The Kramer problem: Micro-macro resolution with a Danish pool

• Kramer (1983) argued that individual assessments of the national economy are biased.
• Therefore, surveys should not be used to study effects of the economy on the vote.
• Using a pool of Danish surveys 1987–2011 we show the fears to be unfounded.
• We find strong sociotropic effects on the vote.
• These micro effects mirror the effects of the macro economy on the vote.

In a seminal paper, Kramer (1983) posed his “problem” for the study of economic voting with election surveys: the items administered can measure neither individual nor national economic wellbeing accurately. Instead these items of economic perception are laden with erroneous judgment and partisan bias. Thus, the investigation of economics and elections should not be a survey research enterprise. Here we show, through varied analyses in an extensive, well-gathered Danish election pool, that these fears are unfounded. The presence of strong sociotropic voting effects from surveys can be established, and reconciled with the observed effects of national fluctuations in the macro economy. Indeed, the micro- and macro- processes mirror each other, so resolving the Kramer problem.