Afghanistan’s Demographic Drought

The Afghan presidential election is finally over. After the first round of elections failed to produce a winner on April 5, the Afghans went back to the polls on June 14 to choose between the two frontrunners, Abdullah Abdullah and Ashraf Ghani. Soon after the run-off, even before the release of the preliminary results, Abdullah’s team claimed that two million votes in favor of his opponent had been fabricated. Afghanistan’s Independent Election Commission (IEC), then-President Hamid Karzai, and Ghani rejected the allegations. The dispute incited a long and bitter political battle — involving even the threat of a "parallel government" — but finally came to an end thanks to pressure from the international community, which was led by U.S. officials. The two candidates compromised and agreed to form a national unity government. In a ceremony in Kabul on Sep. 29, Ghani was sworn in as president and Abdullah as chief executive. Afghanistan survived a possible catastrophe… for now.

One of the most important lessons to be learned from the recent electoral crisis is to deal with the lack of demographic data in the country. This absence of reliable population data has always been a problem for the Afghan government, and it continues to create challenges in many areas, from effectively planning long-term development projects to efficiently running a general election. The impasse caused by the allegations of electoral fraud, and which brought the country to the verge of chaos, partly originated from this problem.

"There are three kinds of lies," Mark Twain famously wrote. "Lies, damned lies, and statistics." In Afghanistan, where there is a chronic poverty of statistical data, almost all official statistics are rather politicized estimations. In the absence of accurate demographic data on the geographic distribution of the population, it is easy for authorities to cover up corruption, favor regional affiliations in the distribution of economic resources, or manipulate votes.

In the recent presidential elections, in Khost province, to mention only one example, the total number of votes in the first round, as released by the IEC, was 113,083 for all eight candidates. In the run-off, the total number of votes increased dramatically to 400,160. As no reliable statistics exist on the population of eligible voters in Khost, it is difficult to prove that the additional 282,077 votes either are or are not fraudulent.  

Unlike the United States, where a national census occurs every 10 years, there has never been a complete survey taken in Afghanistan. Various unsuccessful attempts, however, have been made to count portions of the populations, at least since the reign of Amir Sher Ali Khan (1868-1879), who wanted to count people for taxation purposes. These attempts failed for countless reasons, including war and violence, the lack of political will, and in one instance, because of the public’s traditional suspicion of government employees. In the early 1900s, when Amir Habibullah Khan (1901-1919) wanted to conduct a census in Kabul city, it caused a huge public panic in the Afghan capital. According to Ernest and Annie Thornton, an English couple who were in the service of the amir, when government officials went door-to-door to gather information, all of the men in Kabul fled the city or "hid themselves in diverse strange ways," fearing that their names were required in connection with some kind of crime of which they were not aware. Soon after, the government abandoned the idea of conducting the census.

In 1972, the Ministry of Public Health initiated a sampling survey that was funded by the United States Agency for International Development and carried out by experts from the State University of New York at Buffalo. Although it was the first methodologically rigorous survey in the country, the representative data it yielded was limited in scope, as it was based on interviews with only about 21,000 sedentary households; Afghanistan’s nomad population was not included.

In 1979, when the Communists were in power, Afghans started to conduct a national census for the first time, with technical and financial assistance from the United Nations Fund for Population Activities (UNFPA). That attempt, however, was left unfinished because of threats from the mujahedeen. Although 80 of the census takers were killed by the Islamic fighters, they managed to gather information from almost all of the country’s urban areas, 40 percent of the rural areas, and 15 percent of the nomads before the process was stopped.

Yet, the shortfall in coverage was not the only problem with this project. The Afghan government altered the questionnaire, which was prepared by UNFPA, at the last minute. The night before the census was to begin, President Nur Mohammad Taraki appeared on public television and radio stations and ordered all census takers to remove the question regarding ethnic affiliations from the questionnaire, fearing that with the availability of such data, "doubt would be cast on the official claim to pre-eminence for the Pashtun." Nonetheless, this flawed and unfinished census still serves as the basis of all Afghan population estimations today, despite the fact that the past three decades of wars, migrations, and internal displacements have greatly added to the complexities of the Afghan demography.

In a multi-ethnic country such as Afghanistan, where power is distributed among ethnic groups based on their population size, statistical data is a powerful political tool. Power-sharing agreements based on the current official demographic estimations have been challenged by minorities who believe they have been under-represented in the Pashtun-dominated governments that have been in power since the country was founded in 1747. Pashtun leaders and politicians justify their political domination partly by claiming to be the majority ethnic group. This is why the French geographer Daniel Balland attributes the lack of "political will at the highest levels" of government as one of the main reasons every Afghan government has failed to carry out a comprehensive national census — such an undertaking would affect the very structure of Afghan power politics.

In 2013, the Afghan government, supported by the United Nations, launched a new national census that is set to be completed in six years. Yet even after completion, the project will do little to settle the ethno-linguistic conflicts in the country, as the census takers have been instructed, once again, not to ask people about their ethnicity and mother tongue. Therefore, the ambiguity over the ethnic and linguistic distribution of the Afghan population will continue to stir disputes over power-sharing bargains.

This lack of demographic data is a symptom of a larger problem. In most developing countries, government institutions are weak and traditionally underestimate the value of data. They have a brief history of systematic mechanisms for gathering, storing, and analyzing information. Sample surveys, with varied degrees of scientific precision, are usually substituted for national censuses.

In Afghanistan, where there is no reliable information on economic, health, and socio-demographic aspects of the country, the government is particularly unable to make evidence-based policies for development programs. Therefore, eradicating information poverty should be the first step towards dealing with other forms of poverty and moving towards a functioning, democratic, and accountable government. The new government should make it a priority to conduct a comprehensive national census. As long as the Afghan ruling elites continue to use the uncertainty in the demographic data as a tool for grabbing political and economic resources, holding a transparent election, among other things, will remain a remote possibility.

Ali Karimi is a Vanier Scholar at McGill University in Canada, where he is completing a Ph.D. in Communication Studies. Follow him on Twitter at: @TheAliKarimi.