My bet with Bryan Caplan – update


Back in 2009, I made a bet with Bryan Caplan that the average unemployment rate in the EU-15 over the following 10 years would be no more than 1.5 percentage points above that in the US. Before talking about the bet itself, I’d like to note that while we disagree about a lot of things, Bryan and I both take a strong stand against war, with a limited exception for self-defence. As Bryan says here, that takes a lot of sting out of the possibility of a losing bet for either of us – agreement on war and peace is more important than disagreement about labor markets in my view. 

Now, on to the bet.

First the numbers. Until now, I’ve been consistently ahead. EU-15 and US unemployment rates were very close during 2009 and 2010. A gap has opened up in the last few months and is now about 1.5 per cent. Given the dismal prospects for the EU economy in the coming year, and the likelihood of some kind of recovery in the US, I expect to be losing ground on the bet this year, and probably for a couple of years after that.


Having said that, the original premises of the debate have pretty much ceased to apply. I was expecting a longish recession followed by a gradual recovery in both the US and EU. Macro policy has been worse than I expected in both cases, but dramatically more so in the eurozone, where the ECB seems determined to turn a recession into a depression. It’s this, rather than differences in labor market performance that’s driving the present divergence in unemployment rates.

Looking specifically at the US labor market, it’s hard to see any signs of the flexibility benefits that might be expected to offset the negative effects of a combination of employment at will, an almost non-existent union movement and minimum wages that are substantially lower, in real terms, than they were 30 years ago.  While unemployment has fallen from its peaks, there is no sign of a recovery in the more relevant measure of labor market performance, the employment-population ratio.

15 thoughts on “My bet with Bryan Caplan – update

  1. I don’t understand the bet. The way that figures are calculated is different in each country, and visibility of unemployed is different. – Unemployment in europe is significantly higher than the USA, and the cost of everything is higher, and people can afford to consume far less.

    So the fact that you’re arguing that 1.5 points describes the difference between two completely different sets, makes sense as a measure of the relative change in each set of numbers. But total unemployment, underemployment and underconsumption in, say, Germany or Italy versus the US, is staggeringly higher.

    1) You can’t draw the conclusion that the institutions are superior from the difference between the numbers you’re using.

    2) Both ‘institutional methods’ (the communal method of the smaller european states, versus the ’empire’ method of the vastly larger, similarly diverse USA) are in relative decline versus the methods of the totalitarian capitalists – and there is no certainty whatsoever that the european countries can persist as legal and cultural entities under their current models any more than there is the certainty that the USA can persist as a legal and cultural entity — certainly not to the degree that the more racist and culturist authoritarian states appear to.

    Love the bet. But your conclusion doesn’t follow from it tho. It takes as normative that which is exceptional.

  2. “But total unemployment, underemployment and underconsumption in, say, Germany or Italy versus the US, is staggeringly higher.”

    Sadly, no. Look at the employment-population ratios, as I said in the post (it’s useful to focus on prime-age males since very few of them are engaged in study, childraising or other productive non-employment occupations). What you’ll find is that the US is in the middle of the pack.

    The only feature where the US is exceptional is in the very long annual working hours of full-time workers. That fact accounts for the entire difference between US output per person and that of the leading European countries.

  3. I second Doolittle’s comment. You are likely to lose as a mere consequence of the difference in the way the rates are calculated. The amount of people employed has hardly budged but the unemployment rate is going down in the US.

    At least you are owning up to it!

  4. While theories of unemployment, microeconomic change, and long working hours are on the table, I’d like to ask Prof Q for an update on something. In his recent paper for the RBA on “Productivity – the lost decade” Saul Eslake mentions:
    “Quiggin (2000, pp 269–270) suggested that much of the apparent increase in measured productivity growth during the 1990s could be explained by measured (or unmeasured) increases in working hours and by increases in ‘work intensity’. To the extent that either of these suggestions are accurate (and they were contested by other participants at the conference – see Gruen and Shesthra (2000, pp 275–276)), the increases in recorded productivity growth rates for which they accounted would also have been unsustainable.”

    I skimmed the 2000 collection some time back and IIRC the chief argument against JQ’s suggestion was the lack of evidence one way or the other – “interesting hypothesis but we don’t have the data”.

    My question is, given that Fear and Denniss at The Australia Institute have now quantified the extent of unpaid overtime in Australia, how well does the “unmeasured increases” argument now stack up, vs the “microeconomic revolution” argument?

  5. @Garrett You clearly don’t understand the data. If the amount of people employed has hardly budged but the unemployment rate is going down then people are giving up and withdrawing from the labour force.

  6. It is nice to see agreement between people on both sides of the political spectrum on warfare.

    However it seems ironic to think that you could well lose your bet because the economy you speculated against followed moderately expansionary policies while the other embraced austerity. In this sense maybe you and Bryan should consider reversing positions (assuming he is an advocate of austerity)?

  7. @John Quiggin
    Yes, I know that the difference in productivity is the result of hours worked – and that in the use, it’s the upper classes (educated) that are both working extra, and earning extra. But while I understand that you’re using the employment ratio data for the USA, I don’t see that you’re using the same for Germany (or Europe for that matter). Now, it’s possible I don’t understand the Eurostat data, because I only look at it once or twice a year. Germany is more disciplined about keeping it’s lower classes productive. But everything I’ve seen is the the unemployment rate there and here is understated (which you demonstrate with the BLS ratio data.)

  8. “everything I’ve seen is the the unemployment rate there and here is understated (which you demonstrate with the BLS ratio data.)”

    Correct. And correctly measured, unemployment is about the same in the US as in most EU countries – there are some long-term outliers like Spain, and peripheral countries like Ireland where the crisis has hit very hard, but if you compare the US with Northern Europe you won’t find much difference *except* in hours worked per employed person.

    Here’s my post on the topic

  9. I would love you to follow up with a longer post some time about the slow drop-off in US unemployment despite record corporate profits and, as you say, despite the labour market flexibiity that is (theoretically) supposed to keep unemployment low.

    This runs counter to all the arguments thrown about in Australia about Work Choices and how the additional flexibility will reduce unemployment. Given that various right wing idealogues are back arguing in favour of Work Choices lite (ish), it would be great to have some proper analysis of the reality.

    I have seen no discussion about this anywhere – apart from your brief comment at the end of this article. Maybe I’ve missed it, but surely it’s fundamental to the whole economic argument.

  10. Thanks for that Prof Q. It’s always striking to see how many of our “empirical conclusions” are based on various statistical illusions, measurement mismatches, pattern-seeking behaviour and the like, and it sounds like you made a good case that the productivity gain was one more of those.

  11. @John Quiggin I think you misunderstood my comment. My point was the reason you were losing not because of actual progress in reducing the number of unemployed, but because the official unemployment rate is not counting unemployed people for the reasons you mentioned.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s