Voters literally know less than zero about economic policy – we would have better policies if they just voted randomly.  But people who believe in “retrospective voting models” often retort that voters’ policy incompetence doesn’t matter.  They don’t have to understand economics; they can simply reward politicians who deliver the goods, and punish those who don’t.

Sounds good.  But does it work?  Tim Harford reports on some interesting research showing that voters actually reward politicians more for prosperity that they didn’t cause.  In other words, voters know less than zero about causal responsibility:

The question is, can the voters tell the difference between an
incompetent government and an unlucky one? Andrew Leigh, an economist
at Australian National University, thinks not. In a recent article in
the Oxford Bulletin of Economics and Statistics,
he looks at 268 elections held across the world between 1978 and 1999.
He estimates how much of a country’s economic performance is due to
booms in the world economy and how much is due to competent government
– and whether the voters can tell the difference.

Both matter,
but as far as the voters are concerned, it is better to be a lucky
government than a skillful one. For instance, a one-percentage point
increase in world economic growth above the norm is associated with a
hefty rise in the chance that incumbents will be re-elected – from the
typical chance of 57 per cent to a more than decent 64 per cent. A
stellar domestic performance, outpacing world growth by one percentage
point, contributes less than half as much to the chances of being
re-elected, raising them from 57 to 60 per cent.

Admittedly, things could be worse – imagine a world where voters actually punish skillful incumbents.  Compared to policy-based voting, retrospective voting is the lesser evil.  In fact, I think it’s an important reason why policy isn’t much worse.  Nevertheless, retrospective voting doesn’t work nearly as well as its fans would have us believe.

HT: Eric Crampton