Are wages in Africa too high?

Today I sat down with a Washington Post reporter and several CGD colleagues to discuss under-reported topics in globalization and development. Inevitably we came around to the topic of Chinese development and manufacturing, and whether special conditions there (low wages, an undervalued exchange rate, and so forth) are contributing to de-industrialization elsewhere in the world.

The impact of China on global manufacturing is certainly no under-reported topic, but the discussion usually focuses on the consequences for the US. One of my colleagues, Arvind Subramanian, suggested that African textile and clothing firms have also taken a real hit, and that the culprit could be Chinese exchange rates. Other colleagues, like Vijaya Ramachandran and Kim Elliott, saw the decline as relatively mild and wondered, if Arvind is right, why we haven’t seen more of a fall.

Of course, the big fact that jumps out at you when you look at industrial production in Africa is not the ups and downs, but just how little of it there is.

Rather than ‘blame China’ (regardless how well that blame may or may not be deserved) I wonder whether there are other culprits that account not just for the decline in, but the persistently low level of, Afincan industry.

One thing that has always struck me in the African countries I have worked is that the real wages (i.e. wages adjusted for the cost of living) of African formal sector workers seem to be incredibly high, at least compared to that of workers in China or India. Given that firms in China and India seem to be more productive than their African counterparts, it creates a double disadvantage for African workers, and raises the question of why the situation continues. Why don’t manufacturing wages fall in Africa, stimulating more jobs for more people at wages still higher than those available in agriculture or informal business?

Why, when I run a survey in rural Uganda, do youth with the same education and experience expect a wage three to four times higher than the youth I worked with in India? I don’t begrudge anyone anywhere a living wage. It’s the relative differential that puzzles me, and that could be keeping Africa from doing business globally.

There are probably lots of plausible reasons. Perhaps we ought to consider (and get data on) the informal sector in Africa, which could be larger and have more moderate wages than the formal sector ones. It may be that all my notions and data about African wages are erroneous.

Another possibility, however, is that the largest employers of skilled workers in most African countries are international NGOs and the local government. They are competing, in many cases, for the same pool of skilled and semi-skilled workers as the manufacturers and service sector firms. Neither the government or NGOs, moreover, seem to set wages according to the local market or local conditions, and it requires little imagination to wonder whether they set their wages higher than the market would normally do.

Could the government and NGOs be distorting local wage markets and pricing African industry out of the world market? I don’t know, but this is a question some economist ought to start investigating.

I have seen the mechanism at work on a small scale in Kitgum, the town where much of my northern Uganda research was based. Large NGOs and UN agencies have begun to drive up local wages as they offer salaries and benefits many times in excess of the local norm. I don’t know if firms have been crowded out by this rise, but I have first hand experience how smaller local NGOs (and research projects) cannot compete. I also can’t help but notice that the best and the brightest pursue degrees in social work, not business. This is not necessarily a bad thing, but it does not feel terribly sustainable. You can’t build a national economy on NGOs.

A similar concern pervades aid in general, of course. Some macroeconomists point to the undervaluation of the Chinese currency (its exports look cheap to the rest of the world) and contrast it to the overvaluation of many African currencies (so that their goods look expensive). The reason? Hard to say, but some blame aid inflows that are large relative to the size of the economy.

By this argument, the incoming funds create much needed roads or deliver humanitarian aid, but at the cost of stifling jobs and industrial growth because of rises in real exchange rates. Even farmers are hit hard by these forces, since their agricultural produce begins to look expensive to the rest of the world.

I don’t know what the problem is, let alone the solution, but this is one subject not being sufficiently examined in academia or the media. I welcome references to existing work in the comments section.

7 Responses

  1. I got quite a giggle when I read this little piece. I moved to South Africa from Sweden in 1981. After a few years there my ex-wife got a contract to provide psychological counseling services to the American Embassy in Pretoria. Because of this contract our family became more of a part of the diplomatic community socially than would have otherwise been the case.

    One thing that I noticed was that the State Department was shopping old American union organisers and officials all over southern Africa through the USAID programme, I think, to “empower African workers”. I think that was the phrase used. At first I thought that this was a programme peculiar to the American policy towards South Africa. As time went on, however, I discovered that these recycled American union people were setting up credit unions and helping labour organisers in a variety of southern African countries and not just South Africa.

    It struck me at the time as a peculiarly caring sort of job creation scheme to help out the union organisers at a time when the Reagan administration was doing its level best to bust unions in the US. (Recall the air traffic controllers’ strike where Reagan simply fired the lot of them).

    In any case, my guess is that the author’s observed level African manufacturing remuneration as a function of productivity might have been at least partially a result of that American effort.

    I know that at the time of the transfer of power from the National Party to the ANC in the early 1990’s it was well understood amongst anybody who actually cared that South African manufacturing firms would not be able to compete with the Far East because of the high pay for workers relative to their productivity. :-(

  3. I think your observation is closely related to one made in Rajan and Subramanian, “Aid, Duth Disease, and Manufacturing Growth.”

  4. There was a nice paper on this by Lindauer and Velenchik some years ago. There is actually a lot written on African wage levels. My sense from the World Bank Enterprise Surveys data (worker surveys) is that they are high relative to India and China, reflecting both formal sector rigidity AND a shortage of educated higher skilled workers. But also, they are not THAT high—it doesn’t drive the manufacturing story entirely.

  5. Some macroeconomists point to the undervaluation of the Chinese currency (its exports look cheap to the rest of the world) and contrast it to the overvaluation of many African currencies (so that their goods look expensive). The reason? Hard to say, but some blame aid inflows that are large relative to the size of the economy.

    I thought there wasn’t much of a mystery here. The (very convincing) story I’ve always heard was one of government voluntarily and intentionally having overvalued currencies to support their middle classes (who import much) and their industrialization schemes (machinery is imported). Even if active pro-overevaluation policies are less prevalent with the end of marketing boards and capital controls (and the devaluation of the CFA Franc at Ivory Coast’s request), the tendency remains since oil, minerals and certain cash crops tend to be viewed as “ressources” less responsive to curency variations and are a huge input of foreign money in African economies anyway (who cares about exporting rice when you have enough money to buy it ?).

    I don’t think aid flows are that important. Unless you take into consideration aid/loans that allows government to continue the aforementionned policies.

    Another possibility, however, is that the largest employers of skilled workers in most African countries are international NGOs and the local government.

    You forget extaction companies. Wages in oil, mines, some cash crops and their service clusters tend to be higher than they should. The reason ? Well, with powerful unions and governments (and citizens) perceiving those sectors as leased state propriety at best, those companies, public or private either mantain high wages to avoid future trouble or are forced to pay high wages by government intervention.

    And when you think of government wages, don’t limit it to the official wage. The revenue one could get from a government position explains why there were more people passing the Custom’s exam than the teacher’s exam, even if in theory, teachers are better paid.

    And Paul has a point. Even if it’s easy to argue that those same wages in the formal sector are probably behind the lack of green revolution in Africa. After all, it gave the urban middle-class the ability to import, killing the demand for local agriculture.

  7. I always assumed the wage difference was due to the lack of green revolution and efficient agriculture. A subsistence wage in China and India isn’t enough to live on in many parts of Africa.