I’ve now seen the same annoying elementary (but quite common) mistake twice in two days, and I’d like to knock it on the head before it gets repeated.

According to a blog post yesterday by Malaka Gharib at ONE, Daniel Yohannes, the CEO of the U.S. government’s Millennium Challenge Corporation, believes this:

If Africa captured captured only a small percentage of global trade, it would make a big difference. In fact, in 2008, 1 percent of global trade was worth $195 billion, more than five times the development assistance sub-Saharan Africa received that same year.

And Ali Hewson and her husband Bono use a similar statistic in this short video (which Chris Blattman posted yesterday):

I support the basic point that is being made: trade is good for developing countries.  I can’t recall ever meeting anyone from Africa who would not prefer more trade to aid.   I agree with Bill Easterly’s argument in the FT that rich countries should be more focused on reducing trade barriers.

But the video and remarks by Mr Yohannes grossly overstate the benefits of exports.

Before we get on to the economics, let’s try to get the numbers right.

The total value of world trade is about $16,000 billion a year (measured as the total value of exports of goods and services).   Exports from sub-Saharan Africa are about $380 billion a year, so they constitute a bit more than 2 percent of world trade, as Ms Hewson rightly says.  And aid to sub-Saharan Africa from OECD countries is about $40 billion a year, roughly as Mr Yohannes implies.

So that means:

  • If Africa’s share of world trade grew by one percentage point, its export earnings would grow by about $160 billion a year.  So the figures quoted by Ms Hewson’s $(70 billion) and Mr Yohannes ($195 billion) are in the right ballpark.  (They may get different numbers because they are using different years for their estimates of world trade.)
  • One percent of the value of world trade ($160 billion) is about four times aid to sub Saharan Africa ($40 billion).  On this, Mr Yohannes is roughly right (he says five times, but he is using older figures).
  • If exports from sub-Saharan Africa increased by 1%, as Bono puts it in the video, the increase in Africa’s export earnings would be about $4 billion a year. This is only about a tenth of aid to Africa.  For Africa to secure a one percentage point increase in world trade would require a 40% increase in Africa’s exports.

So it’s true that if sub-Saharan Africa could increase its share of world trade by 1 percentage point, its export earnings would grow by about $160 billion a year, and that’s about four times what it gets in aid.

But the net benefit to Africa of increasing its exports is not the same as the increase in its export earnings.  That is the same mistake as equating a firm’s turnover with its profit.

Put another way, Mr and Ms Hewson and Mr Yohannes are all making the fundamental error of mercantilism, which is the idea that the prosperity of a nation is increased by the accumulation of bullion obtained from overseas.

They seem to have forgotten that exports have a cost.  You have to put labour, land, energy and other inputs into making the thing you are going to export – inputs which could be used for something else.  You have to grow tea, dig diamonds out of the ground, or make the shoes or silicon chips.   You may have to transport the goods. There may be workers on the factory line, and the firm will need administrative staff, drivers and security guards.  You may have to make or buy machines, tractors or lorries.  All this is a cost.  Even oil has costs of exploration, extraction, refining, transportation and damage to the environment.

These may well be costs worth bearing.  When you export those goods and services, you get foreign currency in return. This foreign currency enables you to buy more imports.  (That’s the point of exports.) But they are costs nonetheless.

Nations benefit from trade in other ways too, in addition to the extra imports that they can buy.  Trading nations tend learn from abroad; their firms are forced to be more competitive and so more productive; and the imports can be of machines or technologies that help the economy to be more productive.   So being an open, trading nation may help a country to grow faster.

So if the value of increasing exports is not the increase in export earnings, what is it?

The value to a firm of a new contract is not the price on the contract, its the profit it will make meeting it.  In the same way, the value to Africa of exports is the difference between the cost  of making what it has exported, and the value of what it can import as a result.

Of course, we don’t really know what those exports cost, or what the imports are really worth, so we don’t know the true net benefit to Africa of increased exports.  But the value added is more likely to be about 10% of the turnover than 100% of it.  If that’s a reasonable estimate, an increase in Africa’s share of world trade by one percentage point would be worth $16 billion a year.  This is, as it happens, less than half of what it currently receives each year in aid.

And remember that increasing Africa’s share of world trade by one percentage point would be no mean feat – it would require a 40 percent increase in Africa’s exports.

It is true that nations benefit from exports (or, to be more precise, they benefit from the imports that exports enable them to buy).  But you can’t measure the benefit to a nation of exports by the value of its export earnings, any more than you can measure the profitability of a firm by its turnover.  Getting this right is important because, as Paul Krugman pointed out in Peddling Prosperity, if you lose sight of the fundamental economics you get drawn into stupid policy prescriptions such as export subsidies or import tariffs.  For example, if you make it a policy goal to increase export earnings, you may subsidise exports, and so start to export goods that cost the nation more to make than someone abroad is willing to pay for them.  Though export earnings have gone up, the nation is worse off, not better off, because the exports have cost the nation more than it earns from them.

Exports are good for development. There are almost no examples of successful development that have not been built on export-led growth.  We should support reductions in trade barriers; and we should support investments that help developing countries to be more productive and so export more at lower cost to themselves.  Developing nations would rather have trade than aid, and most taxpayers in donor nations would agree.

But let’s not overstate the case by claiming that the net benefit of exports is the same as the value of turnover.  Even if Africa could increase its exports by 40%, which is a lot, and so achieve the (more modest-sounding) increase of 1 percentage point of world trade that Mr and Mrs Hewson, Mr Yohannes, and indeed all right-thinking people would like to see, the benefit to the continent could still be quite a bit less than it presently receives each year in aid.

If you enjoyed this post, please consider leaving a comment below, and perhaps sharing this with other people using the buttons on the left. You can also sign up to have blog posts sent to you by email.

9 Responses to What is the value of exports?

  • Too often, policy makers and politicians rest back with the neo-liberal. neo-classical economics view of trade being and important part of gorwth.You’ve hit the nail right on the head on the actual degree of importance of exports. Africa is most laden with agricultural exports whihc naturally don’t always mean development which price fluctuations. In my post http://ipeanddevelopment.wordpress.com/2010/08/16/ipe-id-first-clash-agricultural-trade/ I bring up the arguments by H-J Chang and some others who see the value of protectionism instead of rapid global reduction in trade barriers. Chang also makes a highly controversial (rarely used) argument that agriculutural protection in theWest benefits Africa’s consumers—and thus should be kept?

  • Yeah, it also implies that the benefits of private enterprise trickle down to those at the bottom in a more efficient way than aid does. I’m not sure there is really evidence of this being true – indeed, in most circumstances exports end up making a very small number of people very rich and bring only slight benefits in income to everyone else involved. We’re constantly told that Trade is Better than Aid, usually by those who are in a strong position to benefit from said trade. Clearly that paradigm is broken – it depends on exactly what kind of aid and what kind of trade you are talking about.

  • You’re right it’s a gross error to crudely compare export revenues with aid flows, as is routinely done.

    I guess from a broader aid versus trade perspective, the argument hinges on the existence (or not) of positive externalities from exporting, the contrast (aid versus trade) in the incentives politicians face and so forth. I’d imagine the consensus view is that a dollar of export earnings (i.e. profit) is better than a dollar of aid because of beneficial externalities better incentives etc. [I’m not saying the consensus if correct: too hard a question for me].

    My impression was that there’s reasonably good empirical support for positive ‘learning’ externalities from export, and also stronger empirical support for the idea that trade benefits growth than for the idea that aid benefits growth.

    I haven’t really looked around much, but one recent thing I saw on exports and growth is here and here’s one on trade and learning.

    Owen replies: Yes indeed, there do appear to be positive externalities from trade. (I’m definitely in favour of trade.) But it isn’t very clear if those are externalities mainly from exports or from imports. They seem as likely to be the latter as the former. If so, you get those benefits from aid (all of which is spent on imports) as you do from exports.

  • Nice post. I agree people are overstating the case about export driven growth. But aren’t you somewhat understanding the real value at $16 billion. The costs generating those profits are generally good developmental costs, as you say, jobs, infrastructure, know how etc. Don’t you need to value them as well beyond the $16 billion.

    Second, it always seems to me to be a mistake to talk about “Africa” as one place. Some countries could benefit very quickly and be in position to benefit from trade. But there are many backwaters that don’t yet have the capacity to get much from exports. So even if you could increase African exports by 1 percent the distribution by country would be very skewed.

    Finally, I would love to hear your thoughts on the whole protectionism as a route to trade-led development. There’s lots of debate about how much protectionism/subsidies South Korean used in growing its economy. The comment about AG subsidies protecting African consumers seems contrary to logic, but interesting assertion.

  • @David

    Work is a cost, not a benefit. The benefit is the income you get from it.

    If a worker has a job making stuff for export, this is a cost to the economy, because that person’s time and energy is tied up producing things that nobody in that country will consume. The cost to the economy is the value of what that person could be doing instead (eg farming, making things for domestic consumption). The benefit to the economy is export earnings. The net benefit to the economy is therefore the export earnings MINUS the value of what the person could otherwise have produced. So export jobs are a cost to the economy, not a benefit, to be set against the export earnings.

    Protectionism requires a longer post. In short, my feeling is that I understand the theoretical case for infant industry protection in markets characterised by large returns to scale; but I am more nervous of government failures than I am of market failures in this particular case.

  • True, work is a cost if you have something close to full employment.

    But Africa has huges numbers of young people who are under employed, or unemployed… or in the informal economy. The whole argument for business-led development, of which trade is one part, is job creation. Your treating a job as purely a cost seems odd to me. Assuming the export led job creation is not a zero-sum (or negative sum) game (just switching from a domestic economic activity to export activity of equal or lesser value to the economy), then shouldn’t you consider the additional goods to the economy, such as the income from the new job that is being spent domestically (helping domestic formal and informal businesses), and the e collection of taxes that can support public services.

    I may be using the terms cost and benefit incorrectly, but the good to the economy of private sector job creation from exports is more than just the export earnings minus the value of what the person otherwise could have produced — in the example above, $16 billion. There are other tangible and intangible goods that should also be considered. That’s my only point. Am I missing something here?

    David – yes, I regret that you are trying to double count the benefits by including both export earnings and jobs. The benefit to the economy of exports is the additional imports that the resulting export earnings enable the citizens to buy. There are no other economic benefits. The costs to the economy include the inputs that go into those exports – including land, labour, energy, and other natural resources. The net benefit is the difference between those benefits and those costs.

    A job is a bundle which includes a benefit (the worker gets an income and perhaps some job satisfaction) and a cost (the worker has to give up their time, which they could be using for something else). For anyone in a job, the benefits to them presumably exceed the costs, otherwise they wouldn’t be doing it.

    People employed making exports are clearly better off with a job than without it. They are better off because they get an income from it, and that is presumably more than they would doing something else or doing nothing. The net benefit to them of the job is the income they receive, minus the value of what they would have been doing otherwise (either some employment of less value, some smallholder farming, or in rare cases, leisure). In other words, the net benefit of each job is the income less the opportunity cost of the labour.

    The reason you are double counting the net benefits of jobs is that the incomes paid to export workers are paid for from the export earnings. So you can’t count the net benefit of a job as additional to the export earnings. The export earnings describe the entire gross benefit to the nation of the activities, including the benefits to workers who have the jobs.

    So when you aggregate across the economy as a whole to calculate the impact of exports, the net benefit is the export earnings minus all the costs, which include the use of land, natural resources, energy and the opportunity cost of the labour. You can’t add “jobs” as a separate item in this reckoning, because the net benefits of each job are already included: the benefit of the each job – namely the income it provides – is included in the export earnings.

    Your point about unemployment suggests that you think that the opportunity cost of labour may be negligibly small. If those people would otherwise be doing nothing, you’d be right. (It wouldn’t be zero, because most people prefer leisure to work, so there is some cost to working even for people who would otherwise be unemployed). But I think you would be wrong to assume that people working in export industries in developing countries would otherwise be unemployed. The people who get these jobs are generally smart and capable and they would be doing something else. In most cases, in the absence of a welfare state, they would have to be doing something else to survive. So I don’t think you can assume the opportunity cost of their labour is zero. The benefit of the export job is therefore the difference between what they earn doing this, compared to the value of what they would otherwise be doing, and that net benefit might well be quite small – certainly much less than the total value of the exports.

Follow Owen




Email subscription

I want to receive new posts by email

Owen on Twitter
Owen on Google Plus