Making policies work

GREAT insights Magazine

Trade liberalisation and poverty: Did we learn anything in the last ten years?

December 2015

Winters, L.A. 2015. Trade liberalisation and poverty: Did we learn anything in the last ten years? GREAT Insights Magazine, Volume 4, Issue 6. December 2015/January 2016.

Recent research reinforces the evidence that trade liberalisation boosts growth but potentially hits people in import-competing sectors. Labour mobility is key to sharing the gains from liberalisation equitably. Women seem to gain more than men from trade liberalisation.

In the mid-1990s I started to worry about how much we actually knew, rather than just thought we knew, about the effects of trade liberalisation on extreme poverty in developing countries.  In the early 2000s, my colleagues and I concluded that there is no simple general conclusion about the relationship (Winters et al., 2004). There is a strong presumption that trade liberalisation will be poverty-reducing in the long run and on average through its effects on the level of national income, but there is no guarantee that the static and microeconomic effects will always be beneficial to the poor. Trade liberalisation will almost inevitably hurt some people (at least in the short term), and some of these may be poor. Since 2000, ‘trade and poverty’ has become a bit of an industry. I ask here whether a decade of research has taught us anything new? (Winters and Martuscelli, 2014). The answer is that we haven’t learned anything fundamentally new, but that we now know a bit more about the mechanisms and can be a bit more confident about some of the earlier conclusions.

Trade liberalisation is good for growth

The literature of the past decade reinforces the presumption that trade liberalisation generally raises average incomes (i.e., at least temporarily boosts economic growth). The challenge has always been to show that liberalisation causes growth rather than vice versa, but at least some studies have done this to a reasonable degree of confidence. Recent research also suggests that the growth effect varies with a series of complementary conditions; among these the quality of institutions appears to be important, as does the ease with which factors of production can migrate between sectors, which, in turn, depends on conditions such as labour market flexibility, education levels and the ease of firm entry and exit. The conditions suggest that the growth effect of trade reform will be stronger for richer countries and that poorer ones may even lose. The latter conclusion is, however, as yet unproven and investigating it more fully seems to be a priority for future research. For example, the statistical association may just reflect Africa’s poor growth performance when the Bretton Woods organisations were encouraging them to reduce trade barriers, and the analysis depends at present on a number of strong and untested assumptions.

Ask how the poor earn their living

The new literature confirms that the effects of trade liberalisation at a household level cannot be guaranteed to be benign: they depend especially on the nature of the trade policy that is reformed and on how the poor earn their living. The poor spend a large proportion of income on food but many also depend heavily on agriculture for their livelihoods. Either way, reforming agricultural trade is potentially important for poverty: roughly speaking, if liberalisation raises the price of a food product, the rural poor will tend to benefit but the urban poor will suffer and vice versa for price falls. In many countries, the urban poor are the more numerous, so one should not conclude that driving down farm prices will always be regressive.

Similarly, trade restrictions may permit a host of small inefficient companies to remain in existence in particular manufacturing sectors; when the restrictions are removed there are general gains for consumers of the goods concerned as imports drive down prices, but the workers in those firms may still lose overall. Whether they do or not will depend critically on what alternative sources of income they can find.

Mobility is critical

This last point illustrates one of the issues that recent research has really brought to prominence. Factor, and especially labour, mobility is key not only to reaping aggregate gains from increasing international trade but also to sharing them reasonably equally and thus to the probable poverty impacts of trade reform. If labour cannot move between sectors or between locations, the effects of trade liberalisations are restricted to the places in which they have their initial impact and are potentially very large: if firms cannot bring more workers in, workers in the export sector will be able to extract much higher wages and if workers in import-competing firms have no other sectors to move to, they are likely to face significant wage cuts. That is, if mobility is low, the effects are deep but narrow. If, on the other hand, successful firms can get more workers and failing ones shed them to other sectors, the wage effects of shocks are smaller in size but greater in extent. That is, they are more equally shared. Several recent studies have shown that large positive or negative effects of trade shocks correlate strongly with the absence of mobility. I conclude from this that a major policy objective for governments seeking to modernise their economies through trade reform (or any other means) should be to break down the barriers to mobility.

Two new results

Recent research has thrown up at least two other interesting conclusions. First, trade liberalisation may increase wage inequality. Recent advances in trade theory at the firm level suggest that liberalisation will disproportionately benefit relative well-performing firms. Good firms are likely to have better workers – they may pay more and they may be better at selecting workers – so if these firms benefit most, better (and hence better paid) workers will benefit more than weaker ones. However, very few studies show weaker workers actually lose – they just gain less from liberalisation than do better workers. Moreover, remember that wage inequality is not the same as household inequality because most households have several sources of income.

The second recent result is that women seem to gain relatively more from trade liberalisation than men; this is mainly because international trade seems to require relatively more brain than brawn than do traditional activities so that increasing trade switches demand towards female workers.


We have come to understand the connection between international trade and policy alleviation better over the last decade. The basic story has not changed: there is a strong presumption that liberalisation will stimulate growth which will boost the incomes of most people. And at a micro-economic level, trade liberalisation will probably hurt some people, and some of these may be poor; but its effects are relatively straightforward to predict, depending, for example, on how the poor earn their living and exactly what trade policies are reformed. Thus if we garner the information required, liberalisation may be tailored to avoid the worst poverty impacts and may be accompanied by compensatory measures. What has become clearer over the last decade is that fostering labour mobility between sectors and over space will allow the benefits of trade liberalisation to be shared more equally and so help to turn into reality the theorist’s claim that everybody could gain in the long run.

About the author

l alan winters

L. Alan Winters is Professor of Economics, University of Sussex.


Winters, L.A. and A. Martuscelli, 2014. Trade Liberalization and Poverty: What Have We Learned in a Decade? Annual Review of Resource Economics, Vol. 6: 493-512. And CEPR Discussion Paper No. 9947.

Winters L.A, N. McCulloch and A. McKay 2004. Trade Liberalization and Poverty: The Evidence so far. Journal of Economic Literature, vol. 42: 72-115.

Photo: The Volkswagen South Africa plant in Uitenhage is the largest vehicle factory in Africa. Credits: Volkswagen South Africa, Media Club South Africa.

This article was published in GREAT Insights Volume 4, Issue 6 (December 2015/January 2016).

Share Button
Economic Transformation and TradePovertyTrade

External authors

L. Alan Winters