In addition to the two working papers from my dissertation (see Dissertation section) I am also currently working on the following papers:
Why are some local governments so much better at delivering basic public services to citizens than others? I argue that the answer lies in the extent of voter attachments to political parties. Where voters are strongly attached to political parties, elections fail to discipline poor performance of public officials. Using a unique dataset from a UNICEF project on public service delivery at the local (district) level in Ghana, I show that the supply of, and demand for public services are significantly lower in districts where voters consistently cast their ballots for the same party in national elections compared to those where significant numbers of voters are less committed and thus switch their votes between different parties over time
What explains the timing of transition elections in Africa? I argue that incumbent dictators timed the first multiparty elections to coincide with the presidential elections of the United States. Using data on all transition elections in Africa in the 1990s, I show that in countries where incumbent governments planned to manipulate the transition process, the first multiparty elections were significantly more likely to occur around the time of the US presidential elections.
We study electoral fraud in a relatively new and highly competitive democracy. A field experiment implemented during Ghana’s 2012 presidential elections produces measures of electoral fraud and estimates of direct and spillover effects of non-partisan domestic election observers. We find that fraud is higher in electorally competitive constituencies and it is committed at similar rates by the incumbent and the opposition parties but they specialize in different types of electoral malfeasance. We also find that observers reduce fraud in the polling stations where they are present but political parties respond strategically and relocate fraud to unobserved polling places, especially in their stronghold districts.
This is another paper from the Ghana election project. We study the effects of domestic election observers on fraud and election-related violence. We theorize that the response of party agents to observers will be conditioned by the level of local electoral competition, which serves as a proxy for the control parties have over local social and political networks. We expect parties are able to respond to observers with fraud in non-competitive constituencies, but only with violence in competitive constituencies. Data from a field experiment conducted during Ghana’s 2012 presidential election support our theory. We use data collected from individual polling stations (N=2,310) to construct measures of fraud and violence. We find that observers significantly reduce fraud (a 6 percent reduction) and violence (a 60 percent reduction). The results also show that the strategies that political parties adopt in response to observers varies depending on the level of local electoral competition.