The Quebec election provided a good opportunity to test the importance of leadership attributes, the strength of party brands and the degree to which voters felt affiliated with the parties, irrespective of their voting intentions.
These fundamentals " party affiliation, leadership and strength of the party brands " are important attitudinal indicators in any election.
We found (question 1) that nearly as many Quebecers (23.9 percent) did not identify with any of the major par- ties as associated themselves with the Liberals (25.8 per- cent) or Parti Québécois (24.3 percent). While the Action démocratique du Québec trailed at 15 percent, the high number of voters who were unsure or would state no affil- iation turned out to be a leading indicator of how ”œunde- cideds” and ”œdiscreets” would vote on election day. A majority of these unaffiliated voters clearly broke to Mario Dumont and the ADQ. Dumont also enjoyed a comparative advantage on election day " alone among the leaders, his name appeared on the ballot, as ADQ-Équipe Mario Dumont.
SES Research was in the field for Policy Options on March 14 and 15, the two nights following the leaders’ debate. We conducted 500 telephone interviews over the two nights, with a margin of error of plus or minus 4.4 per- cent, 19 times out of 20.
Since our findings would not be published until after the election on March 26, we saw no point in sampling vot- ing intention. Nor did we test the dissatisfaction rate with the Charest government, which all the polls put at 50 per- cent at the outset of the campaign, and as high as 60 per- cent by the end of it " very close to a tipping point, as the election results indicated.
We tested five leadership attrib- utes " vision, trust, competence, character and shared views. Jean Charest won all five of them handily against André Boisclair. But Mario Dumont was very competitive, finishing second on all five leadership attributes. And on two of them " trust and shared views " Dumont trailed Charest within the margin of error. Boisclair finished a distant third on all five questions, showing a reverse coattail effect for the PQ.
Charest scored most strongly on competence and character " two key components of who would make the best premier. On question 2, ”œWhich leader is the most competent,” Charest outscored both opponents by a margin of about 2-1, at 38 percent, versus 20.8 percent for Dumont and 17.9 percent for Boisclair.
On question 3, ”œWhich leader has the personal character to be the premier of Quebec,” Charest was again the clear winner at 35.6 percent, compared with Dumont at 24.7 percent and Boisclair at 18.2 percent. Again, Charest outscored Boisclair by a 2-1 margin, but Dumont narrowed the gap on character, and would narrow it even more on the other leadership questions.
On question 4, ”œWhich leader has the best vision,” Charest came out on top at 30.5 percent, while Dumont scored 25.5 percent and Boisclair trailed at 21.4 percent.
Asked which leader they trusted the most (question 5), Charest came in first at 29.0 percent, Dumont was a close sec- ond at 25.1 percent, and Boisclair was again third at 20.6 percent.
Asked which leader had views most like their own (question 6),
Charest was first at 27.1 percent, Dumont pulled to a virtual tie at 25.8 percent, and Boisclair was again third at 21 percent.
When we asked which of these five leadership factors was most likely to influence their vote, compe- tence was by far the most important at 29.8 percent, followed by vision at 17.3 percent, views most like the voter’s own at 14.4 percent, trust and under- standing Quebec both at 13.2 percent and personal character at 7.3. percent. Competence was by far the most important leadership attribute likely to influence Quebecers’ vote, and Charest enjoyed a huge 2-1 competence advan- tage over both his rivals. It isn’t for pollsters to determine why Charest and the Liberals were unable, in the end, to capitalize on their clear com- parative advantage on leadership.
When we examined the strengths of the parties, the Liberals enjoyed an even greater advantage on voter prefer- ence for their brand.
Asked which party had ”œthe strongest team of candidates” (ques- tion 7), fully 48.9 percent said the Liberals, with 22.9 percent for the PQ and only 7.4 percent for the ADQ. These were slam-dunk numbers for the Liberals.
However, they also suggested that even Dumont’s voters knew he had by far the weakest team, but they were going to vote for him anyway.
Again, when we asked which party had the strongest policy platform (question 8), the Liberals easily came in first at 36.5 percent, compared with 23.4 percent for the PQ and 15 percent for the ADQ
When we asked ”œWhich party is best at ensuring Quebec receives its fair share within Canada” (question 9), the Liberals were again a strong first at 41.5 percent, the PQ a distant second at 27.6 percent and the ADQ a bad third at 12.7 percent.
This question of getting Quebec’s fair share is not dissimilar to the litmus test question of which party or leader is the best defender of Quebec’s interests in Ottawa. While the PQ might have been expected to do better, the Liberal strength probably reflects the current context of negotiating enhanced fund- ing for Quebec in the federal budget, which would address the fiscal imbal- ance four days after we were in the field. When we asked which of these party factors was most likely to influ- ence their vote, the party platform was seen as the most important by 39 per- cent of Quebecers, obtaining Quebec’s fair share within Canada was a close second at 36.6 percent, while the best team was a distant third at 14.6 percent.
Again, as on leadership, the Liberals enjoyed an important advantage on brand preference, but were unable to use either to their advantage on election day. Indeed, a post-election poll of 600 Quebecers by Léger Marketing, published by Le Journal de Montréal on April 1, found that 48 per- cent of Quebecers thought a minority government led by Jean Charest would be ”œefficace,” pretty close to the com- petence question.
Clearly, the Liberal campaign failed to communicate the strengths of the leader and the party. In running on the record, the Liberals risked allowing their opponents to drive dissatisfaction to tipping-point levels. And in promis- ing a $700-million tax cut with new equalization money in the federal budget, Charest risked reminding vot- ers of his broken 2003 campaign prom- ise of $1 billion a year in tax cuts.
There was another important factor in the outcome. In a word: Mario. The fact that his party ran a bad third on all the brand equity ques- tions was not beside the point, but helped make the point that he was the brand " indeed, he had his name on the ballot.
Inasmuch as Dumont’s supporters were voting for him rather than his party, his strong second-place showing in the leadership numbers provided a road map to his strong second-place show- ing on election day.
Dumont is a populist in the old- fashioned sense of the word, in that support for his leadership runs way ahead of support for his party, and even for his ideas.
We also found that Charest shouldn’t have expected to get much of a bounce from the federal budget, no matter how much money he got out of it. Only 20.9 percent of our respondents said they would take a more favourable view of Charest as a result of gains for Quebec on the fiscal imbalance, while 38 percent said their view of him would be the same, and 37.8 percent said they would have a less favourable view. Clearly, Quebecers who weren’t voting for him anyway weren’t about to be impressed.
Stephen Harper fared somewhat better with Quebecers when we asked about the $350-million February envi- ronmental announcement to support Quebec’s Kyoto compliance plan, as well as the expected good news on equalization in the federal budget. Here, 27 percent said they would have a more favourable view, while 33.5 percent said their view would remain the same, and 36.6 percent said they would view the Prime Minister less favourably. Not surprisingly, Harper scored more favourably with Liberal voters (36.7 percent) than with PQ and ADQ voters (20.5 percent and 27.8 percent).
Coming back to the opening ques- tions in our poll on association with par- ties, we also tested this at the federal level in Quebec (question 10), and found that 28.3 percent self-identified with the Bloc Québécois, 18.7 percent with the Liberal Party of Canada and 15.3 percent with the Conservative Party. As in the question on identity with provincial par- ties, there was a large pool of voters, 24.1 percent, who identified with none.
In essence, one-fourth of the Quebec electorate is in play and could hold the key to the outcome of the next federal election in the province.