DDP Research Memo #2: The Climate Change Conundrum

Executive Summary

Over our first two surveys, two clear electoral issues are emerging: climate change (for Liberals and the left-leaning parties) and ethics (for the Conservatives). But if the Conservatives are having trouble finding an audience for ethics concerns, even among their core supporters, the Liberals appear to have an even more difficult conundrum: The very people who actually believe in their main issue don’t appear to have much interest in doing anything about it.

Key Findings: Issues and Policies

1. The environment remains a top-three political issue for most Canadians, along with the economy and healthcare, with ethics ranking further down the list for supporters of all three major parties.
2. Climate change is the most important component of the environmental discussion for politicians, journalists and members of the public alike.
3. Public support for reducing greenhouse gas emissions is high, but support for an increase in the carbon tax is low. In general, public opposition to a carbon tax rises as the proposed cost of the tax increases.

Key Findings: Media and Information

1. Media coverage about the environment is also more likely to involve climate change than other environmental issues such as single-use plastics or conservation. But while news organizations might be covering the environment, and journalists we monitored on Twitter frequently shared that coverage, there was far more disproportionate sharing of ethics-related stories, especially around the SNC-Lavalin story.
2. It is possible to make Canadians better informed about the facts that underpin policy issues, regardless of their political leanings. However, even if they are provided with correct information, it is unlikely to influence their beliefs about the policies needed to address those facts.

Survey Results

This is the second report from the Digital Democracy Project, a partnership between the Public Policy Forum and the Max Bell School of Public Policy at McGill University. The project uses data from both public opinion polling and online media analysis to examine the media habits of the broader Canadian public as well as the political and journalistic class, with an eye to understanding the various relationships between media use, partisanship, political knowledge, and concern over policy issues.

As with our first report, the environment was a top-ranked issue for Canadians, with climate change dominating the discussion. Seventeen percent of our survey respondents said the environment was the most important political issue to them, second to the economy (20%) and tied with healthcare (17%). Only 5% rated ethics as their most important issue. When asked to choose three important environmental policy issues, climate change came out on top (61%), followed by pollution at 49%. Journalists and candidates were far more likely to share media content involving climate change than other environmental issues, and a scan of online news sources shows that far more stories about the environment made reference to climate change compared to other environmental topics.

While a large majority of Canadians accept the scientific consensus on climate change, a quarter (27%) express skepticism that the Earth is getting warmer or that human activities are behind rising temperatures. This number is substantially higher among Conservative partisans (45%) compared with Liberal (22%) and NDP supporters (16%). The journalists in our sample were less than half as likely to tweet links to stories about the environment than stories about ethics. This probably reflects less a lack of interest in the environment than it does underscore an endemic feature of political journalism—namely, an overriding professional interest in scoops and scandals.

Experimental Results

1. While public support for emissions reduction is high, there is little consensus on what policies should be used to get us there. There is some support for renewable energy subsidies, but—somewhat surprisingly given the ostensible concern over climate change—widespread opposition to increasing carbon taxes, with only 36% of Canadians in favour.

We also found that Canadians are more likely to oppose a carbon tax as the tax rate increases. In the first of two experiments we conducted as part of our survey, we assigned respondents to one of three randomly assigned groups. Some were asked about a proposed carbon tax of 5 cents per litre of gasoline, while others were told it would be 10 or 15 cents. Forty-two percent of respondents are estimated to oppose a carbon tax at 5 cents per litre, but that rose to 51% at 15 cents per litre. A carbon tax gets majority opposition at around 13 cents per litre.

Conservative supporters were the most sensitive to proposed carbon tax increases: 60% oppose a 5- cent carbon tax, which increases to 75% at 15 cents per litre. However, supporters of left-leaning parties (i.e. Liberals, NDP, Greens, Bloc) were not responsive to changes in the proposed tax price. One can only speculate as to what motivates this clear partisan difference.

2. Our second survey experiment took a deeper look at a finding from our first report. There, we found that Canadians who read or watched more news media were more likely to give an incorrect response to policy questions, and strongly partisan respondents were particularly susceptible to giving more wrong answers as their media consumption increased. This suggests that highly motivated partisans may reject information that is inconvenient to them.

For this report, we wanted to see if providing survey respondents with correct information would affect their answers to questions about their policy knowledge or opinions. We informed half of our survey respondents that Canada was not on track to meet its Paris Accord targets. Members of this group were more likely to correctly answer a related question, regardless of their political leanings, than the other half who did not receive the fact.

However, we saw no evidence that having correct information about climate change would affect support for some types of mitigation policy. For example, support for the carbon tax was virtually identical for respondents who received the correction about the Paris Accord (35%) and those who did not (36%). What this suggests, at the very least, is that correct information about facts and related policies plays a limited role in determining one’s support or opposition to those policies.

Methodology

This Digital Democracy Project report draws from two primary data sources. First, our survey data team conducted an online panel survey of 1,554 Canadian citizens 18 years and older using Dynata. The sample was gathered from Aug. 17-23. Data was weighted within each region of Canada by gender and age to ensure it adequately represented the Canadian public. Survey respondents were asked questions related to basic demographics, as well as their partisan, ideological and issue preferences.

Second, we continued to collect Twitter data using the Twitter Search and Streaming APIs, leading to a cumulative sample of approximately 2.8 million tweets from June 3 to Aug. 23. Data used in this report was collected from accounts belonging to two categories of Twitter users: 1) major party candidates, for a total of approximately 950 candidates (including the two high-profile independent candidates, Jody Wilson-Raybould and Jane Philpott); 2) a list of approximately 450 political journalists.

Finally, we collected 38,735 English- and French-language news stories from major Canadian news organizations’ RSS feeds from June 1 to Aug. 22. To identify all the stories pertaining to the environment, the full collection of stories was checked for explicit reference to a party leader, legislation, the legislature or the election itself, leaving 11,257 politics-related stories. These stories were then checked against a broad range of environment-related terms (e.g. climate change, Paris accords, wildfires) and only those that contained a minimum of one environment-related term were retained, leading to a dataset of 2,172 articles (representing approximately 19% of politics-related coverage). All the environmental stories in this dataset were checked against environment-related key-terms aligned to the eight issue areas we discuss.