What Governments Do
We don’t have a lot of knowledge about the effects of election timing on local government policies. To date, few studies have attempted to assess the link between election dates and the broad range of policies enacted by local governments. This is definitely an area where more research is warranted.
There are, however, signs that the move to even year elections can benefit broad segments of the public. The most compelling research that we have focuses on the extent to which the views of elected officials match the views of the public. That research finds that after cities shift to on-cycle elections the views of local leaders tend to more closely match the views of local residents (Dynes et al 2021, Hartney and Hayes 2021, Schaffner et al 2020). That is true at the city government level, where the move to even year elections leads to the election of city government officials whose preferences are more in line with the preferences of the median resident (Schaffner et al 2020, Dynes et al 2021). Similar effects have been identified at the school district level where research shows that on-cycle elections reduce the ideological distance between school board members and district residents (Harney and Hayes 2021).
In terms of policy itself, there is strong evidence that on-cycle elections reduce the impact of special interest groups (Anzia 2014, Dynes et al 2021). By expanding the electorate, the belief is that on-cycle elections compel local government to be responsive to a broader set of residents.
That appears to be especially true for the interests of smaller groups like teacher’s union. Anzia () has shown that the move to on-cycle elections leads to significant drops in teachers’ salaries. Work by Dynes and his colleagues (2021) also indicates that local government salaries more closely match the ideological leaning of residents when elections are held on-cycle. One other study also finds better alignment between school board member’s support for charter schools and residents’ views on charter schools in on-cycle communities (Hartney and Hayes 2021). Combined this work suggests that on-cycle elections may reduce the influence of teachers unions and government employees.1
Other research has also shown that timing shapes spending patterns – typically in ways that lead to greater responsiveness to the average residents. The most compelling of these studies finds that the shift to on-cycle dates makes government policy more responsive to the median resident (Schaffner et al 2020, Dynes et al 2021). That greater responsiveness occurs both for the level of government spending and taxation as well as the extent to which governments try to address the needs of disadvantaged populations. However, these same studies find limited effects of timing on environmental or LGBTQ policies (Schaffner et al 2020, Dynes et al 2021).
There is also some evidence on the effect of election timing on disparities in representation by race and class. In both cases, the research indicates that more advantaged groups (Whites and the wealthy) tend to have more influence over policy than less advantaged groups (residents of color and working class Americans). But the research also shows that holding even year elections reduces those disparities (Schaffner et al 2020). By broadening the electorate, on-cycle contests reduce Black-White, Latino-White, and rich-poor gaps in local government policy responsiveness. In line with that study, research by Hajnal (2010) finds that higher turnout elections tend tend to enact policies that are likely to be viewed favorably by people of color and other less advantaged groups (Hajnal 2010). Specifically, higher turnout cities spend more of their budgets on welfare, public housing, and other social services than do cities with lower turnout elections.
​
Actions in different policy areas are impacted by election timing. More research would be beneficial to help us learn the extent to which election dates affect a range of different policies.
​
1 Subsequent work does, however raise doubts about the magnitude of these effects. In particular, Kogan and his colleagues (2018) show that the effect of on-cycle elections on the share of voters who are teachers is relatively small – largely because teachers represent such a small percentage of the overall population – typically only 2 or 3 percent. The same study found that on-cycle elections tend to have slightly higher passage rates for tax and bond referenda (Kogan et al 2018). The more liberal outcomes for on-cycle contests may be because they increase the share of young voters with school age children (Hajnal et al 2022).