Are developing countries engaging in “social dumping”?

Guillermo de la Dehesa

24 May 2007



Increasingly, the charge of “social dumping” is heard as a rational for protectionist measures against developing country exports. Many businessmen, labour unions and politicians in developed nations believe that lax working regulations and conditions, as well as weak political and social rights, provide developing-country exports with an unfair advantage. To counter this, they argue that either developing-country exports of cheap manufactured goods should be subject to antidumping duties to eliminate the “unfair” competitive factors, or these countries should impose higher labour standards.

These beliefs are wrong on several counts.

Is it dumping when low wages compensate low productivity?

First, according to basic international trade theory, each country should specialise in what it can produce competitively, exploiting its comparative advantages. Developing countries have lower wages due to their lower labour productivity and relative scarcity of physical and human capital. Their productivity disadvantage, however, is not spread evenly across all sectors. In labour-intensive sectors, their lower wages more than compensate for their lower productivity, so they tend to specialise in goods that are intensive in their use of low-skilled labour, cheap land, or abundant natural and energy resources. In other sectors, the higher wages of developed countries are more than compensated by their high productivity which stems from their abundant physical capital, technology and innovation; in these sectors it is the developing countries that cannot compete with the developed nation exporters. Moreover, both nations gain from focusing on what they do best and trading for the rest – a proposition that holds true regardless of the source of the comparative advantages.

Is it dumping when you pay higher wages in export sectors?

Second, “social dumping” is not dumping. Under WTO rules, dumping means the country in question is exporting at a price below its domestic market price or below its production cost. This is exactly the contrary of what it is happening in developing country exports to developed countries. In these countries, wages in labour-intensive industries producing for exports tend to be higher than in those firms producing only for their domestic market and, as a consequence, their export products tend to have higher costs than similar products sold domestically. Many, if not most, exporting firms in developing countries are subsidiaries or affiliates of developed country multinationals or local firms that have outsourcing or off-shoring contracts with them. Robert Lipsey has marshalled conclusive empirical evidence that foreign-owned companies pay higher wages in both developing and developed countries.1 Thus, only when exports from these foreign and domestic companies in developing countries were found to be subsidised by governments should a countervailing duty be justified.

Does imposing labour standards help poor nation workers?

Third, activists have for decades pressed for enforcement of “fair labour standards” to “level the playing field.” While I believe it is an undeniable moral fact that everybody should embrace the goal of improving the wages and working conditions of workers in developing countries, it is also an undeniable economic fact that forcing these labour standards on developing countries may only make things worse for their workers while failing to have the desired effect of keeping more jobs in developed countries. Drusilla Brown, Alan Deardorff and Robert Stern have shown that, beside good intentions, there is no obvious case on welfare grounds for pursuing universal standards or even the harmonisation of standards on labour, the effect of which will most probably be negative for the very people that are in the greatest need for assistance.2

One common argument suggests that low-labour-standard countries reduce employment in rich nations via direct-export competition and the offshoring of jobs. This is refuted by empirical research. Mita Aggarwal has demonstrated that labour standards are lower in non-tradable sectors than in exporting ones and that developed countries imports of low labour standard products from poor countries are a very small percentage of total imports from developing countries.3 On the FDI side, Dani Rodrik has shown that low labour standards are a hindrance rather than an attraction, for foreign investors.4

Imposing higher labour standards (higher minimum wages, higher severance payments, lower number of working hours, very high labour security etc.) on poor countries does not improve wages and working conditions, just the opposite. By definition, raising the cost of labour above its level of productivity is an oxymoron; workers tend to become unemployed when their wages are raised above the market value of their productivity. Furthermore, imports from developing countries have had a lower impact on employment and on widening of the skilled/unskilled wage differential than technical change in developed countries. A large body of empirical evidence establish both these findings not only for developing countries but also for developed countries.5 One case in point is the large increase in unemployment in East Germany that occurred with Western wages were extended to Eastern workers with lower productivity as shown by George Akerlof, Janet Yellen and Hans Hessenius.6 Will Martin and Keith Maskus also show that, contrary to the argument that weak labour standards provide an illegitimate boost to export competitiveness and a “race to the bottom,” low labour standards tend to reduce competitiveness when they are the result of workers discrimination, abuse of market power etc. in the exporting country.7 Competitiveness only increases in the medium- to long-run with productivity improvements.

Policy to improve the situation of poor nation workers

Protection based on charges of “social dumping” is not the right way to pursue the undeniable moral fact mentioned above. Poor country workers do not benefit from the imposition of higher labour standards but rather from internal economic and social development. There are better ways to try to improve worker rights and labour standards in poor countries than to impose them, while protecting workers in rich countries, shown by Robert Stern alone and with his co-author Katherine Terrell.8

The proper response in rich countries to the challenges posed by the rapid growth of developing country manufactured exports is more public sector assistance for workers who must transition from jobs in declining sectors into growing sectors. In poor countries, there is a need for multinational corporations to continue to develop internal codes of conduct to improve labour standards and working conditions in their affiliates and subcontractors in host developing countries. There is also a growing need to target foreign aid more efficiently through NGOs working on the spot in developing countries, on health, education and basic infrastructures.

Another way forward is to increase awareness of poor labour practice among consumers in rich countries. These consumers have already proved willing to pay higher prices for goods that are certified and labelled as having been manufactured by firms that applied higher labour standards. The same willingness is seen in the rapid spread of consumer-level demands for “fair trade” distribution outlets for such kind of goods.

Would social dumping protection help the child labour problem?

Fourth, “child labour” is often used as an argument in the “social dumping” arguments. Child labour should be fought on moral, ethical and human rights grounds, which are fully understandable and shared by everybody. The association with social dumping – for example, demands for an import ban on all manufactures produced using child labour – may make matters worse for the children. Restrictions on child labour are already embedded in the International Labour Organization (ILO) Convention 138, which mandates countries to specify a minimum age of employment for children (not less than 15 years) so child labour below that age is considered illegal. To date, 141 countries have signed this ILO convention and some others, which have not signed the convention (included the US) effectively outlaw child labour as well in their national labour regulations.

In spite of the almost universal presence of legal bans, child labour continues to thrive The only countries which have advanced in banning child labour are those where the adult unskilled workers face labour market competition from children but, at the same time,9 do not rely on child labour income in their own family. A large majority of countries introduced the ban because it was politically advantageous, but enforcement may well be ineffective.

The empirical evidence contradicts the belief that child labour is important in promoting exports. Most empirical evidence, for example, by Eric Edmonds and Nina Pavcnik, shows a positive correlation between more international trade and less child labour.10 Moreover, Eric Edmonds shows that only 1 in every 200,000 economic active children in the world, aged 10 to 14, are employed and only two countries in the world go above 1 in very 10,000.11 Of that tiny fraction only 12% of total working children in developing countries are employed in exporting companies and they have higher wages than those 88% involved in domestic production. Banning child labour would only have a net positive income effect in nations where three conditions apply. One, if the reduction in the supply of working children would imply a similar increase in the demand for adult workers (which are natural substitutes); two, as Kaushik Basu and Pham Hoan Van show, if this larger demand of adult workers would increase their wage levels compensating totally the loss of children wages, and three, if adult workers, now employed, would use their higher wages to send their children to school.12 In these three conditions are not fulfilled, such a prohibition would make working children worse-off, total welfare would deteriorate and eventually working children could end up having working children through their “family poverty trap” impact on lower level of education. The main factor forcing child labour in those countries is that they live in extreme poverty, where “a low paid job is better than no job at all”, as Paul Krugman correctly pointed out.13 Their parents, in order to survive, need their children’s’ wages to complement their own meager incomes.

Better ways to combat child labour

There are better ways to deter child labour than an outright banning of children working for exports, which historical experience, both in Europe between 1880 and 1910 and in developing countries now, shows do not produce a decline of child labour but only depress wages and may cause more children to work. One is letting human rights-conscious consumers in developed countries to purchase products labelled as “child labour free” at a premium to reduce the incentives to employ children. Another is lowering the cost of schooling in developing countries to attract more children to school, either through conditional cash transfers to their families for making their children to attend the school or through lower education fees for poorest families sending their children to schools. PETI and Bolsa Escola in Brazil, Mid-day meals in India or Progresa in Mexico are effective programs to discourage child labour.

Finally, the last instance and real solution to this problem is not banning manufactures produced by children, but to reducing drastically the protectionists laws and practices by developed countries (and some emerging countries) against agricultural and labour-intensive manufactures produced by developing countries, which are exactly those goods they can export more competitively to the rest of the world and which impede their parents from finding a reasonable job in the export business and from being able to send their children to the school instead of forcing them to take a job.


1 “Home and host country effects of FDI,” Robert Lipsey, NBER Working Paper 9293, 2002.
 2“Child labor: theory, evidence and policy,” Drusilla Brown, Alan Deardoff and Robert Stern, Discussion Paper 474, School of Public Affairs, University of Michigan, Ann Arbor, 2001).
  3“International trade, labour standards and labour market conditions: an evaluation of the linkages”, Mita Aggarwal, USITC Office of Economics Working Paper 95-06-C, 1995.
  4“Labour standards and international trade: do they matter and what do we do about them?” Dani Rodrik, in Robert Lawrence, Dani Rodrik and John Whalley (eds.) “Emerging agenda for global trade: high stakes for developing countries”, Washington D.C., Overseas Development Council, 1996.
  5For developing nations, see “The cost of job security regulations: evidence from Latin American labour markets”, NBER Working Paper 7773, 2000; “The consequence of doubling the minimum wage,” Industrial and labour Relations Review, Vol. 54, 2001. For developed nations see “Trade, jobs and wages”, Paul Krugman and Robert Lawrence, 7 NBER Working Paper 4478, 1994, and “International Trade and core labour standards: a survey of the recent literature”, Drusilla Brown, Labour Market and Social Policy Occasional Papers Nº 43 (October) Paris, OECD, 2000.
  6“East Germany in the cold”, Brookings Papers on Economic Activity nº 1, George Akelfoff, Janet Yellen and Hans Hessenius, 1991.
  7“Core labour standards and competitiveness: implications for global trade policy”, Will Martin and Keith E. Maskus, Review of International Economics vol. 9 Nº 2, 2001.
  8“Labour standards and trade agreements,” Robert Stern, Gerald Ford School of Public Policy Discussion Paper nº 496, University of Michigan, Ann Arbor, 2003, and “Labour standards and the World Trade Organization,” Robert Stern and Katherine Terrell, Gerald Ford School of Public Policy Discussion Paper Nº 499, University of Michigan, Ann Arbor, 2003.
  9“Origins and consequences of child labour restrictions: a macroeconomic perspective”, Matthias Doepke and Dirk Krüger, CEPR Discussion Paper 5953, 2006; also see “Child labour: cause, consequence and cure, with remarks on international standards,” Kaushik Basu, Journal of Economic Literature Vol. 37, Nº 3.1999.
10 “Does globalization increase child labour?” Evidence from Vietnam,” Eric Edmonds and Nona Pavcnik, NBER Working Paper 8760, 2002.
11 “Child labour,” Eric Edmonds, NBER Working Paper 12926, 2007.
12 See “The economics of child labour,” Kaushik Basu and Pham Hoan Van, American Economic Review, Vol. 88, Nº 3, 1998.
13 “The accidental theorist,” Paul Krugman, New York, W.W. Norton, 1998



Topics:  Development

Tags:  child labour, social dumping, labour standards

Chairman of CEPR, Member of the Group of Thirty, and author of "Europe at the Crossroads".