Some African labour markets don’t work, but competition helps

Image: Flickr, SuSanA Secretariat
Image: Flickr, SuSanA Secretariat

Labour market regulation is a thorny issue in African economic policy debates. In many African countries, labour movements played an important role in liberation struggles.

Labour is therefore often well-organised, and labour market regulation is now remarkably well developed in many places on the continent. But do the labour markets they regulate work? From a social standpoint, the answer is often “yes”. Worker and broader social rights are well protected (often the International Labour Organisation’s guidelines are followed to the letter), and the critical link between an organised labour movement that is free to associate and negotiate and healthy democratic development is undeniable.

But from the economist’s standpoint, the answer is “no”, for two evidence-based reasons. First, these markets do not ‘clear’. That is, there is insufficient flexibility in wage levels to allow supply and demand to work. In a perfect world, wages adjust to a level that generates full (formal) employment. Second, and perhaps because of these ‘rigidities’, Africa’s informal sectors employ masses of people.

But this doesn’t get at the underlying problems. Why don’t labour markets clear? One way of addressing this question concerns the demand for labour in African economies. As we all know, African economies don’t create enough jobs, for all the myriad reasons always put forth to explain Africa’s poor growth and development record.

Another way of assessing the problem restricts itself to the functioning and regulation of labour markets themselves. In all market-based economies returns to labour (wages) depend on the productivity of that labour. The overall productivity of the firm, which is generated by the mix of labour, capital, land, technology, management, and any other inputs to the production process, is also important to worker remuneration, but in less direct ways.

The rule, call it an economic law if you like, is simple: returns to labour must track changes in that labour’s productivity. If mismatches develop, the demand for labour by existing firms will drop, and the incentives facing new firms deciding on their optimal labour-capital-technology mix tends to swing in favour of more machinery and less people.

Recent empirical research into this relationship in Africa reveals the following interesting insights.

In more developed parts of the world the wage-productivity ‘law’ seems to operate well. In the US, for example, the gap between wages and productivity is only significant with respect to one of the five commonly-studied ‘characteristics’ of the labour force: gender (the other four are experience, educational attainment, time in the current job, and formal on the job training). That is, women in the US are on average 16% less productive than men, but get paid on average 45% less. So if one limited one’s analysis of the US labour market to men only, thereby excluding gender considerations, one would find no mismatch between wage and productivity growth.

In France, Israel, and Norway, even the gender-based wage-productivity gap is insignificant.

This naturally does not mean that people are not rewarded for investing in themselves. On the contrary, people with higher educational attainment levels, more formal on the job training, more experience, and more time in the same job almost always earn returns higher than those that cannot boast similar achievements. This is because they tend to be more productive. If this wasn’t the case why would anyone bother getting an education? And why would firms invest in their workforce?

The research is simply saying that in these rich economies, if two people are identical in some or all the five respects, their remuneration is appropriately calibrated to their productivity levels. Should some be less productive, they will earn less, regardless of equivalence in these other respects.

In African countries that have been studied using the same methodologies, this is not the case. In Ghana, for example, women are found to be up to 62% less productive than men (perhaps because they are denied the same educational opportunities), but are paid only 12% to 15% less.

In Tanzania, wage premiums (rewards for having more education, etc) also deviate substantially from productivity. In particular, experience and schooling is rewarded far in excess of what the productivity levels of these people suggests is economically viable. On the other hand, more firm-specific attributes like formal training and staying in the same job do not elicit significant remuneration increases, despite having clear and positive impacts on productivity.

In Kenya, a richer and more diverse economy, these gaps are lower, and the economy as a whole, as well as the average firm, is more productive. But Kenyans on average receive fewer years of educated than Tanzanians.

It is by now considered immutable fact that education is a pre-requisite for improved economic growth. But in Tanzania, relative to Kenya, more schooling does not translate into higher output per worker. And why, when Tanzanian firms do invest in their workers and generate greater productivity, do they not then reward those workers for achieving precisely that?

Clearly education isn’t everything. A range of other factors are important, underlining once again that policy must address all challenges concurrently if the right results are going to emerge.

Competition is one area that does seem effective in reducing these wage-productivity gaps. Specifically, those firms engaged in international trade, meaning they face far greater competition in product markets, are more disciplined in their labour market interactions. These firms must produce good quality in as efficient a manner as possible. Efficiency depends heavily on productivity – the more worker ‘bang’ you can get for your wage ‘buck’, the more efficient is your firm. Facing such imperatives, it is far more difficult for these firms to pay excessive premiums for more educated workers when that education does not translate into higher productivity.

The empirics support this reasoning – in the African countries studied, worker characteristics (like education and experience) in firms engaged in international trade are rewarded far more in line with productivity levels. In firms exposed to less competition, these relationships unwind.

Of course this does not imply that in those economies where there is greater international competition, unemployment is non-existent and everyone is prosperous. It also does not mean that these countries have noticeably more flexible labour market arrangements. But it does mean that competition disciplines firm behaviour in beneficial ways, and reduces the impact that paying excessive premiums has on aggregate demand for labour.

Fostering greater and fairer competition in an economy is no easy task. It requires a judicious mix of a range of policies, some of which are politically difficult. This includes allowing greater international competition in product markets, and greater flexibility and competition in labour markets. But it seems that unless this road is not travelled, the wedge between wages and productivity in Africa and the associated negative impacts on the demand for labour will persist.

15 Oct 2005