Bryan Caplan and I have now agreed on the settlement conditions for a bet on US_EU jobless rates while also agreeing to differ on the interpretation. The stake is $US100 and the agreed criterion is that, for Bryan to win, the average Eurostat harmonised unemployment rate for the EU-15 over the period 2009-18 inclusive should exceed that for the US by at least 1.5 percentage points.
Since the implied difference in the proportion of the population who are unemployed is almost exactly equal* to the difference in the proportion of the population who are incarcerated, I interpret my side the bet as follows
Averaged over 2009-18, the sum of incarceration and unemployment rates in the US will exceed that in the EU-15
Caplan wants to leave incarceration out of the discussion and focus only in unemployment. Since we’re agreed on how to settle the bet, there’s no problem with differing on how to interpret the result.
I’ll do a longer post on the more substantive policy questions about the merits of labour market policies when I get some free time. But, for the moment I’d just like to make a “revealed preference” observation about the bargaining process.
Caplan initially proposed 1 percentage point, which I rejected with a counteroffer of 2.5. He rejected that and we ultimately settled on 1.5. Assuming our offers truthfully reveal our beliefs, that suggests Bryan’s point estimate for the average difference in unemployment rates is between 1.5 and 2.5 percentage points, while mine is between 1 and 1.5. Considering we have very different understandings of the economy, these estimates are pretty close.
and http://michaelnielsen.org/blog/?p=609). So, 1.5 percentage points seems like a pretty good estimate for the average over the next 10 years.