17 years and no census in Pakistan — A country running on guesswork

Published September 7, 2015
The census needs a champion in Pakistan.—Illustration by Khuda Bux Abro
The census needs a champion in Pakistan.—Illustration by Khuda Bux Abro

Every sensible parent knows that the timely and frequent measurements of their newborn’s height and weight is essential for the child’s mental and physical development. A national Census is no different.

Conducted at regular intervals, a census allows governments, businesses, and others to take stock of the socio-economic health of the nation.

Without Census data, socio-economic planning is not much better than guesswork.

Rumours are ripe that the federal government is either hesitant or even reluctant to hold the decennial Census, which has been overdue since 2008. The last Census in Pakistan was conducted in 1998, only after a delay of seven years.

Also see: Census 2016

The Census should receive the same priority as the timely measurements of our newborns. Or else, like malnourished children, we will continue to raise a socio-economically malnourished nation.

Demographers, economists, and social scientists unanimously favour Census data because no amount of customised surveys or other databases, including Nadra, can be a substitute for the Census.

It appears that the government has not formally cancelled the Census. However, the funds needed to complete the task by March 2016 have not yet been released. An estimated 14.5 billion Rupees have been requested to hold the Census. The Pakistan Army will receive more than half of the requested amount (7.4 billion Rupees) to provide security for the Census.

Excluding transfers to the Pakistan Army, an estimated 40 Rupees have been budgeted to enumerate one individual. I find this estimate on the low side. In addition, large sums are needed to turn the Census data into research deliverables and insights and in research-ready formats for researchers in academia and public sectors.

Given the changes in the Statistics Act (Act number XIV), which was promulgated in 2011, the Federal Government is not bound to hold the Census at regular intervals. Part VI of the Act states:

The Federal Government may, from time to time, by notification in the official Gazette, declare that a census of population and housing conditions of Pakistan shall be taken by the Bureau during such period as may be specified therein.

Taking the Census is now more of a function of convenience than obligation. This would allow the governments to take the Census as per their short-term political needs. Such preservation of political self-interest is likely to harm the national interest.

Read on: Census challenges

Is there a substitute for the Census?

Some have erroneously argued that small surveys and other national databases could substitute for the Census. They are gravely mistaken. Neither in Pakistan, nor in an advanced economy, such as Canada, can there be a substitute for the Census.

In 2011, the Conservative government of the Canadian Prime Minister Stephen Harper replaced the long-form Census with a National Household Survey. The politically motivated decision by the Conservative government attracted rebuke from provincial and local governments, as well as academic and business researchers who knew that there could be no substitute for the Census.

The decision to scrap the long-form Census prompted the resignation of Canada’s chief statistician, Dr Munir Sheikh, who observed that a voluntary survey could not substitute the Census.

What about Nadra?

Some well-meaning individuals have advised the government that Nadra’s database could be a substitute for Census data. This advice, despite being well-intentioned, is based on ignorance of statistical methods, especially random sampling.

Nadra’s database is over-represented by those who need a national identity card or a passport. It systematically misses millions of destitute Pakistanis who see no value in the national identity card or a passport.

The primary data collection philosophy between the Census and Nadra differs. Individuals approach Nadra to be included in the database, this results in the response bias. The government, on the other hand, approaches all individuals to include them in the Census, thus limiting the response bias.

Another technical aspect of the Census is that it provides the sampling frame for all other surveys conducted by the government and other agencies. In the absence of a recent Census, the sampling frames are drawn from the 1998 Census. Furthermore, the delineation of political and administrative boundaries will have to be based on the dated 1998 Census.

Yet another key limitation with Nadra’s data is its proprietary nature. Nadra does not share its data with municipal or provincial governments. Academics and other researchers have no way of accessing Nadra’s data.

The Census, on the other hand, has a history of data sharing agreements with other tiers of governments, researchers, and others interested in public policy.

Lastly, Nadra does not collect the same details about households that are collected in the Census. Also, the systematic biases in coverage render Nadra’s data of little use for socio-economic planning. No wonder, analysts, such as Haris Gazdar, oppose any equivalence between Nadra and the Census.

Will Muslims outnumber Hindus in India? Not if you consult the Census

India conducted the national Census in 2011. It recorded 1.2 billion individuals. The recently released breakdown of the Census data revealed that the population of Hindus in India declined by 0.7 per cent. Hindus now constitute 79.8 per cent of the population. At the same time, the share of Muslim population increased by 0.8 per cent reaching 14.2 per cent of the total. This puts the Muslim population in India at around 170 million.

There was no reason for these numbers to rile up right-wing Hindu fundamentalists like they were. They are merely using these numbers to project the false image of a rapidly Islamising India, so as to strengthen the Hindu vote in the upcoming State elections.

Know more: 5 charts that puncture the bogey of Muslim population growth in India

If one consults the Census, one would realise that since 1991, the Indian Muslims inter-Census growth rate has been declining at a rate much faster than that of the Hindus. It is quite likely that the population growth rates of Hindus and Muslims will converge soon.

Amir Ullah Khan, who works with the Bill and Melinda Gates Foundation and was quoted in the Financial Times, believes that the convergence of growth rates has already taken place in the developed and literate States of Kerala, Tamil Nadu, and Karnataka.

If it were not for the Census, Hindu fundamentalists would continue spreading false alarms about the Islamisation of India.

Source: Indian Census data reported in the Financial Times.
Source: Indian Census data reported in the Financial Times.

What should Pakistan do?

Census needs a champion in Pakistan. I see no one more qualified and adequately placed than the Minister of Planning, Ahsan Iqbal, to champion the Census from within the government.

He should lead a parliamentary committee that is proportionately representative of all parties in the Parliament. The Committee should act as an advocate to take the Census and lobby for the necessary resources required to keep the completed task. Such a committee will help develop the political will needed for any task of such scale and scope.

It is also important that this Census use the state-of-the-art in technology for enumeration and tabulation of data. This will eliminate the errors when data are digitised later. Aerial photography and GPS units should be used to demarcate Census geography.

The government should consider collaborating with Google, which has extensive experience in collecting such data. In fact, such tasks could be outsourced to companies like Google so that data are digitised and archived using the global best practices. It may even be cost-effective.

In 2006, I established the GIS laboratory at the Population Census Organisation (PCO). UNFPA funded the project. I was tasked to train the employees at the then Federal Bureau of Statistics in digitising the Census geography.

I ran a training workshop in Islamabad for 20-plus demographers and statisticians. It is my understanding that not much came out of the exercise, because Nadra pulled rank and took the task away from the PCO. Little is known about digitising Census geography since then.

At the same time, every effort should be made to build in-house capacity at the Pakistan Bureau of Statistics, its counterparts in the provinces, and its sister organisations, to collect, archive, analyse, and disseminate Census data.

In addition, for sound socio-economic planning, data liberation must be a central theme for the next Census in Pakistan.

It is the primary responsibility of the government to report on the state of the nation. Without Census data, the government cannot advise the citizens on the state of the economy and society.

The government might plan for the future. However, in the absence of a census, ignorance, and not knowledge, will drive that exercise.



Security lapses
Updated 13 Apr, 2024

Security lapses

Ensuring the safety of foreign citizens is paramount, not just for diplomatic relations but for our economic future.
An eventful season
13 Apr, 2024

An eventful season

THE Senate chairman and deputy chairman were elected unopposed, and 41 new senators were sworn in on Tuesday,...
Living rough
13 Apr, 2024

Living rough

WE either don’t see them or don’t want to see them — not even when they are actively trying to get our...
Saudi investment
Updated 10 Apr, 2024

Saudi investment

The state has to address barriers that stand in the way of attracting foreign investment, and create a pro-business environment.
Charity for change
Updated 11 Apr, 2024

Charity for change

PAKISTANIS are large-hearted people who empty their pockets at the slightest hint of another’s need. The Stanford...