You are here: Home / Academia / Some Desi Electoral Data
Some Desi Electoral Data

Some Desi Electoral Data

I was asked to run some numbers using local electoral data today, and a couple of potentially interesting things came up. The first is a visibly pronounced blip in party seat and vote-share between 1997 and 2002, and the second is an observation on where larger parties stand on the rural-urban seat issue.

As most stories go, General Musharraf’s mauling of election laws, back in 2001-02, produced a number of breaks with previous electoral patterns. Some of the more apparent ones are: the number of first-timers – almost as many as were seen in the 1985-88 assembly under Zia; the high degree of horse-trading – Patriots jumping ship to ‘secure passage for Mohtarma’; and a fall in the average age of the members – the last one being a direct outcome of the graduation requirement that gave many daughters, sons, and younger brothers their chance at playing politician, politician.

A less studied, but academically more interesting consequence though is the sharp disruption in Pakistan’s trajectory towards greater concentration of votes and seats between two political parties. By initiating a fat, heavily bankrolled process of local government, and by carving out a section of loyalists from existing parties (PML factions, mostly), Musharraf ensured a diffusion of political and social capital, which probably made his job as the chief executive a lot easier.  I’m sure there’s an appropriate scientific analogy here, (something to do with molecules and atoms splitting and convalescing), but since I don’t know ‘SCIENCE’, I’ll show you a graph instead.

 

Vote-share for Top 2 and Top 3 parties, per election

Vote-share for Top 2 and Top 3 parties, per election

 

Seatshare over time

Seat-share for Top 2 and Top 3 parties, per election

 

Yes, it’s easy to point out that the vote-share trend was already downwards prior to the 2002 election, but the gradient changes very sharply. More than that, the vote-share for the top 2 parties takes a far steeper fall, as opposed to the same figure for the top 3 parties – an obvious consequence of PML-Q’s creation, and then its success at the polls. Just by looking at the two subsequent elections it seems that the almost-two-party model of yesteryears is nowhere close to being replicated. In 2008, the PML-Q put up a decent showing, and kept itself relevant as the third party in parliament. However after five years of falling prey to factionalism, leadership quarrels, and scavenger attacks by other parties, the party fell apart, and its political space – in numbers alone – was taken up by the PTI in this year’s general election.

Does this mean there’s organic room for a large, almost-national third party in this country? Going by voxpops from the urban middle class, one would think there is (sab chor hain, naik qayaadat ki zaroorat hay etc). But the truth is we don’t quite know. A military intervention grossly distorts natural political trajectories, and creates new political economies, new stakeholders, and new incentive structures. Absorbing all these new stakeholders back into a two party federal system might take a long time, and in all likelihood, several electoral cycles. So for example the suburban/aspirational middle class in Punjab – i.e. a beneficiary of Musharraf ‘s decade in power, and now part of PTI’s voter base,  – is a fairly new stakeholder in the electoral process, and one which will maintain an independent identity for some time to come.

The most obvious question that comes out of this is whether a two-party system is even desirable. That’s a hard question to answer, and I would be very tempted to say that it isn’t – especially in countries as diverse and fractured as Pakistan. The most we can hope for are two stable coalitions of aligned interests, composed of parties with fewer internal contradictions.

The second thing I wanted to bring up was the issue of categorizing voters along urban-rural lines. Data-gaps, especially on basic census stuff being as salient as they are, it’s very hard to divide constituencies into either-or category. What a few researchers have come up with though is an urban-rural-mixed typology that’s based on census projections and ECP data. A constituency is classified as rural if 80% of its population resides in areas classified as ‘rural’ under the 1998 census; urban if greater than 60% of its population resides in towns/municipalities/cities; and mixed if it falls anywhere between the two. Now these are fairly arbitrary, but it is the closest anyone’s come to producing a usable categorization.

This is what the classification looks like for each province:

Rural Urban Mixed Total NA Seats
Punjab

76

27

45

148

KP

26

2

7

35

Sindh

20

22

19

61

Balochistan

9

1

4

14

FATA

11

0

1

12

ICT

0

1

1

2

TOTAL

142

53

77

272

And in percentage terms:

Rural (%age) Urban (%age) Mixed (%age)
Punjab

51.4

18.2

30.4

KP

74.3

5.7

20

Sindh

32.8

36.1

31.1

Balochistan

64.3

7.1

28.6

FATA

91.6

0

9.4

ICT

0

50

50

These pie charts below show the percentage breakdown of rural, urban, and mixed seats won by the three largest parties:

Urban Rural partyshare PMLN urban rural partyshare PPP Urban rural partyshare PTI

 

Simply put, no party can afford to ignore urban voters, regardless of which province they got their seats from. The PPP has the greatest amount of space in this particular configuration – given the de-urbanisation patterns currently being witnessed in Sindh – but for a party with national ambitions, it would have to adapt to Punjab’s changing geography. The PML-N, the party currently lording over the greatest number of seats in Punjab, would have to work very hard in these five years to retain its existing electoral gains, while the PTI would have to invest in strong organization structures at the ward/union council level in urban areas.

What I find slightly strange though is that despite a fairly even split in terms of seat-attainment, the larger parties aren’t very vocal in outlining their own rural or urban character. The PML-N’s leadership is all fairly urban in its outlook, but the party still won a large number of rural seats. How do those rural winners figure into the party’s policy-making, and planning? With PTI, it’s even more strange, given that their outward projections scream ‘urban middle class’, while they actually won in a province with very few urban seats. Remains to be seen whether these parties can fix their identity issues over the next five years.

About Umair Javed

Umair is a political economy researcher based at the Center for Economic Research in Pakistan (CERP).

Comments are closed.