Elections 2013: survey indicates close contest

4

elections_ballotboxes_670

Pakistani voters appear divided on many questions of the day – including who to vote for in the upcoming elections and what issues are most critical for the country at present – according to the Political Barometer, an opinion survey conducted by the Herald in partnership with the Sustainable Development Policy Institue (SDPI), an Islamabad-based think-tank.

Of those respondents who say they have registered for the upcoming elections, 29 per cent expressed an intention to vote for the Pakistan Peoples Party (PPP). 24.7 per cent pledged support for the Pakistan Muslim League–Nawaz (PMLN) while 20.3 per cent indicated a preference for the Pakistan Tehreek-e-Insaf (PTI).

graph-1-finalMind the generation gap

The survey’s findings indicate that the PTI’s support is derived from all age groups – 22.9 per cent of those between 18 to 35 years, 18.6 per cent of those between 36 to 50 years, 18.4 per cent of those between 51 to 70 years and 7.7 of those above 70 years support the PTI, dispelling the notion that its vote bank is rooted in the younger generation.

The highest proportion of those aged between 36 to 50 years (32.5 per cent) indicate a preference for the PPP. Similarly, 46.2 per cent of those aged over 70 expressed a preference for the PMLN.

Compared with respondents’ voting histories, the PMLN’s vote bank appears to have remained stagnant while the PPP’s seems to have declined significantly.

It appears that the PTI has a stronger urban base, while a higher proportion of rural respondents indicated that they would vote for either the PPP or the PMLN in the upcoming elections.

graph-2-finalThe ethnic vote

Predictably, the highest level of support for the ruling party was pledged by Sindhis, 55 per cent of whom said that they would vote for the PPP in the impending elections.

This was followed by Seraiki-speakers at 46 per cent.

Forty-four per cent of Hindko-speakers said that they intended to vote for the PMLN, closely followed by Punjabis at 43 per cent.

The same proportion of Hindko-speakers – 44 per cent – also expressed an intention to vote for the PTI, indicating a close contest between the two parties (PMLN and PTI) within that particular demographic.

It is worth noting that while 34 per cent of Pakhtuns stated that they would vote for PTI, only 11 per cent expressed the same vis a vis the Awami National Party (ANP).

47 per cent of Baloch said that they would vote for the Balochistan National Party–Mengal.

Money matters

On average, approximately a third of those earning up to 30,000 rupees each month indicated a preference for the PPP whereas, among those earning more than 30,000 rupees, support for the party dropped to 10.8 percent.

This is in keeping with the party’s traditional pro-poor image.

No such trend could be determined for the PMLN, whose level of support remained similar across all income levels.

Those earning in excess of 250,000 rupees each month (the highest identified income bracket in the survey) expressed the maximum intention to vote for either the Muttahida Quami Movement (MQM) or the PTI, at 33 per cent each.

While this figure may appear anomalistic in the MQM’s case – support for the party within the second highest income bracket (those earning between 100,000 and 250,000 rupees each month) was only four per cent – it was possible to identify a rough direct trend between level of income and support for the PTI.

In general, it appeared that support for smaller parties declined with increasing levels of income.

graph-3-finalIt’s the issues, stupid

From a given list, respondents identified — in the following order – poverty, corruption, power crises, illiteracy and extremism as the top five issues crucial to the country today.

No issue received more than 17 per cent of the vote, a possible indication of a divided electorate.

A marginally higher percentage of respondents belonging to lower income brackets identified poverty as an issue of concern, while a higher proportion of those at higher levels of income identified – albeit again by a small margin – corruption as a ‘crucial issue’.

A similar proportion of urban and rural respondents (12 per cent each), considered corruption to be an issue that has not been effectively addressed by the current government — an assessment perhaps linked to the prevailing perception (55.6 per cent, according to survey responses) of the ruling party as being the most corrupt.

While 59 per cent of respondents rated the performance of the government as either ‘poor’ or ‘very poor’, the ruling PPP nonetheless emerged as the party that the highest number of respondents (27.1 per cent) said would be most effective in addressing the identified issues.

Urban respondents appeared more dissatisfied with the current government’s performance, with 63 per cent rating it as either ‘poor’ or ‘very poor’. Dissatisfaction was also slightly more pronounced with increasing education-levels of respondents.

10 per cent of those earning less than 3,600 rupees a month deemed the current government’s performance to have been ‘excellent’ while 23 per cent, conversely, rated it as ‘very poor’.

In contrast, 54 per cent of those earning in excess of 250,000 rupees a month considered the government’s performance to have been very poor.

Indeed, based on the survey’s findings, it is possible to state that those at higher income levels appear relatively more disgruntled by the current government – a conclusion somewhat surprising given that among the top five ‘crucial’ issues identified by respondents, at least two – poverty and illiteracy – are of greater concern to those earning lower levels of income.

For respondents with higher levels of education, extremism, political instability and interprovincial problems appeared to be issues of greater concern, whereas those with little or no education deemed inflation, gender discrimination and food shortages as bigger problems.

Getting out the vote

Approximately 21 per cent of respondents admitted to never previously having voted in an election.

There appeared to be a negative correlation between inclination to vote and levels of income – 38 per cent of those earning above 250,000 rupees a month stated that they had never voted before, as compared to 13 per cent of those earning less than 3,600 rupees.

Despite this, however, those within the highest income bracket were most likely (at 38 per cent) to have been a member of a political party.

Similarly, while a greater number of respondents from urban areas say that they have been members of a political party (17 per cent) and claim to have been active participants of an election campaign (23 per cent), their political involvement did not necessarily appear to translate into electoral participation.

29 per cent of urban dwellers profess to have never voted in an election while 85 per cent of rural respondents have voted in at least one election with 46 per cent having voted in three or more elections.

Moreover, while it would appear that higher levels of education may result in greater political participation, 87 per cent of those with no education claimed to have voted in three or more elections, while approximately 38 per cent of those with a Bachelor’s degree or higher professed to have voted in the same number of elections.

Ninety four per cent of respondents, however, have said that they have registered to vote in the upcoming elections, a figure that bodes well for a country with plummeting rates of voter turnout.

Based on survey findings, it would seem that the electoral playing field will be closely contested this time round between the PPP and the PMLN, with the PTI not far behind. In light of this, who will be at the helm of the incoming government?

Five scenarios

According to Dr Abid Suleri, executive director at SDPI, things may unfold in in a number of ways come elections, giving shape to five distinct scenarios.

Scenario 1
The PPP forms an electoral alliance with its current allies: the Awami National Party (ANP), MQM and the Pakistan Muslim League–Quaid-e-Azam (PMLQ); while a grand anti-PPP alliance, comprising the PMLN, Jamiat Ulema-i-Islam (JUI) and Jamaat-e-Islami (JI) but not the PTI, simultaneously takes shape.

Based on the findings of this survey, in this first scenario, the PPP and its allies may be able to secure 38.1 per cent of the vote. A grand anti-PPP alliance, which discludes the PTI, may secure 29.5 per cent. Together with the PTI, this grand opposition alliance may give the PPP a difficult time.

Scenario 2
Alternatively, the PPP forms an electoral alliance with its current allies, the ANP and the PMLQ – but not the MQM, which opts for the grand anti-PPP alliance. Here too, the PTI chooses to remain alone.

In this instance, the PPP and its allies would likely capture 33.9 per cent of the votes. An opposition alliance could secure with ease 33.7 per cent. This scenario would also result in the formation of a minority government, albeit a weaker one with a more formidable opposition. Either the PPP and the PTI would form the opposition to a PMLN-led government or the PMLN and the PTI would do the same against a PPP-led government.

Scenario 3
In this scenario, the PPP forms an electoral alliance with its current allies, the ANP, the MQM and the PMLQ – while the PMLN and the PTI form an opposing alliance. In this case, the PTI and the PMLN would, with 45.0 per cent of the votes, comfortably sail through the elections, forming a comparatively stable government at the centre. They may be joined by a number of other anti-PPP parties.

Scenario 4
The PPP and the PTI form an electoral alliance, while the PMLN forms a grand opposing alliance. In this instance, the PPP and the PTI would receive 49.3 per cent of the votes, and easily form a stable central government.

Scenario 5
The PPP forms an alliance with its current allies, while the PTI and the JI form an alliance; concurrently, the PMLN forms an alliance with the JUI and other anti-PPP parties. In this instance, the PPP and its allies would receive 38.1 per cent of the vote, the PTI and JI pairing would capture 23.9 per cent of the votes, while the PMLN led alliance would receive 25.9 per cent.

Verdict

So, what electoral outcome seems most likely?

Politics in Pakistan is predicated on uncertainty but, according to Dr Suleri, there’s a high chance that the third and fourth scenarios may never materialise.

Moreover, given that the country is far from being a homogenous entity where support is spread uniformly across all constituencies, the actual outcome may be entirely different from the ones described above.

Indeed, in the first-past-the-post electoral system, percentage of votes scarcely translate into a similar proportion of seats in the National or provincial assemblies.

It is evident, however, that no single party currently stands to sweep the upcoming polls.

It also appears, based on survey findings, that the PPP may have to retain its current allies to maintain its present political clout – and that, amidst the traditional PPP-PMLN toss-up, the PTI is emerging as a political reality.

What is most certain, says Dr Suleri, is that whoever does manage to form the government will most likely have to contend with a strong opposition. Moreover, if responses collected in the Political Barometer are any indication, it may also find it difficult to figure out what issues to tackle first, in order to soothe an electorate clamouring for change.

Methodology

Whereas many surveys conducted in Pakistan base their sample along provincial demographics, the Political Barometer’s sample of 1,283 respondents was based on Pakistan’s ethno-linguistic identities — the Baloch, Hindko-speakers, Pakhtuns, Punjabis, Seraiki-speakers, Sindhis, Urdu-speakers and others.

Stratified sampling was used to reflect voter preferences across ethnicities, genders, age groups, urban or rural localities, education levels and income brackets.

An abridged version of this story was first published in Dawn. Full results of the survey, which was carried out in 54 districts by a team of around 100 interviewers, can be found here

4 thoughts on “Elections 2013: survey indicates close contest

  1. The sample size of this survey (less than 1300) may not reflect the actual election fever that is visible across the country. The stratified sampling and its extension to180 million heads could mislead those who observe politics via Pakistan’s booming but elite oriented electronic media. The family-personality-blend coupled with the Religious-Sectarian Beliefs may prove decisive as well this time around. In the KPK, the political parties’ Af-Pak War/Drone related policies will affect the voters’ final decisions. Candidate-oriented constituencies will show results reflecting the (changing) loyalties of the candidates than those of the voters. The family system in the rural areas will belie even well-calculated projections, which are based on the one-head one-vote ideal democracy.
    Releasing pre election surveys were once considered something valuable but the BAN on these survey during the LAST US Presidential Elections proved beneficial. Lets establish research centres on Pakistan’s rural electoral psyche and behaviors…than mere issuing surveys and forecasts:-)

  2. These are technically flawed surveys with special design to promote the image/popularity of certain parties and to understate that of others. Frankly speaking (I m not a PPP worker) but I know the vote bank they have in Sindh, KPK, Balochistan and even in Punjab. PPP has won majority of by-polls even in Punjab. I wonder who undertake these surveys for these so calledsurvey institutes? How do they take sampling? How do they ensure credibility and reliability? They are seriously misleading, and could even lead to chaos and bloodshed if actual results do not produce the picture they paint.. It is also the responsibility of the media to check the credibility of such surveys before making such reports on air.

  3. Poorly laid out graphs. can you please revamp them so they are readable. Also the colors are so bad that its hard to make out even when expanded. Love the article thou.

  4. dont go on surveys the result always turn out into unpredictable one cuz pakistani survey are mostly based on few thousand of people and the silent voters only come into action when the day come

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>