but in general get ignored by the mainstream because, as Noam Chomsky (one of the most important in this group), the establishment media help to "manufacture consent" for the policies of the elite and the plutocracy? So is it fair to keep snapping that professors--who, if you mean tenure-track faculty, are a dwindling portion of the higher education system--are off "in France"? Why do you keep saying this? And why don't you acknowledge the many professors--scholars, critics, bibliographers, writers, artists, etc.--who do speak out, forcefully and effectively, but never can get a hearing on the One True Channel or elsewhere?

    Also, Reinhart and Rogoff are both linked directly to billionaire Peter G. Peterson, one of the chief plutocrats seeking to destroy Social Security, Medicare, and all social insurance programs.

    ReplyDelete
    Replies
    1. Typing too fast...

      "Didn't other professors repeatedly request Reinhart's and Rogoff's data sets, only to be rebuffed?"

      and

      "as Noam Chomsky (one of the most important in this group) has argued, the establishment media help to "manufacture consent" for the policies of the elite and the plutocracy?"

      Delete
    2. Exactly. In fact, you needn't cite Krugman (Princeton professor) or Chomsky (MIT professor, though not an economist, of course). Just about every Econ professor anywhere, of any note in the econ academic world, has been full of despair at the business and journalistic world's ignoring them re austerity (just read Krugman!). Don't know where Bob S's larger diatribe v. academics comes from (gee, didn't he graduate from Harvard? did he learn nothing of value there, or is he some genius autodidact?), much less this specific diatribe v. economists who are academics. But it gets boring and, more important, is bad analysis.

      In a phrase: Ad hominem.

      Delete
    3. It is quite clear from Somerby's style that he is speaking to the undercurrent in news reports that believes that if an idea comes from a professor, then it automatically has credibility. Of course, he goes over the top to try to shake up some who reflexively believe this, but do not see it in themselves.
      Exposing to us these unspoken mechanisms is his most important contribution. He explains in great detail how writers use various methods to convince readers to believe ideas without actually having a discussion about the ideas themselves. Quoting studies from professors is one of these tricks. If a reader has a healthy skepticism, this will not be enough to convince, even if the professors are from Harvard.
      Those who cry that he is being prejudiced against professors are probably just as prejudiced positively for them. But Bob is joking.

      Delete
  3. I think there are a lot of mistakes in academic publications. Here are 3 examples:

    -- My wife published a letter in a medical journal pointing out errors in the statistical analysis done by Dr. Paolo Zamboni regarding a method he promotes for treating multiple sclerosis. This issue means a lot to my wife, because our niece suffers from MS and was considering Zamboni's controversial and dangerous treatment.

    -- A friend, also a statistics professor, published an article essentially accusing a group of Indian researchers of falsifying some data in their medical studies.

    -- Michael Mann originally derived his "hockey stick" via a faulty statisitical approach that was shown to produce a hockey stick shaped output regardless of the input.

    These three errors might have been avoided if a statistician had been a part of the research group. American medical journals typically require that any statistical analysis be validated by a qualified statistician. I think that's an excellent practice. Statistics looks easier than other fields of science, but it has traps that can lead to subtle errors.

    ReplyDelete
    Replies
    1. Medicine is an applied field. Doctors do not typically receive much training in research methods or statistics in the traditional M.D. program. For that, one needs to do a concurrent Ph.D. or take extra training in research methods. That means that most articles in medical journals are easy pickings for someone with statistics training. However, one cannot generalize from medicine to all research fields.

      Delete
    2. One of the issues in academia, or at least among academic economist, is the dearth of replication research. When I was in grad school, we were encouraged to do replications by our very knowledgeable professors. However, being smart and forward looking graduate students, we saw that you couldn't get shit published if it was a simple replication even if the results called into question the original research. So very few of us undertook such low yield research. This is a major problem!

      Delete
  4. From a policy POV, although the 90% tipping point was wrong, Krugman honorably acknowledges that other researchers typically found some correlation between high debt and slow growth. That's the difficulty faced by the US.

    In the short run, our huge annual deficits may be helping to goose the economy. Evidently Congress and the President are unwilling to take the drastic steps necessary to end the deficits. However, in the long run the growing level of debt slows the growth of our economy.

    ReplyDelete
    Replies
    1. http://www.cepr.net/index.php/op-eds-&-columns/op-eds-&-columns/crashing-the-90-percent-club-the-importance-of-the-reinhart-rogoff-error

      This column [by Reinhart and Rogoff] is careful to halfway walk back the main claim of their famous paper, telling us:

      "Our view has always been that causality [between high debt levels and slow growth] runs in both directions, and that there is no rule that applies across all times and places."

      It is good to hear the reference to causation from slow growth to high debt and that "no rule applies across all times and places." However it is worth noting that Reinhart and Rogoff never felt the need to use their access to the NYT's opinion pages to correct all the politicians who used their paper to argue the exact opposite: that their paper implied that countries with high debt levels could anticipate long periods of slow growth.

      Delete
  5. Don't worry about the long run, DinC. If you use the money you're spending to create the deficit on employment (instead of say, tax breaks for those who don't need them--i.e. the well-off), the deficit will shrink in the long run, because more people will be working and making enough money to pay income taxes (not just the 53% who are not unemployed/ under-employed).
    The alternative is to tax those who make a lot of income (or better yet, tax wealth) at a higher rate, and if we do that we get the whining about "taxing the rich".
    If you were really an actuary, I'll assume you know enough about math to see how obvious this is.

    Berto

    ReplyDelete
    Replies
    1. Berto, I know enough to distinguish between theory and definite reality. You've stated an economic theory. The reality is that since around 1999, the ratio of employment to population has been plummeting, with a brief, temporary lull during the Bush Presidency. Will the ratio of jobs to population go back to where it was during the Clinton Administration? Or, is lower employment now a permanent feature of our economy? Whatever the answer is, it's not obvious.

      Delete
    2. BTW Berto, the fact that I'm worried about the long term is a clue to my profession. Worrying about the long term is what actuaries do.

      Delete
    3. That graph don't show what you say it does, Mr. Self-proclaimed smartguy.

      What it shows is a continuing uptrend in employment-to-population ration, interrupted by periods of recession.

      The most recent recession was hugely damaging to employment, but if there's any evidence that the employment rate following the recession is a structural feature of the economy you haven't come close to showing it.

      Choosing, as you do, a peak year just prior to a recession is the work of a manipulator or a fool. We know which one you are, David.

      Delete
    4. if there's any evidence that the employment rate following the recession is a structural feature of the economy you haven't come close to showing it.

      Past recessions have been followed by periods of sharp growth in employment. You can see that on the cited graph for the recessions that ended in 1982 and 1992. The current recession ended 4 years ago. Yet, the percentage of people employed is lower than it was at the end of the recession. That seems to me a good reason for concern that we may not see the kind of employment recovery that we did in the past.

      Delete
    5. Well, this time local and federal government as a net has imposed contractionary policy following the recession -- that's not typically the case.

      But in any event, you have no valid response to the accusation of cherry-picking your start date, 1999.

      And so your reputation as a thoughtful commenter here takes another well-merited beating...

      Delete
    6. Anon -- I picked the start date I did, because that's when the downturn inemployment started. I showed 65 years worth of data. The severity of the employment drop in recent years is unmatched during the entire period.

      The point is, by calling the last 14 years just another recession, I think you are implying tht employment will automatically pick up again. I hope you're right, but I think future levels of e employment are undertain.

      Delete
    7. DinC,
      Am I supposed to believe that the idea that more working people, making more income will increase tax revenues is just a "theory".
      If that's the case, then if I had an apple, and I bought 2 more apples, it's just a "theory" that I would then own 3 apples.

      Berto

      Delete
    8. Dude, *I'm* not calling this "just another recession" -- it's you who are implying that.

      But before you begin flailing about, attacking me, you might do well do admit your own failures of thinking and logic. Yeah, cold day in hell, etc...

      To know how this time's very different, we might begin by looking at how very differently government reacted to the recession.

      Local and federal government employment, for example is sharply lower in this recovery than the one under Bush. Under Bush government employment grew as a response to recession. Under Obama we have not grown government employment.

      We've imposed austerity on government in reaction to this recession.

      Maybe that was a big fucking mistake. And one imposed by the followers of your school of thinking (if I may be so bold as to call your sad efforts "thinking") and that of the touters of R&R's wretched work.

      Delete
    9. Your problem is that you assume that austerians are reacting to events in the world. Instead it is a religious war against government. Government is always bad. Smaller government is always better than bigger government. The slippery slope starts with the first law. Freedom is less government.

      Of course, this is true, but only for the rich. The citizens without wealth need government to protect them, because history has shown again and again that the rich will eat us all, given a chance. This is why the current U. S. Supreme Court is such a cruel joke: they are creating a set of protections to safeguard the only group in our culture that doesn't need it, the rich.

      Delete
  6. "For reasons we don’t understand, three long, high-profile years went by..."

    Oh, come on!!

    You and everyone else know full well what the reasons are. The plutocrats, Bob, after all have their own "tribalism" too!

    When you are being told what you want to hear (or more exactly, what you want the rubes to hear) you're not very likely to go checking that all the Is are dotted and Ts crossed...

    ReplyDelete
  7. You all should read this:
    http://www.japantimes.co.jp/opinion/2011/07/25/commentary/toadies-to-the-debt-to-gdp-ratio/#.UX7KjcocN5o

    There were professors saying: what the ...

    ReplyDelete
    Replies
    1. Thanks for the link, Anonymous. Shiller's an awfully smart guy. It's always good to see what he has to say. He's worried that we'll demand fiscal-austerity programs too soon. That people are asking governments to cut expenditure while their economies are still vulnerable.

      Well, IMHO there's little or no chance of the federal government implementing austerity. As the Fact Checker points out , federal spending used to be around 20\% of GNP. Under Obama it's been around 23% to 25%. It's gone up every year:

      2008: $2.98 trillion

      2009: $3.27 trillion

      2010: $3.46 trillion

      2011: $3.60 trillion

      2012: $3.65 trillion

      2013: $3.72 trillion

      With interest on the National Debt getting more and more expensive and the same for entitlements, and with ObamaCare about to kick in, I don't see any chance that federal spending will go back down to 20% of the GNP.

      Delete