The mismeasurement of urban Egypt

The short version of this post is that despite recent reports, urbanisation in Egypt has not gone into reverse in the last 30 years. The Egyptian government systematically under-estimates both the size of and the growth in Egypt's urban population for adminstrative/political reasons. For the full story, see below the fold.


David Leonhardt of the New York Times wrote an interesting analysis of Egypt's social and economic malaise a few days ago (picked up by Matt Yglesias and others), kicking off as follows:

It would be easy to look at the images coming out of Cairo over the last few weeks and think of Egypt as a highly urbanized society. It would also be wrong.

When Hosni Mubarak took power in 1981, Egypt was indeed more urban than the rest of the world. About 44 percent of its population lived in cities. In East Asia, by comparison, only 26 percent of people lived in cities.

Since then, the cities of Asia have expanded rapidly, drawing in millions of peasant farmers looking for a better life — and, more often than not, finding it. Almost 50 percent of East Asians now live in cities. And Egypt? It is the only large country to have become less urban in the last 30 years, according to the World Bank. About 43 percent of Egyptians are city dwellers today.

I think the rest of Leonhardt's analysis is broadly correct, but these opening claims (accompanied by a trademark NY Times infographic) are wrong. It's not really his fault, since the World Bank does indeed publish the statistics he cites. But the World Bank gets them from the UN, which doesn't use a consistent measure of urbanisation but simply relies on whatever measure its individual member countries happen to use. And in Egypt's case the government systematically under-estimates both the size and the growth in its urban population.

But don't take my word for it - I'm getting all this from the World Bank itself, specifically its 'Urban sector update' on Egypt from 2008. A long extract follows, with emphasis added.

Levels of Urbanization and the Problem with the Official Definition of Urban Place
2.9 Official figures report that in 2006 a total of 42.6% of Egypt's population was urban, residing in 214 urban places. As Table 2 shows, this percentage grew slowly but steadily until 1976-1986, declined slightly over the 1986 – 1996 period, and remained virtually unchanged over the 1996 – 2006 period. Can it be that Egypt is the only developing country where urbanization rates have recently been declining or remaining constant? The answer is no, and such an aberration points directly at the problem of the definition of urban areas in Egypt. The Census records urban and rural populations according to an arbitrary administrative definition of what is an urban place. (see Box 1) As Denis and Bayat have documented in 2000, such administrative definitions have led to a gross underestimation of urbanization, one which is progressively more and more out of touch with reality

Table 2: Egypt Urban and Rural Populations 1947-2006 (in thousands)

Year	Urban	Rural	% Urban	% Rural
1947	6,363	12,604	34%	66%
1960	9,965	16,12	38%	62%
1966	12,033	18,043	40%	60%
1976	16,036	20,59	44%	56%
1986	21,216	27,038	44%	56%
1996	25,286	34,027	42.6%	57.4%
2006	30950	41631	42.6%	57.4%

            Source: CAPMAS Census various years; www.capmas.gov.eg

Box 1:  The official definition of urban places in Egypt

The official Census definition of urban areas in Egypt is purely administrative and thus is problematic.  Urban areas considered to be either:

(1) urban governorates – limited to Cairo, Port Said, Suez and, until recently, Alexandria;

(2) agglomerations which have been declared “cities” and have a city council, or

(3) the capitals of rural districts (marakaz) and capitals of rural governorates. 

This definition has no relation to the size of the agglomeration’s population or its importance as an urban area. As a result, the urban population of Egypt has been located largely in the same geographic space for decades.  The redrawing and reclassifying of Census areas by CAPMAS are rarely carried out, and the Ministry of Local Development is less likely to decree new urban areas for administrative purposes.[1] 

([1] Declaring a rural agglomeration to be administratively a city commits the government to provide higher levels of services and changes in representation in Parliament.) 

2.10              By any definition, Egypt is already overwhelmingly urban, with estimates varying widely depending on the definitions adopted.  Denis and Bayat carried out an analysis of the 1996 Census and, using the definition of urban places as settlement agglomerations of more than 10,000 inhabitants, calculated that Egypt was at least 66.8% urban, residing in 628 urban places. It could be argued, on this basis, that the urban population was 39.8 million in 1996, which if projected to 2006 would represent an urban population of at least 49 million inhabitants, or 67.5% of the national population ...

2.12 There are two main types of recent urbanization patterns that have been missed by Census enumerations:

2.13 Overspill from urban centers into village agglomerations in the agricultural hinterland: This could also be called spontaneous urbanization of agglomerations on the periphery of the large cities, in larger villages and in small towns...

2.15 Emerging small towns: Many small agglomerations in dense rural areas throughout Egypt have begun growing to reach well over 10,000 inhabitants, and their economic functions have begun to diversify away from purely agricultural activities. In fact, these “urban villages” or small towns can be considered market towns and the loci of trade, petty manufacturing, and services for the larger rural hinterlands, as well as the location for certain footloose enterprises.

2.16 The phenomenon of these emerging towns underscores an important point about the economics of Egypt’s countryside. Even in the most “rural” governorates non-agricultural activities predominate. For example Census figures for 1996 show that in all of rural Egypt, only 50.5% of working persons aged 15+ (8.40 million persons) were engaged in agriculture, fishing, and related activities.

I would also add that that when the World Bank came up with a new 'agglomeration index', designed to more accurately and consistently reflect the real extent of urban concentration, Egypt came out with an extremely high score of just over 90 out of 100, about as high as the West Bank and Gaza, and as far as I can see exceeded in the Middle East or North African only by Bahrain.

In light of all this, I think Leonhardt's argument needs to be tweaked a little. He writes:

This urban stagnation helps explain Egypt’s broader stagnation. As tough as city life in poor countries can be, it’s also fertile ground for economic growth. Nearly everything can be done more efficiently in a well-run city, be it plumbing, transportation or the generation of new ideas and businesses.

I think the issue is that Egypt's urban population is large and growing despite its cities not being 'well-run'. People may still prefer urban to rural life, but having a lot of city dwellers won't achieve much by itself if the overall economy is sclerotic and dominated by a corrupt, authoritarian government with no real interest in fostering education and innovation.

The fact that Egypt's urban population is much larger than a lot of people think also means that the potential for growth and development may be all the greater if and when the dead weight of the current regime (writ large, as opposed to just personified by Mubarak) is lifted.