The population census has finally begun in Pakistan, nine years overdue, and nineteen years after the last census. As is always the case, it is a highly contentious exercise, with the country’s ethnic, religious and sectarian minorities expressing fears about under-representation, and marginalized groups such as people with disabilities and transgender persons once again in fear of being left out altogether, or being counted at par with the majority population. In a country as riven by multiple orders of exclusion as this one is, these fears are not unfounded. Combined with these fears is the fact that the census is key to determination of delimitation of constituencies, and allocation of seats in national and provincial legislatures. The fear of under-representation thus pervades the exercise. Of the many factors which make this exercise important, one which receives relatively little attention is the division of resources in the federation, which is also closely linked with the population count. The bulk of taxes in Pakistan are collected by the federal government, including a major share of the all-important general sales tax. A part of this collection then has to be filtered down to provincial governments, who barely raise 15 to 20 percent of their required revenue, and are thus dependent on the federal government for the transfer. The part of the tax collection that is transferred to provinces is called the federal divisible pool, and how it is divided across the four provinces has been a contentious issue since Pakistan came into existence. Four distribution awards were announced in Pakistan before 1971, where the focus was on the division of resources between the western and eastern wings of the country. The main basis for division was population, with the former East Pakistan getting 54 percent of the divisible pool, and the former West Pakistan getting 46 percent. At that time, using population as the main criterion was less contentious, as development levels in East Pakistan, which benefitted from the use of population as a criterion, were significantly lower than in West Pakistan. Despite this, the National Finance Commission (NFC), the body that determines the award, was severely criticised in East Pakistan, for not taking into account the disparities in development across the two wings when deciding on the award. Post-1971, with the role of the NFC now enshrined in the constitution, population continued to be the criterion for division of resources across the four provinces constituting Pakistan. But in the post-1971 era, this decision had very different implications. Now, the use of population as the sole criterion favoured a province that was already, as a whole, relatively prosperous when compared to the others. Six awards announced since 1974 granted roughly 58 percent of divisible pool taxes to Punjab, just below a quarter (or about 24 percent) to Sindh, 14 percent to the frontier province (as it was then known) and about 5 percent to Balochistan. Pakistan was more or less unique amongst countries with a federal form of governance, in using population as the sole criterion for distribution of resources across the federating units. Neighbouring India, for example, uses criteria as diverse as forest cover, area, ability to generate revenue, and income distance from the richest major state, in addition to population and population growth rate. Interestingly, it assigns the most weightage, almost 50 percent, to revenue generation and income distance, while population gets just 17 percent weightage, and a high population growth rate is actually included as a negative factor in the overall weighting scheme. After years of negotiations, and much hand-wringing from the smaller provinces, the 7th NFC Award announced in 2009 changed the formula for distribution of resources and shifted from just population to four criteria: population, poverty/backwardness, revenue generation and inverse population density. In effect, the key demands of three smaller provinces were incorporated, as Khyber Pakhtunkhwa (KP) had plumped for the inclusion of a poverty criterion, Sindh for revenue generation ability, and Balochistan for inverse population density. But population remains the dominant factor even now, with 82 percent weightage. Poverty gets 10 percent weightage, revenue generation 5 percent, and inverse population density, just over 2 percent. In effect, the use of additional criteria has changed the allocation of federal divisible pool resources across provinces as follows. While Punjab’s share has gone down by about six percent, that of Sindh and KP has barely increased by one percent, while that of Balochistan has increased by four percent. This is less radical, even for Balochistan, than it seems. The province has long been under-serviced, with policymakers routinely ignoring the fact that a small, scattered population demands more cost-intensive service delivery than a dense area. Coupled with the historic neglect of remote regions in Pakistan, Balochistan’s higher allocation represents the least possible redressal of long-standing grievances. The key point is that population remains the most important criterion for distribution of resources, and this is unlikely to change anytime soon. In a sense this creates a perverse incentive — it is in every province’s interest to show significant population growth or at least a greater share in Pakistan’s total population in order to be eligible for more funds. Until the NFC formula changes to tone down the emphasis on population, the census will remain a contentious exercise. The writer is an economist and policy analyst based in Islamabad