Yet, politics has prevented any census from being held in Pakistan in 1991. The last was held in 1981. The four provinces of the country—Punjab, Sindh, the North West Frontier Province and Baluchistan—are strongly suspicious of one another's figures and disagree over the method of conducting a census whose data will be acceptable to them all. The result: outdated figures, conjectures and extrapolations are the foundations on which the country's economic and political planning rest.
Since the Partition, four censuses have been conducted—in 1951, 1961, 1972 and 1981. And going by history, population figures have always been a major instrument in the struggle for power the world over. In Pakistan, it has become a crucial factor. For, any census conducted now would radically alter the existing power structure in Islamabad.
At least, that is what earlier events had portended. The first signs of trouble erupted soon after the 1951 count when there were allegations that East Pakistan (now Bangladesh) had deliberately inflated its population figures in order to shore up its parliamentary and financial quotas. The inter-state tussle was once again evident in the 1961 and 1972 census exercise, when special measures were adopted, both during the count and at the tabulation stage, to protect the power structure from developing serious cracks.
Then, after the last census conducted in 1981, it was widely believed that the troubled province of Sindh had been grossly under-represented because thousands of women in the interior of the province—dominated by powerful feudal chiefs—were not included in the head count.
But matters really came to a head in 1991, after the Nawaz Sharif government completed a provisional household enumeration which showed a stunning 60 per cent increase in the number of households in the Sindh province, with some areas showing as much as a 600 per cent increase in population. If the distribution of Assembly seats was based on these figures, Punjab, which has traditionally dominated Pakistani politics, would have had to relinquish at least 22 Assembly seats to Sindh. This fear was openly voiced by former Punjab chief minister Manzoor Ahmad Wattoo, whose appeal for freezing his province's share of Assembly seats (115 of the 217 seats), was turned down by the Benazir Bhutto Government.
The 1981 census had come up with startling findings. While the growth rate in the national population at 3.21 per cent corresponded to the high birth rate, the Federally Administered Tribal Areas (FATA) showed a negative growth rate (-1.64 per cent), Punjab had a figure (2.75 per cent) below the national average, the NWFP and Sindh (3.23 and 3.56 per cent respectively) were somewhat higher than the national average and Baluchistan revealed a massive increase of 7.08 per cent..
By the time the 1991 household count was organised, a number of factors were in play. All the provinces as well as the ethnic groups realised the power of large population figures and every group wanted to inflate its numbers. Under the 1974 act on delimitation of constituencies, the 207 National Assembly seats were as follows. General seats: NWFP-26, Punjab-115, Sindh-43, Baluchistan-7. Reserved seats: Women-10 and non-Muslims-6. In 1978, General Zia raised the number of non-Muslim seats to eight. In 1984, he further enlarged the National Assembly by increasing Muslim seats to 207, the seats for women to 20 and the non-Muslim seats to 10.
For the 1988 general elections, held after the death of General Zia, the distribution of seats for the four provinces should have been on the basis of the 1981 census. The government, however, ignored the increase in the population of women and non-Muslims, treating their quotas as fixed. Since then, all general elections have been held on the basis of the delimitation that took place using the 1972 census data and any new census results will make the perpetuation of this old formula impossible.
As far as the fears of Punjab are concerned, it is easy to see why it feels threatened. According to the 1961 census, Punjab accounted for 61 per cent of the country's population. By 1981, the figure had fallen to 58 per cent. Its rates of population growth and female fertility are lower than the other provinces and it has not suffered any of the refugee influx that Sindh and Baluchistan can lay claim to. Even if a faircensus is possible, Punjab faces a cut in its National Assembly seats. These fears had in fact come to the fore following the 1991 household enumeration.
Notwithstanding the tentative nature of the figures released by the Federal Interior Ministry, the population increase in Pakistan had broken all international records. The orthodox ratio of increase was 40-50 per cent per decade and, for cities like Karachi, Bombay, Tokyo and London, 70-80 per cent. For instance, the average percentage increase in Punjab in the 1991 household enumeration was only 45 per cent while Karachi's was 89 per cent. But in the interior of Sindh, the increase ranged from 771 per cent in Naushero Feroz to 397 per cent in Larkana, 288 per cent in Jacobabad, 265 per cent in Sukkur, 272 per cent in Shikarpur, 375 per cent in Dadu,315 per cent in Sanghar and 135 per cent in Hyderabad. These results reduced the share of the urban population from 45 per cent in the 1981 census to only 23 per cent in the latest housing enumeration.
Surprisingly, even Sindh, which stands to gain if the 1991 house count is used as a basis for the census, has reservations about it. This is related to the conflict between the Mohajirs and Sindhis. Most of the increase is attributed to the massive influx of illegal immigrants who, the government feels, are mainly responsible for the ongoing bloody ethnic strife. Besides, a major official hike in the number of Mohajirs would help their growing movement for a separate state, not to mention their demand for greater quota in jobs and educational institutions.
A similar situation prevails in NWFP and Baluchistan, where a large number of Afghan refugees are concentrated. Any census which takes them into consideration would automatically grant them legal status as residents. In fact, the Baluchi government has adopted a motion seeking postponement of the census until the Afghans and other illegal immigrants are identified and deported.
The Punjab government believes that it is already being assessed at 5 per cent less than its actual strength of 62 per cent of the entire population. In fact, some people in Punjab fear that the household enumeration, as in 1991, could show a fall to 42 per cent in Punjab's population, reducing the state's National Assembly seats from 115 to 92 with a proportional increase in Sindh's representation. A similar reduction would occur in Punjab's share of the federal divisible pool of taxes. This is a situation no provincial government can afford.
To add fuel to fire, Opposition leaders have alleged that the entire exercise has been planned to increase Sindh's quota of National Assembly seats in order to ensure a comfortable electoral position for the ruling Pakistan People's Party, whose leader Benazir Bhutto belongs to Sindh. Worse still, no one can agree on how to hold the census. There are several suggestions: Punjab has demanded that the national census be held after imposing a curfew throughout the country on a single day, that the exercise be supervised by the army, and monetary allocations be frozen at the 1981 census level to discourage smaller provinces from exaggerating their figures. The postponement of the census in December last year was meant to resolve these doubts and evolve a fresh policy on the census. But the federal government failed to allay the fears of Punjab, and the deadlock continues. For Pakistan, it seems to be time all heads were put together to take count of the situation.
