As you read the title, you may be thinking, “How could one even pose such a question? Of course they matter!” After all, why would they even be run if they didn’t matter? There are plenty of implications if the answer to the title was “no,” not the least of what it would mean for democratic ideals.
Allow me to explain. Of course many campaigns matter. Plenty of candidates that were in tight races or had substantial leads have collapsed because of serious missteps in their campaigns. It could have been a scandal, poor strategy, or something else that completely tanked a campaign that otherwise may have won. Some candidates have beaten the odds on their way to victory. Campaigns also have the ability to persuade some people.
But there are also some races where the writing is already on the wall — races where the losing candidate has basically no chance to win no matter what they do. You can see these at all levels of government. The shoo-in may just have too much of the electorate on their side through approval ratings. They may just be too well-known. They may have significantly more money. In Democratic vs. Republican elections, the shoo-in may just have so many voters in their party base that the opposing party just can’t match. It could be a whole bunch of other reasons or some combination therein.
So, when do you determine when a campaign matters? The answer to that is rather complicated, but we will explore it some here. But first, we go to the studies that first truly posed this question: “Do campaigns matter?”
The Beginnings of Election Research
Modern election research began in the early 1940s with a group of researchers from Columbia University. The University — whose theories and studies would come to be known as the “Columbia school” of thought — featured early elections researchers such as Paul Lazarsfeld and Bernard Berelson. Lazarsfeld, Berelson, and Hazel Gaudet published a book called The People’s Choice in 1944. In this book, the authors studied residents of Erie County, Ohio over the course of the 1940 presidential election, which pitted Democrat Franklin D. Roosevelt against Republican Wendell Willkie. Their goal was to find what effects the campaign had on their opinions. To their surprise, their findings suggested that media exposure to the campaigns had little to no effect on people’s opinions. Instead, people relied on their social group memberships to determine who to vote for, as social groups were predisposed toward voting with one side or another.¹
Those who ascribe to the phrase “demography is destiny” would certainly agree to these results. A follow-up book by Lazarsfeld, Berelson, and William McPhee in 1954, Voting, would come to much of the same conclusion. Studying Elmira, New York during the 1948 presidential election between Democrat Harry Truman and Republican Thomas Dewey, they wrote of both sides of the political aisle:
“Both exhibit stability and resistance to change for individuals but flexibility and adjustment over generations for the society as a whole. Both seem to be matters of sentiment and disposition rather than ‘reasoned preferences.’ . . . Both are characterized more by faith than by conviction and by wishful expectation rather than careful prediction of consequences.”²
According to the authors, voters were mostly operating on feeling rather than thinking a great deal about who is the best choice.
Of course, this wasn’t a consensus opinion. Studies over time would come to challenge, critique, and modify these theories about the electorate. But after the publication of 1960’s The American Voter — which identified party affiliation as the major determinant of vote choice — Holbrook writes that there were was relatively little scholarly research into political campaigns until a resurgence of interest in the 1980s and 1990s.¹
Where Do We Draw the Line on When Campaigns Matter?
Recounting all of the major findings over time would be outside the scope of this article, but there have been many different angles of study into how much and when campaigns matter. Studies have explored the effects of campaign rallies, certain types of media exposure, political socialization, and networks of similarly-minded people, just to name a few examples.
The Columbia studies were limited and exaggerated the situation about campaigns, but they did give us valuable information and set off noteworthy inquiries. Saying campaigns don’t matter as a blanket statement would be faulty. Yet campaigns can only do so much in actually persuading voters to pick a particular candidate, or even to get them to come out to vote at all. There are some voters who are simply set in their ways, and no amount of campaigning to change their minds is going to work. Some campaigns will be over before they even begin.
So the real question isn’t whether they matter in general. The real question is this: When and how do campaigns matter?
The latter part of this question is easier to address, although it is a lengthy answer that isn’t straightforward. Campaigns have all sort of strategies that are used and an innumerable amount of factors to consider, and the effectiveness and importance of these depend on the election in question. Being able to send the most effective messages to the groups that most need convincing, and reaching them in the correct manner at the proper time, is a necessity. There are some voters who are so adamantly opposed to a candidate that it makes no sense for campaigns to target them, because they’ll never change their mind. Likewise, some voters are so enthusiastically supportive of a candidate that it does not make sense to target them either — they’re going to show up for that candidate anyway, so there is no reason to use campaign resources toward them. It is those who are less firm in their support that need to be reached.
The former part of the question, however, is not at all simple. In fact, there is no real answer. There is no clear line for when and where campaigns matter and can win. There are some campaigns that just have no chance at winning, but there is no statistical measure that blatantly tells you this.
To illustrate this, let’s take the examples of Governors Charlie Baker of Massachusetts and Larry Hogan of Maryland. Both governors are Republicans, but registered Democratic voters heavily outnumber registered Republican voters in their states. The latest data available for Maryland (June 9, 2018) show Democrats as well over half the population, 59.58%, while Republicans are only 27.89%.³ In Massachusetts, August 2018 data shows Democrats at 33.22% and Republicans at 10.37% (NPAs actually make up the largest share of the state at 55.08%). Judging by these numbers, Republicans are already at a significant disadvantage, and many observers would write off their chances at winning a statewide office.
Yet not only did these two candidates win, but they are the two most popular governors in the country right now. Baker has a 69% approval rating compared to only a 17% disapproval rating. Hogan has a 68% approval rating compared to only a 17% disapproval rating.⁵ Such approval ratings are already hard enough to get when the Republican/Democrat divide is more even or even favors a governor’s party. Because of this disadvantageous partisan disparity, Baker and Hogan had to run more moderate campaigns. Certainly, their campaigns mattered.
So, there really is no straight answer to when campaigns matter. Some candidates have won despite huge odds against them. But in some cases, candidates have lost when plenty of factors were on their side. While it is hard to know whether it would have lasted, Gary Hart was widely perceived as the front-runner at one point for the 1988 Democratic nomination for president. His public profile took a serious hit, however, when the media uncovered evidence that Hart was having an extramarital affair, and he eventually dropped out of the race.
Even so, a lot of times elections are predictable. Incumbents win at very high rates, with percentages in the 80s and 90s depending on the type of office. Some candidates are too disciplined and/or have too much at their disposal to lose. Some candidates just aren’t going to be competitive enough. In some state legislative races, you can see one major party or the other not even try to put up a candidate because they don’t foresee a chance at winning. In those cases, primaries are often the “real” election because they will decide the winner if multiple candidates run for that party’s nomination.
Campaigns don’t always matter. But many do. And when they do matter, they matter a lot.
- Thomas M. Holbrook. “Do Campaigns Really Matter?” In The Electoral Challenge: Theory Meets Practice. Chapter 1. Stephen C. Craig and David B. Hill, eds.
- Bernard Berelson, Paul Lazarsfeld, and William N. McPhee. Voting. 1954.
- Maryland Department of Elections. “Eligible Active Voters by County.” https://elections.maryland.gov/press_room/2018_stats/GP18_Eligible_Active_Voters_by_County.pdf (accessed October 11, 2018).
- Massachusetts Secretary of State. “Elections: Massachusetts Registered Voter Enrollment.” https://www.sec.state.ma.us/ele/eleenr/enridx.htm (accessed October 11, 2018).
- Morning Consult. “America’s Most and Least Popular Governors — July 2018.” https://morningconsult.com/2018/07/25/americas-most-and-least-popular-governors-2/ (accessed October 11, 2018).