Be Careful Counting Chickens: Three Reasons the Liberals Should be Cautious About their Recent Polling Success

When it comes to polls the federal Liberals have had plenty to be happy about. Throughout the summer they have maintained a healthy lead over the Conservatives, with some polls even putting them in the majority government range. This has been a significant reversal from their third place finish in the 2011 election. There is, however, a long way to go until October 2015 and a lot might happen to change Liberal fortunes. The Liberals should be cautious about these recent numbers for three reasons: the degree to which past polls have under-predicted incumbent vote share, the effect that increased scrutiny might have on their support, and the difficulty they will face competing as a centre party in the 2015 election.

The inaccuracy of polls has become a bit of a refrain in Canadian politics over the past few years. Polling predictions have been disappointing in a couple of recent provincial elections including the 2013 British Columbia election, the 2012 Alberta elections, and the 2012 Quebec election. Federally polls have not faired particularly well either. In 2011 pollsters accurately predicted the growth of the NDP and fall in support of the Liberal party, but under-predicted the Conservative vote share putting it around 35% instead of at the 39% that they actually received. That margin may have been the difference between a minority and majority government. In 2008 as well polls had the Conservatives below the 38% they received in that election. Given these inaccuracies it is tempting to dismiss the results of polls outright, but there may still be something to the numbers. The errors in both the provincial and federal polls have consistently been in the same direction, under-predicting the incumbent’s vote share. This has not always meant the more right-wing party has benefited. In the 2012 Alberta election the Progressive Conservatives were under-predicted vis a vis the more right-wing Wildrose Alliance and in 2012 in Quebec the federalist/centrist Liberal party was under-predicted vis a vis the separatist Parti Quebecois and more right of centre Coalition Avenir Quebec (the Liberals did lose the 2012 election but not by the large margins predicted). The reasons incumbents are doing better than polls expect are unclear. It may be that voters are more willing to tell pollsters that they are ready to vote against the government but don’t follow through when they get into the voting booth. It could also be that individuals are reluctant to tell pollsters they support an unpopular incumbent, but are willing to cast their ballots for them when voting in secret. Finally these inaccuracies may be a function of the turn-out models and sampling strategies that some pollsters use, though I have yet to see evidence that a particular sampling strategy or weighting method has consistently avoided the errors of recent years. Regardless of what is responsible for the polling failures, their errors give one reason to believe that the Liberal lead may not be as large as the polls suggest.

The second reason that the Liberals should be cautious about their polling numbers has to do with the way that media scrutiny increases during elections. This increased scrutiny is likely to benefit the Conservatives in two ways. First, the Liberals are going into their first election with Trudeau as leader while it will be Harper’s fifth election as leader of the Conservatives. In their book on the 2012 U.S. election, The Gamble, Johns Sides and Lynn Vavreck noted a trend in the Republican primary where new candidates would be “discovered” by the media as their poll numbers increased and then see their numbers fall as they were subject to increased scrutiny by the media*. The candidate that was immune from this pattern was Mitt Romney, who had contested the Republican primary in the previous Presidential election. The fact that Romney had run in a previous election meant that there was much less about him to discover in the 2012 primary, and so his support was less sensitive to changes in the favourability of his media coverage. American primaries are obviously quite a bit different from Canadian general elections, but there is a general concept that should apply to both cases. A large amount of criticism of Harper would have come up in previous elections. It is unlikely that we are going to find out that Harper is a bad debater or about a scandal from Harper’s past in the coming election. For better or for worse, Harper is a know quantity. The story is different for Trudeau. He has not been subject to the same high degree of scrutiny as Harper has in previous elections, and so there is an increased chance that scrutiny will uncover problems with Trudeau’s leadership or policy ideas. These problems do not have to be scandals, it may be that Trudeau is exposed as a poor debater or weak in particular policy areas. Because less is known about Trudeau, it is more likely that the poll numbers for his party will shift as the changes in the tone of media coverage expose things about him that were not well know before.

The other scrutiny advantage that Harper has comes from the fact that he is currently in government. Outside of an election campaign the media will apply much greater scrutiny to a governing party than to an opposition one. Scandals in government affect the way that policy is delivered and are more interesting than scandals within opposition parties who have limited control over what happens to the country. That changes during an election campaign as opposition parties begin to be treated more like potential governments, and the media becomes more interested in problems with them. This is especially the case if polls make it look highly likely that an opposition party might win the election. None of this is to say that the government is immune from scandal during an election, just that we are probably as likely to hear about a government scandal or problem at any time between elections, while we are more likely to hear bad things about opposition parties during elections. As a result, Conservative polling numbers should be a lot closer to their floor than Liberal numbers. The Liberals probably have further to fall than the Conservatives do.

The final reason the Liberals should be cautious lies in their position on the left-right spectrum and what it means for the strategies that the Conservatives and NDP are likely to pursue. Because Liberal voters are generally in the centre of the political spectrum, it easier for both the Conservatives and NDP to grow by stealing Liberal voters than each other’s voters. Even in ridings that are close NDP-Conservative races, it is likely easier for either party to convince a Liberal voter supporting a third place candidate to move a little over to the right (in the Conservative case) or to the left (in the case of the NDP) than it is to convince a NDP voter to switch all the way over to the Conservative party or vice versa. As a result, the Liberals should expect to be the focus of attacks from both the Conservatives and the NDP. They can also expect the Conservatives to take policy positions designed to appeal to centre-right Liberals, pulling them away from the Liberal party, and the NDP to do the same with respect to centre-left Liberals. Traditionally the Liberals have been able to avoid this predicament by emphasizing issues off of the left-right spectrum in elections. In the 1960s and 1970s the Liberals were the defenders of Quebec’s place within Canada, in 1988 they were the party that stood against free-trade (which was needed for the to stave off a quite serious NDP challenge prior to that election), and in the 1990s and early 2000s they were the defenders of national unity and a strong federal government. Given the focus of recent elections on left-right politics though, it is unclear what issues the Liberals might be able to leverage to move the focus of the 2015 election away from the left-right spectrum. Even if the parties had equal resources, and it is likely the Conservatives have more, the Liberal party will have to match the campaigns of both the Conservatives and the NDP, while those parties can concentrate primarily on the task of winning over Liberal voters. This puts the Liberals at a disadvantage when it comes to holding on to the gains that they have made since 2011 during an election.

None of this is definitive. Over the course of the next year the Conservatives could dig themselves into a bigger hole (especially if the senate scandal gets worse) and the Liberals could run a clean an effective campaign in 2015 that simultaneously holds off both the NDP and Conservatives. They are in a difficult position though, and there current lead in the polls does not diminish the challenges that they will face in 2015.

* Sides, John and Lynn Vavreck. (2013). The Gamble: Choice and Chance in the 2012 Presidential Election. Princeton: Princeton University Press.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s