Election observers and electoral fraud

Election observation - the deployment of trained personnel to monitor the compliance of political parties with electoral laws - has become a prominent feature of polls in countries across the world. Observers are usually deployed by multi-national organizations (international observers) or by local civil society groups (domestic observers). International election observers were present in 86 percent of the national elections organized in 95 newly democratic, or competitive authoritarian regimes, between 1989 and 2002.1 Today, domestic election observation groups are active in at least 60 countries.2 The main purpose of election observers is to detect and deter electoral fraud and violence. Do observers achieve this goal? At the national level, establishing the causal effect of observers on fraud is difficult because governments usually self-select into the treatment (i.e. allow themselves to be monitored). In this case, an association between observers and low levels of fraud does not prove observers reduce fraud, because the countries that choose to have observers may have cleaner elections to begin with compared to countries that are not open to observers. Similarly, within countries, if civil society groups deploy observers to historically troublesome regions, a comparison of outcomes at monitored and unmonitored polling stations may underestimate observers’ impacts