What the 'Do Not Want to Reveal' response actually revealed
Democracy International (DI) has recently published its latest Bangladesh Opinion poll. A quick look at the results provides good news for the ruling party. The opinion about the direction of the country is a case in point. However, the most comforting news for the ruling Bangladesh Awami League (AL) is that support for its arch-rival seems to have completely withered away. In a predictable response, the opposition Bangladesh Nationalist Party (BNP) spokesperson rejected the poll results.
The survey conducted in late October (with a sample of 1,453 citizens) has given a clear thumbs up to the present regime ('Key Findings: Bangladesh Computer Assisted Telephone Survey October 2016', Democracy International). A staggering 70 percent of the respondents feel that the country is moving in the right direction, a dramatic turnaround from April 2013 when only 14 percent were so optimistic. Earlier surveys were conducted by DI in April 2013, July 2013, January 2014, January 2015, and July 2015. As for the direction of the country, the percentages of optimistic populations were 14, 37, 29, 53 and 58 percent respectively. Evidently, the positive trend began in January 2015 and is holding, according to the latest poll.
The number that has raised eyebrows of many people on the one hand, and provided cause for celebration among the ruling party supporters on the other, is the infinitesimal support for the main opposition party, the BNP. The survey result shows that the ruling AL has the support of 38 percent of respondents while the BNP has the backing of only five percent, the lowest ever share of popular support in any opinion poll. This should be of great comfort to the AL, for it has been questioning the bona fide of the BNP for the past years, put the BNP leadership under enormous pressure and has subjected the party activists to harassment. These, in conjunction with the strategic mistakes of the leadership, have put the BNP in disarray. As such, the ruling party, taking the survey at face value, may conclude that its tactics to decimate the opposition are succeeding.
The reason that the number has raised some eyebrows is that, if true, this represents a tectonic shift in Bangladeshi politics. For decades Bangladeshi politics has been characterised by a two-party system. The BNP had 35 percent support according to the DI survey in October 2014. The number was quite consistent with its earlier survey of January 2014. Immediately after the controversial election boycotted by the opposition parties including the BNP, the DI survey asked respondents, "If the election of January 5 was fully participatory, which political party would you vote for?" The support for BNP was 35.1 percent. Therefore, unless we accept this as one of the outlier surveys, which is not unusual in the realm of opinion polls, we have to conclude that in less than two years the BNP has lost almost all of its support.
A pertinent question then is, where did the BNP supporters go?
The support for the AL, as per the recent survey, is 38 percent, the exact share it had in October 2014. But in the January 2014 poll, 42.7 percent of respondents said they would have voted for the AL. These results show that between January and October of 2014, the AL lost about 4 percent support. The variation is quite reasonable. As for the BNP supporters, have they just withered away? Perhaps we can find the answer if we look elsewhere.
The poll shows that 35 percent of respondents didn't want to reveal who they would have voted for if the election was held today. The number is astounding since almost two years ago, in December 2014, the number was only five percent. Besides, 14 percent in 2016 said they 'Don't Know' compared to only one percent in 2014. Intuitively, we can say that the BNP supporters have moved to the 'Don't Want to Reveal' column. But of course we cannot say that with certainty. At best we can only suggest that they are the 'secret BNP voters' akin to the 'hidden Trump voters' who eluded pollsters in the recent U.S. presidential election. I must acknowledge that this is sheer speculation and must caution that the BNP shouldn't consider it as a 'consolation trophy.' Can these respondents be the supporters of the Bangladesh Jamaat-i-Islami (JI), the Jatiya Party (JP) or the AL? Given that the JI is facing an adverse situation it is a plausible explanation, but lest we forget, the JI has lost only one percentage point between these two surveys. The likelihood of them being the supporters of the JP or the AL is slim to none, because both are at the helm of power and neither made any substantial loss in their support base – the JP has lost three percentage points while the AL has maintained its base solidly.
Speculations aside, it is my understanding that the 'Don't Want to Reveal' category is quite revealing. It is not revealing their choice of a party but of the prevailing situation in the country. That more than one third of respondents do not want to reveal their preference between legitimate constitutional political parties in a country which claims it to be democratic and insists that dissent is tolerated, is a message in itself. One doesn't have to read between the lines here. It is loud and clear that a sense of fear is prevailing which precludes citizens from expressing their opinions. Similar results were found in a survey conducted by the Asia Foundation between October and November of 2015 ('Bangladesh's Democracy: According to its People', Asia Foundation, March 2016). It reported that "a third of respondents did not feel free to express their political opinions or were unsure." The DI survey informed us that 38 percent of respondents have said that most people in the area where they live do not feel free to express their political opinions. The response to a similar question in the 2015 Asia Foundation survey revealed that 30 percent of the respondents said people in their locality felt the same way. The Asia Foundation survey summarised the prevailing atmosphere saying, "Freedom of speech has (sic) declined significantly since 2006."
While the ruling party may take comfort in the 'diminishing support' for the opposition BNP, there is a message for it too: its support hasn't increased but instead declined between January 2015 and October 2016. A simple majority of the respondents who didn't reveal their preference can sway the result if a free and fair election is held.
(Postscript: The latest DI survey used Computer Assisted Telephone Survey System or CATSS. Randomly generated mobile phone numbers were used as a sampling frame. Discussion on the sampling method is a matter of different discussion).
The writer is Professor and Chair of the Department of Politics and Government at Illinois State University, USA. His recent publication is titled Bangladesh: A Political History Since Independence (London: I B Tauris).