Losing Their Census
In the absence of a fresh census, the projection of earlier trends into the nineties points towards a major shift in urban-rural demographic ratios, leading to significant political repercussions…
Articles in the national press, letters to the editor in a number of newspapers and magazines, speeches and discussions in seminars, and statements by interested political leaders, have all expressed concern over the fact that the deciminal population census that was to be held in 1991, has not yet been conducted and has perhaps been indefinitely postponed.
Behind all the controversies are some major political stakes. The persons voicing this concern feel that there has been a major increase in the urban population during the last ten years. In fact, some sources place the urban population of Pakistan at over 45 per cent of the total population. In the case of Sindh, figures of well over 50 per cent have been quoted by certain political leaders and denied by others. It is felt by the authors of these concerns, that the holding of the census would lead to a major re-demarcation of political constituencies in favour of the urban areas. It is also felt that the results of the census would allow the urban areas to demand a greater share of the finances allocated for the Annual Development Plan (ADP). The supporters of this point of view feel that a census and a revision of political constituencies and financial allocations, based on the census results, would completely change the policies of Pakistan. They further claim that the census is being delayed precisely because the government wishes to maintain a status- quo in favour of the feudal interests in the country.
The deciminal census has been held regularly in the country since 1901. It has always been a fairly routine affair. In the aftermath of the Indo-Pakistan war of 1971, however, the census could not be held that year. But a year later in 1972, it was held even in very difficult circumstances.
The census is much more than a head count and an assessment of population growth in urban and rural areas. It establishes demographic trends, household size; sex ratios; age structures; marital status of the population and an age-sex pyramid. In addition, it establishes literacy and educational attainments; enrollment ratios in schools and the levels of education at every administrative level in the country. It also quantifies rural-urban, urban-rural, inter-provincial and out-country migration. Statistics are also developed for the labour forces by occupation and for unemployment. These figures are developed not only for national and provincial levels, but for every district of the country. As a result, the government can plan effectively for the changes that are taking place in society and can evaluate its past performance. Not only can national and provincial programmes be developed from this data but district and tehsil level projects can also be made appropriate to the social and economic issues that are identified by the census.
A number of government agencies carry out a regular census and surveys, other than the population census. These include the housing census, agriculture census and livestock census. In addition, surveys are carried out to establish labour statistics; household income and expenditure; trade statistics and transport, communications and services sector statistics. Socioeconomic and physical data is also generated for various cities and districts for master plans and development plans. However, except for the agriculture and housing census, most of this data is developed through sample surveys. But these samples are too small and in the past have been known to conflict with the census results. To be effectively used they must be seen in the larger context of a national census. In fact, even the housing and agriculture only makes sense, and can be related to each other, if a larger demographic picture is available.
As mentioned earlier, the major debate in the press, seminars and political statements regarding the census, revolves around the extent of urbanisation in the last fifteen years and the present rural-urban ratio. But there is no way that we can accurately assess this phenomenon. However, if we project the growth trends between 1971 -81, established by the last census, into the nineties and onto the year 2001, we get some interesting results.
In 1981, the urban population of Pakistan was 28.3 per cent of its total population and was growing at an average rate of 4.4 per cent per year against a total growth rate of 3.1 percent. According to this trend in 1991, the urban population of the country was 33 per cent and in 2001 it will stand at 38 per cent of the total population. In the case of Punjab, in 1981, the urban population was 27.6 per cent, in 1991 it was 33 per cent and in 2001 it will be about 40 per cent of the total population of the Punjab. At present, Punjab’s urban population would be just above 36 per cent.
In the case of Sindh, in 1981 the urban population stood at 43.3 per cent. At the 1971-81 growth rate it was 47 per cent in 1991; it is 48.5 per cent at present and will be 51 per cent of the total population of the province in 2001. Similarly, the urban population of the NWFP will increase from 15.1 per cent in 1981 to 21.1 per cent in 2001; and that of Balochistan will increase from 15.6 per cent in 1981 to 44.35 per cent in 2001. However, this massive increase in percentage terms in the case of Balochistan, is negligible in numbers as compared to the comparatively smaller increase in the case of the Punjab. Between 1981 and 2001, only 1.5 million persons will be added to the urban areas of Balochistan, whereas in the same period 28.7 million persons will swell the population in towns and cities of the Punjab. Thus, both in political and development terms, percentages of growth at provincial or district levels are far less important than actual figures.
This projection of the 1972-81 growth trends into the nineties points towards a major change in the urban-rural demographic ratio. In the case of Sindh, the new figures suggest that if political constituencies are redefined on the basis of the present population division, a major political change, with all its accompanying problems will take place. This will dramatically change the status quo. Urban Sindh would be able to determine whether the Sindh Assembly has or does not have a quorum at any particular session. To minimise the existing and ensuing conflict between rural and urban Sindh it is necessary to blur their differences as far as possible. A complete end to the quota system in government jobs and educational institutions would be a good beginning followed by an attempt to end the system of political patronage in business and trade. The changes due to urbanisation can only be accommodated peacefully if an overall consensus on a process of accommodation is arrived at between various actors in the Sindh political drama. In the Punjab as well, there would be a need for new political alliances and a change in the style of politics. A discussion on development issues and the policies related to them will have to replace rhetoric, jingoism and fanfare.
However, it is not proper to accept the results of projecting the trends established by the 1981 census into the present and the future. Physical and sociological sample surveys tend to suggest that the urban growth rate established by the 1981 census for the period of 1972-81, has increased considerably during the last 14 years, especially in the Punjab. Official documents are quoting figures of 6 to 8.5 per cent per annum growth for Peshawar and the Punjab’s secondary cities of Gujranwala, Faisalabad, Gujrat. And socio-economic surveys of katchi abadis and informal settlements, lend support to these figures.
It makes sense that the pace of urbanisation has increased in the more developed areas of the country and especially in the Punjab. In the last 15 years, a consolidation of capitalist farming and of green revolution technologies, introduced in the sixties, has taken place in the irrigated areas of the country. In the Barani areas, the development of a cash economy is forcing an increasing number of small farmers and landless labourers to send at least some members of their families to the urban centres so that they can survive on their remittances. Fifty per cent of all such rural-urban migration took place in the Punjab between 1972 and 1981. Indications are that this trend has increased in the Punjab and decreased in Karachi over the last few years.
The population of Karachi has also been the subject of considerable speculation and discussion. Figures of between 10 to 15 million have been quoted by political leaders, press articles, professionals and development agencies. On the other hand, there are those that insist that the population of the city is not more than 8 million. However, if the growth rate established by the 1981 census is projected, then Karachi’s population of 5.2 million n 1981 was 8.4 million in 1991 and will exceed 13 million in 2001. As such Karachi’s population was 27.33 per cent of Sindh’s population in 1981, 32.86 per cent in 1991 and will be 34 percent in 2001. Satellite imagery and socio-economic surveys carried out for the Karachi Master Plan suggest that the rate of growth increased substantially between 1981 and 1988 when these surveys were carried out. No trends have been established beyond 1988. Thus, there does not seem to be any earth-shattering change to report about Karachi’s population ratio vis-a-vis the rest of Sindh. However, there will be major political repercussions if adjustments are made to reflect the figures established by the 1981 census and the changes that have taken place in the demography of the city since then.
The urban-rural divide has been another popular topic of debate. It is all very well to talk about this divide. But then it is also necessary to determine what constitutes an urban area or an urban population.
Under the current definition, established for the 1981 census, any settlement that has a population of over 5000 and has a town committee is an urban area. However, there are thousands of settlements of over 5000 which do not have a town committee and almost every year the number of such settlements is increasing rapidly. In addition, before the 1981 census criteria, any settlement that had a population of over 2500 constituted an urban area. If we were to follow this earlier definition, then well over 50 per cent of Pakistan already falls into the urban category. But with the development of small scale industry in many rural settlements, mechanised transportation, communication networks, and links with city-based trade and commerce, there are large pockets of urbanised Pakistanis living in the rural areas of the country. The deciminal census would help us, not only to quantify this reality, but also to address the physical, economic and social issues related to it.
For the reasons discussed above, it is important that the government hold a population census. In addition, the undoing of a useful 90-year-old established procedure has already set a bad precedent. It has also deprived Pakistan’s planners of the most useful of all planning tools. Even if the government wishes to maintain the status quo, as it is being accused of doing, the census would help it in doing this more rationally and appropriately. Political and economic inequities are bound to increase if demographic and social changes are ignored and if they become a source of wild speculation and blatant political opportunism as they already have. And in any case, one cannot protect a status-quo the quo of which has lost its status.