Term limits, time horizons and electoral accountability

Abstract

Term limits have been known to reduce electoral accountability by removing the possibility of reelection, thus affecting economic policy choices (i.e., the ‘lame duck’ effect). We show that the magnitude and statistical significance of this effect is influenced by the expected length of a future career. By using incumbent age as a proxy for expected career length, we find that the lame-duck effect is statistically observable only among those politicians with long careers ahead. Using data on US governors from 1950 to 2005, we find evidence that the influence of term limits is heterogeneous, primarily influencing young incumbents who hope to have long careers and thus have stronger incentives to remain accountable to voters. Indeed, we find little evidence of a lame-duck effect among older incumbents, suggesting that their already-shortened time horizons may offset the term limit effect.