Streetwise Professor

August 9, 2010

Earth to Krugman: “Data” and “Anecdotes” are Not Synonyms

Filed under: Economics,Politics — The Professor @ 12:15 pm

George Stigler once wrote “‘data’ is not the plural of ‘anecdote.'” Apparently Paul Krugman didn’t get the memo.

In today’s NYT, he relates a modern-day Dickensian tale of the penury into which states and local governments have been cast as a result of eeeeevvvvviiiiilllll anti-government ideology, allegedly regnant since 1980:

How did we get to this point? It’s the logical consequence of three decades of antigovernment rhetoric, rhetoric that has convinced many voters that a dollar collected in taxes is always a dollar wasted, that the public sector can’t do anything right.

The antigovernment campaign has always been phrased in terms of opposition to waste and fraud — to checks sent to welfare queens driving Cadillacs, to vast armies of bureaucrats uselessly pushing paper around. But those were myths, of course; there was never remotely as much waste and fraud as the right claimed. And now that the campaign has reached fruition, we’re seeing what was actually in the firing line: services that everyone except the very rich need, services that government must provide or nobody will, like lighted streets, drivable roads and decent schooling for the public as a whole.

So the end result of the long campaign against government is that we’ve taken a disastrously wrong turn. America is now on the unlit, unpaved road to nowhere.

If you read the entire piece, you’ll note it is chock full of anecdotes.  Data not so much.  Not at all in fact.

So let’s look at the data.  I know it’s an onerous task to collect it, but I’m willing to make any sacrifice for the truth. The humanity!  I’m sure the compassionate liberal Krugman wanted to spare his overworked armies of RAs the indignity of such backbreaking labor.

In fact, it took me all of 10 minutes.  But I’ll recover.

From the beginning of the Dark Age, with the ascension of Darth Reagan to the presidency in 1980, state and local government spending as a fraction of GDP rose from 15 percent to 22 percent.  Even stopping at the time of the financial crisis, in 2008, state and local government accounted for 19.6 percent of GDP.

In other words, state spending grew faster than the economy during the entire Age of Anti-Government Ideology.  The rate of increase of that percentage was actually higher in the 1980s, under Reagan, than in the 1990s, under Clinton.  Indeed, it was higher still in the 2000s, when the new Dark Lord was president.

Insofar as state and local debt is concerned, in 1990 it totaled $993 billion.  In 2000, $1.28 trillion, and in 2008 $2.25 trillion.  Note the high rate of growth during the allegedly anti-government Bush years as opposed to the Clinton period.

There is clearly no prima facie case that states and municipalities were starved of resources at any time over the past 3 decades, be it the result of antigovernment rhetoric or any other damn thing.

One could argue that spending should have been higher, that the growth observed absolutely and relative to the entire economy, was too small.  In which case, you need to present some empirically testable hypothesis, and then go out and test it.  Which Krugman, of course, does not even try.  Sorry, teary anecdotes don’t cut it.  Period. Have some data?  Fine. We can talk.  Otherwise, forget it.

Put differently, what are the testable implications of the anti-government ideology theory?  How is relative growth consistent with this theory?  Until Krugman specifies those hypotheses, and actually tests them, he’s just blowing gas.

Yeah, like  that’s a new development.

As for governments’ current strapped fiscal condition, here’s an alternative explanation which happens to be consistent with the actual, real, you know, data just discussed.  States and localities have been financially imprudent in recent years, increasing spending at an unsustainable rate.  Governments assumed that the good times would always last–we were in a new paradigm, dontcha know–and entered into commitments that were unaffordable unless les bon temps continuer a rouler.  When the music stopped, they were already greatly overextended.  They are currently in the painful process of adjusting their ends to their means.

Under this interpretation, throwing more money at states and localities is just so much fuel for the moral hazard fire.  It relieves them of any accountability for their fiscal decisions, laying the groundwork for the next crunch.

Krugman rationalizes this with more vulgar Keynesian stimulus reasoning.  But the money has to come from somewhere, and as much as Krugman is loath to admit it, people are sure thinking like Ricardians these days, hence their clear aversion to vastly increased public expenditures.

Print Friendly, PDF & Email


  1. […] This post was mentioned on Twitter by Craig pirrong, Craig pirrong and Craig pirrong, Craig pirrong. Craig pirrong said: Updated my SWP blog post: ( ) […]

    Pingback by Tweets that mention Streetwise Professor » Earth to Krugman: “Data” and “Anecdotes” are Not Synonyms -- — August 9, 2010 @ 12:49 pm

  2. Once Krugman became a media darling he realized there was a lot more money in getting paid by the word and as opposed to being paid by the intelligent comment. Hence, we get many words and few intelligent comments from him. At least he has done something lately to show he understands economics.

    Comment by Charles — August 9, 2010 @ 2:30 pm

  3. […] Earth to Krugman: "Data" and "Anecdotes" are Not Synonyms (Streetwise Professor) tweetmeme_style = 'compact'; tweetmeme_service = ''; tweetmeme_source = 'ReformedBroker'; […]

    Pingback by Have Some Data? (When Economists Tug on the Heartstrings) The Reformed Broker — August 9, 2010 @ 8:33 pm

  4. Speaking of data, can you cite or post yours?

    The reason I ask is that Paul Kasriel @ Northern Trust just had a piece that State and Local spending averaged under 13% real GDP during the 30 years 1977-2007. You two are pretty far off; more than I would expect for a couple of year difference of domain. See page 7 of the link below.

    Thanks… It’s more convenient of me to ask you since Northern Trust doesn’t have a comment feature.

    Comment by Chris of Stumptown — August 9, 2010 @ 10:51 pm

  5. it’s even simpler than you made it…

    The lights are off in Colorado Springs because that’s a choice the citizens made – they didn’t want to pay for it. Simple. Full stop. To suggest (as Krugman did) that they need a Federal subsidy/bailout is LUDICROUS.


    Comment by Kid Dynamite — August 10, 2010 @ 8:12 am

  6. Coming from biomedical science, I’m with you on data vs anecdotes. It’s stupid to claim something is true without empirical evidence. However, I have to say that economics as a field is frustrating to read about. It’s hard to think of any other science in which large swaths of “experts” come to essentially opposite conclusions about basic questions of fact. I honestly don’t know who to believe.

    Comment by Jeff — August 13, 2010 @ 12:01 pm

  7. @Jeff–understood and acknowledged. Now you know how economists feel when reading about medical studies that draw the exact opposite conclusions about basic questions of fact 🙂 But it’s better to be arguing over data and interpretation of data, than to make Olympian pronouncements completely detached from the data at all.

    The ProfessorComment by The Professor — August 13, 2010 @ 4:33 pm

  8. Anecdotes =/= Data.
    Good. Discredits 99% of Western Kremlinology.

    Comment by Sublime Oblivion — August 14, 2010 @ 2:33 am

RSS feed for comments on this post. TrackBack URI

Leave a comment

Powered by WordPress