Monday, December 30, 2013

Is there a small-state effect?

In countries where some parliament chamber allocates the same number of seats to each member state regardless of its population, small states are deemed to enjoy a disproportionately strong influence. One paper that analyzes whether this small-state effect is empirically significant is by Gary Hoover and Paul Pecorino which shows that US states with higher per capita representation also get more federal funding. Does this mean that the open question is now closed? Of course not, as the scientific process would tell us to revisit this to test whether it holds more generally, whether the effect disappears with time, or whether it is robust to different specifications.

Stratford Douglas and Robert Reed (link corrected) address the latter question. They run a robustness exercise that is unfortunately too rare in Economics. They confirm the results of Hoover and Pecorino, but find that when you switch from ordinary least squares to cluster robust standard errors and include population growth that small-state effect vanishes. So we are not done with this question.

We should have more replication studies in Economies. It saddens me that Douglas and Reed felt the need to add the following footnote on the front page: "we wish to express our special appreciation to Gary Hoover and Paul Pecorino for their willingness to allow their study to be subject to critical analysis. Openness and integrity such as theirs is the basis by which science advances." This should be obvious.

No comments: