December 16, 2008

Absurd statistical gimmick in Ethiopia’s 2007 census

By Cens Kerk | December 15, 2008
Absurd statistical gimmick in Ethiopia’s 2007 census and the need for accounting for the missing 3.0 million people in Amhara and Addis Ababa
When the second population and housing census was conducted in 1994 (the first was in 1984), the total population of Ethiopia was 53.5 million, putting the country the second most populous in Africa. There were no surprises or inconsistencies in the results of both censuses. Unlike its predecessors, the 2007 National Population and Housing Census was accompanied by long delays and the results contained horrific inconsistencies and unprecedented errors.

The much awaited population census result was made public a couple of weeks ago. According to the new census, the country is home to only 73.9 million people in 2007, significantly lower than the 79.1 million expected based on the 1994 census and various demographic surveys carried out between 1994 and 2007 (see World Bank, Ethiopia at a Glance). The population growth rate has been reduced from 2.6 to 2.51. What is striking about the census is that the lower overall population number or growth rate is attributed to a sharp decline mainly in two regions, namely Amhara and Addis Ababa.

According to the 2007 census, population growth rate varies markedly among the different regions: the highest annual population growth rate is observed in Gambella region (4.1 percent) followed by Benishangul-Gumuz, growing at a rate of 2.94 percent (Table 1). Oromiya and SNNP are reported to be growing at 2.90 percent per annum over the same period. The growth rate for Tigray is 2.48 percent. 2.55 percent for Somali, 2.40 percent for Dire Dawa, 2.22 percent for Afar and 2.61 percent for Harari. By contrast, the growth rate for Amhara is the lowest, only 1.70 percent. The capital city, Addis Ababa, is also reported to be growing by only 2.01 percent per annum, the second lowest growth rate. The sudden drop in the growth rate of Amhara is totally inconsistent with trends and statistical facts until the current census report. Why are the growth rates so low in Amhara and Addis Ababa? Is there any fundamental economic transformation or any disaster of biblical proportion taking place in the two regions? Similar doubts were also raised when the census report was recently presented to the Parliament but no satisfactory responses were given.

Available demographic and socioeconomic evidences demonstrate that the Amhara region is in no way different from the other regions. Table 2, for instance, portrays total fertility rates by region and it can be observed that the Amhara region has the fifth highest total fertility rate and the third highest mean number of children ever born to a woman of 40 to 49 years old. In other words, Amhara and Tigray regions have the same fertility rates (5.1). The mean number of children born is, in fact, higher in Amhara (7.0) than in Tigray (6.8). Only SNPP (7.5) and Oromiya (7.1) have a higher mean number of children ever born to a woman than Amhara. This is to be expected because women in Amhara are not better educated than their counterparts in other regions: illiterate women in Amhara region account for 69.5 percent, compared to 63.0 percent in Tigray, 66.1 percent in Oromiya and 69.6 percent in SNNP (Table 3). The Amhara region also shares similar features with other regions of the country such as Oromiya, Benishangul-Gumuz, and SNNP in terms of wealth distribution (Table 4). Not surprisingly, close to 84 percent of married women in the Amhara region did not use any contraceptive in 2005 (Table 5), a feature akin to regions like Tigray, Oromiya, Benishangul-Gumuz, Gambella and SNNP. It should be made clear that all the information (tables 2 to 5) is from the 2005 Demographic and Health Survey (DHS) report which is based on a nationally representative data collected only two years before the new census by the Central Statistical Agency (CSA). CSA is, a government institution which also conducted all the census surveys, including the current one.

There is absolutely no justification for the 1.7 percent growth rate in the Amhara region. If we take a conservative growth rate of 2.51 percent, which is the same as the national average, then the true population of the Amhara region is 19.09 million, not 17.21 million (Table 6). Nearly two million people have ceased to exist in the Amhara region not because of any sudden earthquake or tsunami but because of absurd statistical error or unprofessional manipulation. Given that Amhara is among the poorest regions in the country, budgetary and grant allocations which exclude the two million people of the region would only worsen the poverty and deprivation.

With regard to Addis Ababa, it is true that the administrative region has the lowest fertility rate or the lowest mean number of children ever born to women. However, the population growth of the region is driven more by migration than by natural growth. According to the 1999 Labour Force Survey, the rate of net-migration in the city was 3.9 percent per annum. We believe this rate of migration has increased in recent years as evidenced by the rapid expansion of the city over the last 10 or 15 years. The recent boom in the housing construction in all corners of the city is unprecedented. If we add a minimum natural growth rate of 1.0 percent to the 3.9 percent of net-migration (which is an underestimate by all account), the annual growth rate of Addis Ababa’s population is 4.9 percent, which is more than double the growth rate given in the new census report (2.01%). Accordingly, the true population of Addis Ababa is 3.93 million, not 2.74 million as given in the new census report (Table 6). Again, the 1.2 million people missing in the new census do exist and need to be counted.

In short, Ethiopia’s total population in 2007 was not 73.8 million but rather 76.9 million. A total of 3.0 million people in Amhara and Addis Ababa have not been accounted for in the latest census. The anomalies in the report are extreme with serious political, social, economic, legal, ethical and human right implications. This blatant error is unprofessional and deeply embarrassing for the government. We trust that CSA still has statisticians and demographers capable of checking the facts and taking corrective actions without delay. We also call upon multilateral and bilateral organizations operating in Ethiopia, including the United Nations Population Fund (UNFPA), the World Bank, the IMF, the African Development Bank, USAID, DFID, SIDA and CIDA as well as statisticians, demographers and related professionals to look at the inconsistencies and gross underestimations. Any delay in acknowledging the existence of the 3.0 million people would only mean that the error occurred by design, not by accident. This incidence, if not corrected immediately, should also cast serious doubt on all quantitative and qualitative information so far provided by the government. Improvement in per capita income requires a sound development strategy and a concerted effort to increase the numerator or the GDP, not a sudden slashing down of the denominator or the size of the population.

In our next report, we will critically look at the current census report by major religions.

No comments:

Blog Archive

የአቡነ ጳውሎስ "ሐውልተ ስምዕ"

ነጻ ፓትርያርክ ምርጫ ቢሆን ኖሮ ማንን ይመርጡ ነበር? እንበልና ሁሉም ነገር ሥርዓቱን ጠብቆ የተከናወነ የእጩዎች ምርጫ ቢሆን ኖሮ፣ አሁን የምናነሣቸው ጉድለቶች ባይኖሩ ኖሮ፣ 6ኛው ፓትርያርክ እንዲሆን የምትመርጡት ማንን ነበር? (ማሳሰቢያ፦ አሁን ያለው ክፍፍል እና የመንግሥት ተጽዕኖ ባይኖር ኖሮ ተብሎ የሚመለስ ጥያቄ ነው። የምን “ባይኖር ኖሮ ነው” የሚል አስተያየት ካለዎትም እናከብራለን።)