The Case for Globalisation: Reforming the Non-tradable Sector

September 23, 2019        Reading Time: 6 minutes

The-Case-for-Globalisation---Reforming-the-Non-tradable-Sector---Shanta-Devarajan Reading Time: 6 min read

Image Credits: Steven Wei/Unsplash

*Shanta Devarajan

For developing countries as a whole, globalisation—the process of lowering trade barriers and integrating with the world economy—has been enormously beneficial, as seen in figure 1.1

Figure 1: GDP Growth Rates before and after Trade Liberalisation

Note: Red line=average growth rate before (after) trade liberalisation, Blue line=growth rate (3-year MA), Dot=growth rate. Source: Wacziarg and Horn Welch (2008).

GDP growth rates have been about two percentage points higher after trade liberalisation. Investment-to-GDP rates have been almost 10 percentage points higher—and sustained for a long time after liberalisation. Moreover, this higher growth has contributed to faster poverty reduction2 in globalising countries. And there is no systematic relationship between trade liberalisation and inequality. In some globalising countries, inequality rose, while in others it fell.

Why then does globalisation elicit so much criticism from NGOs3 and academics 4, among others? One reason is that the average growth rates hide a huge variation among individual countries. Among major Latin American countries, for instance, the only country that saw a significant increase in its growth rate post-liberalisation was Chile; Brazil and Mexico experienced a decline in their growth rates. Similarly, in Africa, with the exception of Ghana, trade liberalisation was accompanied by a decline in average growth rates in many countries. It is only in Asia where most countries have seen an increase in growth rates after liberalisation. This includes not just the celebrated cases of India and China, but smaller countries such as Bangladesh, Sri Lanka, and the Philippines.

Furthermore, in some of the countries where the growth impact was weak, the employment effects were even more troubling. In Brazil 5, the regions facing tariff cuts experienced significant drops in formal-sector employment and earnings. These effects became more pronounced 20 years after liberalisation.

Finally, for trade liberalisation to have its intended effect, a host of other factors have to be in place. Africa still has a huge infrastructure6 deficit, which means that even if there is trade reform, it remains difficult to ship manufactured goods to ports. India has seen very little growth in manufacturing employment—even though it has a large number of low-skilled workers. The level of education of these people is woefully poor; the share of second-graders in rural public schools who could not read a single word was 80%.7

Do these criticisms imply that globalisation has gone too far? On the contrary, they suggest that it has not gone far enough. For the benefits of trade liberalisation were not simply the efficiency gains from removing a set of tariff distortions in the economy. Simulations with computable general equilibrium (CGE) models showed that, if this were the only effect, the benefits from trade reform would be miniscule.8 Trade restrictions did more than add a distortion to a competitive economy. In many cases, they created domestic monopolies that could exercise their monopoly power behind trade protection. Some of these monopolists were also politically connected, which may explain the resistance to trade liberalisation in many countries. When the presence of these monopolies is incorporated into a CGE model, the beneficial effects of trade liberalisation become much greater.9 The reason is that trade liberalisation subjects these monopolists to foreign competition, breaking down their monopoly power, lowering domestic prices much more (thereby making it cheaper for those who buy these goods), and permitting the exploitation of economies of scale.

But trade liberalisation only affected monopoly power in the tradable sector—manufacturing and agriculture. It did nothing to break down the monopolies in the non-tradable sector—services such as finance, transport, and distribution. To this day, the services sector remains largely unreformed.10 Yet, services such as finance, transport, distribution, and business services are necessary inputs11 into the production of exports, accounting for about 30-40% of value added in exports. If these non-tradable services remain monopolised, then it is difficult for the tradable sector to expand in the wake of trade liberalisation.

That this is not just a theoretical possibility, I will illustrate with three specific examples.

1. Cronyism in Tunisia – Tunisia undertook major trade reforms in the 1990s but export growth remained anemic. This is surprising given Tunisia’s proximity to Europe, fairly good infrastructure, and an educated population. During this same period, the family of the then President, Ben Ali, had interests in certain enterprises. The sectors where these enterprises were situated received protection from both domestic and foreign competition. These sectors included telecoms, transport, and banking, and the prices of these services were artificially high (Tunisia had the third highest telecoms prices in the world). Since these services were required to export, Tunisian exports were not competitive in world markets. The monopoly power enjoyed by these firms can be seen in the distribution of profits: the ‘Ben Ali firms’ relative to the rest of the economy accounted for 0.8% of employment, 3% of output—and 21% of profits.12

2. Roads in Africa – As stated above, Africa’s infrastructure deficit stands in the way of harnessing the gains from trade liberalisation, but a study13 of the major road transport corridors in Africa revealed that vehicle operating costs along these four corridors were no higher than in France. What was higher in Africa were the transport prices—these prices were, in fact, the highest in the world. The difference between transport prices and vehicle operating costs is the profit margin accruing to the trucking companies. These margins were of the order of 100%. This was made possible by regulations in the books in almost every African country that prohibits entry into the trucking industry. These regulations were introduced a half-century ago, when trucking was thought to be a natural monopoly. Today, there is no need for such regulation, but there are huge trucking monopolies in every country that lobby against deregulation. The fact that relatives of the ruling family own the trucking company doesn’t help. Africa’s high transport prices are due to monopoly power in the (nontradable) transport sector, which is in turn standing in the way of the continent’s benefiting from trade liberalisation.

3. Teachers in India – Many second-graders in rural public schools in India can’t read, and this attributed to the fact that for about a quarter of the time, teachers are absent.14 This absence is because teachers run the campaigns of the local politicians. When the politician gets elected, he turns around and gives the teacher a job for which he doesn’t need to show up. The result is that teachers, being providers of non-tradable services, have a small degree of monopoly power that enables them to be absent without major sanctions.

In sum, trade liberalisation has not fully delivered on the promise is because only tradable sectors have been subject to international competition. If this competition can be spread to the non-tradable sectors, there will be greater competition in those sectors and bigger gains from trade liberalisation. Hence, the problem with globalisation is not that it has gone too far; it’s that it hasn’t gone far enough.


1Wacziarg, R and Welch, K. (2008). Trade Liberalisation and Growth: New Evidence. [Online]. World Bank Economic Review. 22(2): 187-231. Available at: [Accessed 05 September 2019].

2Dollar, D and Kraay, A. (2004). Trade, Growth and Poverty. [Online] The Economic Journal. 114(493): 22-49. Available at:
[Accessed 05 September 2019].

3Guttal, S. (2007). Globalisation. [Online] Development in Practice. 17(4-5): 523-531. Available at: [Accessed 05 September 2019].

4Stiglitz, J. (2003). Globalisation and its Discontents. [Online] Available at:
[Accessed 05 September 2019].

5Dix- Carneiro, R and Kovak, B. (2017). Trade Liberalisation and Regional Dynamics. [Online] American Economic Review. 107(10): 2908-2946. Available at: [Accessed 05 September 2019].

6African Economic Outlook 2018. Africa’s infrastructure: great potential but little impact on inclusive growth. (2018). [Online] Available at: [Accessed 05 September 2019].

7The World Bank. Learning to realize education’s promise. (2018) World Development Report 2018. [Online] Available at:
[Accessed 05 September 2019].

8Ackerman, F & Gallagher, K. P. (2008). The Shrinking Gains from Global Trade Liberalisation in Computable General Equilibrium Models. [Online] International Journal of Political Economy. 37(1): 50-77. Available at: [Accessed 05 September 2019].

9Devarajan, S & Rodrik, D. (1991). Pro- competitive effects of trade reform: Results from a CGE model of Cameroon. [Online] European Economic Review. 35(5): 1157-1184. Available at: [Accessed 05 September 2019].

10Borchert, I et al. (2012). Policy barriers to international trade in services: evidence from a new database. [Online]. The World Bank. Available at: [Accessed 05 September 2019].

11Miroudot, S & Cadestin, C. (2017). Services in Global Value Chains. OECD Trade Policy Papers. No.208. [Online] Available at: [Accessed 05 September 2019].

12Freund, C. & others. (2014). All in the family: state capture in Tunisia. [Online]. The World Bank. Available at: [Accessed 05 September 2019].

13Teravaninthorn, S & Raballand, G. (2009). Transport Prices and Costs in Africa: A Review of the International Corridors. [Online]. World Bank Group. Available at: [Accessed 05 September 2019].

14Chaudhry, N. et al. (2006). Missing in Action: Teacher and Health Worker Absence in Developing Countries. [Online] American Economic Association. 20(1): 91-116. Available at: [Accessed 05 September 2019].

*Dr. Shanta Devarajan is a Professor of Practice at Georgetown University’s Walsh School of Foreign Service, Washington DC, and the former Senior Director for Development Economics at the World Bank. The opinions expressed in this article are the author’s own and not the institutional views of LKI, and do not necessarily reflect the position of any other institution or individual with which the author is affiliated. This post first appeared on the Brookings blog, Future Development.

Leave a Reply

Your email address will not be published. Required fields are marked *

Untitled Document