Carly Urban, Montana State University
Sarah Niebler, Dickinson College

Why do individuals contribute to political campaigns? There are many reasons people may give—ranging from the warm glow they feel after contributing to a cause they care about to perceived access to political candidates. Presidential campaigns thrive on individual campaign contributions to keep their messages present with advertisements, radio slots, and get-out-the vote campaigns. With these goals in mind, candidates should aim to maximize the amount of money they can bring in. But how can they do this?

We show that by running advertisements in non-battleground states, presidential candidates could earn additional money, meaning that there are currently “dollars on the sidewalk.” Because of the winner-take-all nature of the Electoral College, U.S. presidential candidates tend to only compete, and therefore advertise, in “battleground” states.  However, in some areas of non-competitive states, media markets overlap with battleground states. For example, Boston is often uncontested, but receives presidential advertisements because its media market overlaps with part of New Hampshire. Our work examines the effect of those “accidental” televised campaign ads on individuals’ likelihood of contributing financially to either of the two major-party U.S. presidential candidates.

We ground our analysis in previous work finding that televised campaign advertisements can have an effect both on voter turnout and on persuasion. We use data on the location, timing, and content of political ads from the Wisconsin Advertising Project in combination with data on who is contributing to presidential candidates from the Federal Election Commission.  Examining these two sources together, and looking only at zip codes in non-battleground states, we are able to isolate the effects of televised campaign ads on contributions.  We match zip codes of non-battleground states where there is no political advertising to zip codes of non-battleground states that are exposed to advertising – advertising really aimed at the neighboring battleground state.

Another example of the accidental advertising phenomenon can be seen by looking at advertising in Illinois in the 2008 presidential campaign. The only three media markets contained solely within Illinois (Rockford; Peoria; and Champaign) did not see any advertising during the 2008 general election campaign.  However, we found that all other media markets in Illinois did see advertising as they also cover areas of neighboring battleground states.  For instance, the Chicago media market covers area of Indiana; the Davenport market covers areas in Iowa; and the St. Louis market covers areas of Missouri.  In our study’s design then, zip codes in Champaign or Peoria were matched with zip codes of similar demographics and socioeconomic characteristics in Quincy or St. Louis.


We ultimately find that in 2008, zip codes that were exposed to accidental advertising contributed, on average, between $6,100 and $7,200 more than zip codes with similar characteristics that were not exposed to advertising.  We further examined whether these findings were likely unique to 2008 and found that they were not.  We estimate that while it was the case that zip codes contributed more, on average, in 2008 than in 2004, that being exposed to advertisements increases aggregate contributions at the zip code level.

Having found that accidental advertising can increase campaign contributions, we aimed to determine next if advertising in uncontested states could ever be economically fruitful for presidential candidates.  After calculating the optimum level of ads in each media market that was solely contained within a non-battleground state, multiplying it by the expected gains to advertising, and subtracting an estimated cost for airing those ads, we found that candidates might expect to raise an additional $3 million by airing ads in these markets.  While this may seem like a small amount of money in a campaign season in which presidential candidates raise, and spend, upwards of $1 billion, we ultimately conclude that targeting advertising not just in battleground states, but also in those media markets that have a high concentration of high-income individuals may prove profitable for future presidential candidates.

About the authors: Carly Urban is an Assistant Professor of Economics at Montana State University and Sarah Niebler is an Assistant Professor in the Department of Political Science at Dickinson College. Their article, “Dollars on the Sidewalk: Should U.S. Presidential Candidates Advertise in Uncontested States?” appeared in the April 2014 issue of the American Journal of Political Science.