Over the last few weeks, I’ve raised the issue of the importance of political polling as we approach the October 23 elections. A new poll released this week raises more questions about the precarious political situation in Tunisia and how voters will respond in this Fall’s ballot. It also avoids key questions that would help us understand how the elections might turn and how political parties can reach out to voters.
Voter registration campaign from the independent elections committee
In an era of misinformation and lots of speculation, both in Tunisia and outside, the newIRI poll provides useful data on how Tunisians view their future (including their hopes and fears), the role of secularism/religion, and how Tunisians get their news. Particularly welcome is there publication of their methodology, which other polling firms here have declined to provide).
Overall, the poll showed widespread support for democracy, but their is an increasingly pessimistic attitude toward the future of the country and the political elite. A few points stand out:
- Almost half (46%) of all Tunisians now believe their country is going in the wrong direction
- Internal security and unemployment remain the critical issues for Tunisians
- 43% believe that political parties are only interested in power
- No individual or party received more than 7% support when Tunisians were asked their voting intentions
- Only one party was identified by Tunisians when asked about their voting intentions (other responses identified party leaders)
The last two figures are key. Political parties are the key to the upcoming elections and probably the single most understood and undervalued element in the transition period in Tunisia. When voters go to the polls in October, they will be voting for their preferred parties, not individual candidates, within their voting district, a system called proportional list voting.
This proportional list system favors a multitude of parties gaining at least some seats. While it does not favor small parties, it doesn’t exclude them either. Parties with strong support will receive a proportional share, while those receiving fewer votes must meet a minimum threshold in order to gain a seat within each constituency.
With over 100 parties competing in the coming elections, even the most well-informed voter could be confused by the plethora of parties. And as I’ve written about previously this week, most parties identify themselves as centrist, meaning that their is very little differentiation among party platforms.
Consolidation is needed, but polls aren’t helping
Unless there is some serious consolidation among like minded parties, there is a great risk that the elections will be jeopardized by spreading votes too thinly across the many political parties, making the drafting of a new constitution and the day-to-day governance of the country nearly impossible.
Imagine a district with 10 seats where 10,000 votes are required in order to gain a seat. Then imagine that the current front runners, Ennahdha (Islamist) and PDP (center-left), each wins 2 seats. The remaining 6 seats will be given to the next closest parties receiving votes. If we use the current polling data, these seats could essentially be randomly assigned, as no other parties receive statistically significant support from Tunisian voters and there is no way to judge the political slant of the electorate (whether they are leaning right or left).
Under these circumstances, a party leader of a small party has an incentive to consolidate his base by joining forces with another smallish party. For those parties that are ideological, alliances should be fairly straightforward, unless there are big egos involved. Far left wing parties should work with other far left wing parties, etc.
However, most Tunisian parties are centrist. Under normal circumstances, these centrist parties would make some calculations for alliances based on support in various regions and by further differentiating based on ideology (i.e., center left parties join with center left parties, etc).
Unfortunately, once again, the polls fail us. We cannot find data that would point a party leader toward making a calculated choice. None of the polls I’ve examined point to regional variation across the country, nor do they look at how voters would like their elected leaders to solve their problems (ideology).
We simply do not know what side Tunisians fall on, left or right. This information gap makes alliance making extremely difficult and very risky for a political party.
As of now, most of these small parties are testing the waters with voters. They are reaching out to voters across the country and they are no doubt looking at enthusiasm for their positions (and their opponents) at rallies, on the newsstands, and online.
With three months to go before the election, the time to start making these alliances is now. Ramadan, which begins at the end of July, will make campaigning problematic until early September, at which point there will only be seven weeks to go before the election.
Perhaps, as this week’s poll points out, Tunisians are right, political parties are only looking out for their own interests. Or perhaps, they need more information from voters.
Also interesting about the IRI poll:
Their are several ambiguities in the poll. For example, while 54% of respondents approved of a secular government, 59% would like to see political parties that are moderately or strongly Islamist in the Constituent Assembly. Meanwhile, the percentage of Tunisians whose political opinions are influenced by religious leaders fell from 46% in March (the time of the previous survey), to 34% in May.
None of the public opinion polls conducted so far have included expatriate Tunisian voters, who will for the first time have the chance to vote from abroad.
Postscript: This is the second time I’ve looked closely at IRI data. Lest anyone think I’m only reading U.S. sponsored polls, I am not. However, with the exception Emhrod, no other polling firms have released their polling methodologies, that I know of. Polls by MediaScan, Sigma, and 3C have failed to publish methodologies and complete lists of questions. It’s possible the information I believe is lacking is in fact out there somewhere. If it is, please send it my way and I will happily revise.