When Politics Get Personal

How marketing savvy and behavioural research have reshaped elections

At the height of the global recession, Mark Carney, then governor of the Bank of Canada, spoke in Yellowknife, Northwest Territories, and reassured Canadians about the national economy. Canada would avoid excess and recover faster than other countries, he told us, because of the Canadian economy’s fundamental strengths.

The same optimism is not being expressed about Canadian democracy.

Last year, in the midst of an investigation into robocalls, Marc Mayrand, chief electoral officer of Canada, spoke in Toronto and cautioned Canadians that our national democracy is under significant pressure. Although, by international comparison, Canada has a historic record of good governance, Mayrand urged Canadians not to be complacent. He cited, as reasons for concern, declining levels of citizen engagement and the potential erosion of public trust in democratic institutions.

Conventional measures of civic engagement underscore his point. For most of the 20th century, voter turnout in Canada fluctuated around 75 percent of eligible electors; since 1993, national participation rates have slid and now hover closer to 60 percent. This represents approximately 4.5 million fewer Canadians casting ballots.

Political parties fare worse. Canadian political parties have comparatively lower levels of membership, and their membership is aging and declining.

These trends are somewhat confounding. Parties remain the primary vehicles for representation and political power. Parties form governments and elected representatives make laws that affect employment, human rights, national security, internet freedom and environmental protection—all issues that citizens say they care about.

So what’s going on?

The two books under review help us to better understand how political competition has changed, how electoral strategies are changing, and what these changes possibly mean for democracy.

A Canadian journalist argues that the political machinery now treats citizens as consumers, and as a consequence people are checking out of democracy. An American journalist argues that the political machinery once again views citizens as individuals, and as a result people are turning out to vote.

Given that these two authors largely describe the same phenomena over the same period of time, it is difficult to reconcile their differing conclusions.

Susan Delacourt, a veteran observer of national politics, reflects upon Canadian democracy in her new book, Shopping for Votes: How Politicians Choose Us and We Choose Them. The goal of her book is to propose a new metaphor for national politics. Delacourt finds tired the customary political analogies drawn from military strategy, theatre and sports. These narratives, she argues, increasingly fall short. Instead, the emergent language of politics is about “brands, products, selling and buying”—in a word: shopping.

This book is not just about a shifting political lexicon. Delacourt examines the more fundamental relationship between commercial approaches to electoral competition and the potential consequences for Canadian democracy.

Almost the entirety of this book is an historical tour of political marketing in Canada from the 1950s through present day. Buttressing her account are concurrent developments in the advertising industry as well as an overview of the period’s political history and popular culture.

According to the history presented by Delacourt, political parties in Canada are relying more and more upon lessons from commercial tactics in marketing, advertising, branding and data analysis. In the mid 20th century, this meant selling politicians on television like soap, whereas today it means undertaking extensive market research to target individual voters with direct messages.

The personal narratives of well-known political advisors such as Dalton Camp, Martin Goldfarb, Allan Gregg and Patrick Muttart are used by Delacourt to illustrate transformations in Canadian politics. Although the skills brought to influence by these figures differ in accordance with the age of persuasion, the author’s portrait structure is replicated throughout: savvy contrarian outsider infiltrates stale partisan clutch to shift paradigm.

Delacourt’s portrait of Martin Goldfarb, long-time advisor to the Liberal Party of Canada, is one such example. The reader is informed that Goldfarb grew up “in a tiny apartment over a grocery store that his parents operated on Dundas Street West in Toronto.” He accumulated a fortune as a market research consultant for corporations such as the Ford Motor Company, but also advised politicians. His first meeting with Pierre Trudeau, then prime minister, was “a short, frosty conversation,” and yet, “between 1970 and 1980, Goldfarb’s firm raked in $1.3 million in government contracts.” The successful application of polling to politics was one of the innovations contributed by Goldfarb. When Jean Chrétien was prime minister, more than 20 years later, Goldfarb was still advising the Liberals.

One learns from Delacourt that political parties that fail to innovate or adapt as quickly as their competitors are unlikely to form government or remain in power.

She illustrates how the Conservative Party of Canada applies market lessons not just to electioneering but also to governing. Dimitri Soudas, former director of communications in the Prime Minister’s Office, illuminating this point, explains why government announcements do not include dollar values: most people have never seen and cannot relate to $10 billion. “Look at PMO press releases,” instructs Soudas, “they always talk about what we are doing, not how much we are paying to get it.” Delacourt further cites the government’s approaches to photo ops, communications planning and the naming of legislative bills as evidence of excessive attention to political marketing.

The influence of individual political actors is central to Delacourt’s analysis. When a political figure is introduced, she has a tendency to precede the person’s name with “a fellow named.” This is one of the stylistic devices that may not appeal to all readers. The passive-verb style is often employed, clichés are not uncommon and little qualifiers dilute arguments; the following sentence exemplifies all three: “Populism and pitchmen had sent Canadians flocking to politics in droves, it seems.” Additionally, readers less familiar with Canadian politics may struggle with the book’s historical chronology not being completely linear.

Toward the end of Delacourt’s historical summary, she introduces the reader to the use of data mining by political parties to construct voter profiles and win elections.

This concept—data mining—is the primary focus of The Victory Lab: The Secret Science of Winning Campaigns, by Washington-based national political correspondent Sasha Issenberg. Whereas the political machinery in Canada described by Delacourt is influenced by cross-pollinating advertising executives, Issenberg’s world of politics has been reshaped by students of science.

Like Delacourt, Issenberg uses the personal narratives of select actors to explain how electoral campaigns in the United States have become data driven. Persons profiled range from obscure figures such as Harold Malchow to well-known operatives such as Karl Rove. With the exception of players like Rove, most of the characters in this story are lo-fi outsiders with high algorithmic intellects.

The same time period is covered here as in the previous book. Issenberg begins mid century not because of a rise in consumerism, but rather due to the behavioural revolution in social science. This was an academic shift away from observational scholarship toward hypothesis testing and randomized field experiments.

The purpose of Issenberg’s book appears to be derived in part from his frustration with conventional journalistic accounts of political competition. Common storylines feature leadership profiles, personal relationships and exorbitant financial contributions. The electoral story told by Issenberg is about data collection, regression analysis and direct mail.

What the big data revolution has wrought for electoral politics is astounding.

Political parties have access to basic personal information about each registered elector. This is not new. The state has long supplied parties with elector names and addresses. But with increasing sophistication, campaign teams now layer on top of this information any data that may indicate a voter’s intentions. This additional information is collected through doorstep interviews, telephone surveys and push polls, as well as commercially available private sector market research.

When combined, these data provide signals about partisan leanings, issue importance and the likelihood that a person will turn out to vote on polling day. What behavioural research has taught, particularly lessons from psychology, is that people respond differently to different messages. A message about the closeness of an electoral contest will cause one person to volunteer for a campaign and another to retreat from perceived conflict. The more a campaign can test varied approaches to messaging, collect results and apply findings against individual profiles, the higher the likelihood of convincing an eligible elector to vote and of persuading that voter to cast a ballot for your candidate.

Sometimes the practical application of elector data is quite sophisticated. A story Issenberg tells early in the book is how, in 2008, during the recount in Minnesota for election to the United States Senate, Al Franken’s lawyers used analysis of the campaign’s elector database to build their legal case. Rather than contest each challenged absentee ballot one by one, legal counsel successfully built arguments based on their team’s data analysis. The Franken campaign “used a complex mix of personal and demographic information, along with polling, to give each voter a score of 1 to 100.” This maximized the efficiency of legal arguments by pointing lawyers toward ballots in unopened envelopes from polling divisions most likely to yield votes for Franken.

More often, though, databases are used not to create more sophisticated campaign methods but rather to enhance the sophistication of campaign tool selection. For example, campaigns are using data to select advertisements for particular bus routes in certain cities in order to maximize the effectiveness of their candidate’s message.

In this regard, Issenberg explains the simple brilliance behind the campaign tactics of then presidential candidate Barack Obama: “the most technologically advanced campaign in history had so thoroughly mastered the politics of individual data and testing that it found new value in electioneering tactics many had abandoned as hopelessly last-century.”

Issenberg sometimes appears in awe of the clever application of science to politics. His perspective is moulded by his coverage of the 2008 presidential campaign and 2010 congressional elections. He describes the Obama campaign as “the perfect political corporation: a well-funded, data-driven, empirically rigorous institution.”

It is true that the first Obama campaign for president brought historically disengaged electors, those thought to be on the margins of national electoral politics—African Americans, Hispanics, youth—into polling stations at record levels. However, the record voter turnout in the 2008 presidential election—the highest since 1968—was not improved upon in 2012. In fact, the recent election yielded participation rates lower than those in 2004. Micro-targeting is most effective at the margins. Data analysis makes the biggest difference in close races where small increases in turnout have a disproportionate effect on the outcome. Data mining alone cannot overcome depressed voter turnout when there are barriers to accessing polling opportunities.

Not mentioned by Issenberg is what all American electors find at the polls: long lines, complex ballots and insufficient levels of accessibility. Earlier this year, Obama issued an executive order to strike a commission examining ways to improve election administration in the United States.

Voters in Canada do not find chaos at the polls. Administration of national, provincial and territorial elections is well ordered. The confidence of Canadians in the integrity of the electoral system is high.1 After the electoral annus horribilis of 2011, which resulted in irregularities at the polls in Guelph being examined by the Supreme Court of Canada and deceptive telecommunications being investigated by the commissioner of Canada Elections, the fact that Canadians still trust the system is instructive.

Canada has an electoral system strong enough to absorb exogenous shocks. This does not mean that problems in the system should not be addressed or that periodic democratic recessions are not possible. Deceptive robocalls need to be dealt with and prevented from recurring. Sound policy responses are needed in order to ensure that the fundamentals of Canadian democracy remain resilient.

Susan Delacourt proposes concrete policy options worthy of further consideration. She recommends that adverts produced by political parties adhere to the Canadian Code of Advertising Standards; this would not end the practice of negative advertising, but would uphold a standard of fairness. Additionally, she joins a growing set of observers who advocate that political party databases should be subject to privacy legislation.

However, I fail to accept the broader conclusions drawn by Delacourt. Referring to deceptive robocalls, she writes: “so loose was the electorate’s attachment to obligations of citizenship … that an unexpected inconvenience would make people opt out of the system.” She argues that such a consequence results from a political culture that prioritizes choice over duty. It is, in her words, “the true, democratic price of seeing civic life in the same terms as consumer culture.”

The facts illustrate otherwise. During the period 1997 to 2010, fewer than 3,000 total complaints were filed with the Commissioner of Canada Elections, an average of approximately 150 per year. But in a single year, 2011, more than 1,400 Canadians contacted Elections Canada to file a complaint about deceptive robocalls. Canadians did not check out. Nor did they respond with violence, as is so often the case elsewhere. They did exactly what should be done. They contacted the independent, impartial national electoral regulator to initiate due process.

Canadian citizens were neither passive nor gullible.

These duplicitous attempts at suppression failed “to divert people from their democratic commitments.” Examining these events from the bench, Justice Richard Mosley concluded that “in no case did the calls ‘affect the result’ for the current applicants in the sense that they were prevented from actually voting.” The one person who did make such a claim in court withdrew her case.

Political parties are indeed shopping for votes. But arguing that Canadians see civic life in the same terms as consumer culture conflates concepts. Just because political parties have borrowed some lessons, strategies and intelligence from the commercial sector does not mean that Canadians—in the House of Commons or at Tim Hortons—equate citizenship with consumption. As was demonstrated following the 2011 general election, Canadians are not too consumed to voice their displeasure when faced with anti-democratic practices.

The shopping analogy does mean that political parties, as with all institutions, learn what works best in order to accomplish a defined objective: winning elections.

Finding certain electoral practices to be distasteful is a necessary, but not sufficient condition for political culture to change. If distaste leads to fewer financial contributions or fewer votes, parties will be quick to adjust.

There is reason to be optimistic. As Delacourt and Issenberg both demonstrate, political parties are always seeking the next big innovation in political marketing. Issenberg closes his book with recent evidence that straightforward, earnest appeals to civic duty can be the most effective messages to raise turnout.

As we wait for political marketers to adopt earnestness, our electoral institutions will be in place to preserve the fundamental strengths of Canadian democracy.


  1. Scores on trust are reported in “Survey of Electors on Communications with Electors,” prepared by Phoenix Strategic Perspectives Inc. for Elections Canada (March 2013).