The relaunching of negotiations on environmental goods: Any breakthrough in sight?

Jaime de Melo, Mariana Vijil, 1 April 2014



The decade-long negotiations on reducing barriers to trade in environmental goods and environmental services at the WTO failed to make any progress (Frankel 2013, Balineau and Melo 2013). Against this stalemate at the WTO, in September 2012, 20 APEC members submitted a list of 54 products for which they would lower applied tariff rates to 5% or less by the end of 2015. And on 24 January 2014 at the Davos meetings, a group of 14 countries including many APEC members plus Costa Rica, the EU, Norway and Switzerland, committed to pursuing ‘global free trade’ for environmental goods starting from this APEC list. The joint statement reads that the group is to “…build on the ground-breaking commitment to reduce tariffs on the APEC list of environmental goods by the end of 2015 [….] to achieve global free trade in environmental goods”. This plurilateral deal “…would take effect once a critical mass of WTO members participates…. and we are committed to exploring a broad range of additional products” (Davos 2014).

Any step forward out of this decade-long impasse is to be welcomed and any reduction negotiated by this group in the ambit of the WTO will be extended to other non-participating WTO members. But how much significance should we give to this initiative for reaching free trade in green goods? New research shows that for environmental goods, unless non-tariff barriers are included not much gain is to be expected. For environmental services, developing countries who would be the greatest beneficiaries of liberalisation because of complementarities with reduction of barriers in environmental goods have committed to market access only in the context of trade-agreement negotiations, usually with a northern partner.

Non-tariff barriers to trade in environmental goods are more important than tariffs

Our research (Melo and Vijil 2014) shows that since the beginning of the Doha negotiations, the time profile for average applied tariffs by income group (see the group definitions in Table 1) drawn from product-level data for the so-called ‘WTO list’ of 411 environmental goods moved in parallel to that for non-environmental goods, showing a small decline in average applied tariffs across all income groups (except for the high-income group where tariffs were already very low at the start). However, no acceleration in the reduction of tariffs on environmental goods (or for other goods) was observed as the Doha negotiations proceeded. Second, the tariff group averages on the left-hand-side of Table 1 indicate that there is not ‘much left on the table’ to negotiate (even though it is the highest, the low-income group is only 7.3%; column 1). With half of world trade taking place among countries having signed a free-trade deal, average applied tariffs could be around half the values reported in the table, that is, except for a few tariff peaks, they are close to negligible for the high-income countries, precisely those which, until the Davos declaration, had been willing to engage in tariff-reduction negotiations. Uniform tariffs instead of the current structure that would leave welfare unchanged are higher than the average tariffs because of tariff dispersion, but they are under 10% even for the low-income group (column 2).

What about an extension of negotiations to reach the ‘critical mass’ of WTO members needed for a plurilateral deal? A comparison of applied and consolidated tariffs shows that a standstill compromise that would consolidate tariffs at the applied rates would have little effect for the high-income group, but the middle-income groups would have to close a gap of 15 and 14 percentage points while the low-income group would have to close a gap of 7 percentage points. While this would not be a breakthrough, it could still be considered a step forward.

Table 1. Overall protection by income group (core list)

Source: Melo and Vijil (2004), Tables 1 and 2; Kee et al. (2008, 2009).
Notes: Environmental goods are defined following the core list of 26 products. Average tariffs and average tariffs + non-tariff barriers are import-weighted. Imports are mean values for 2010-11. Average values for each income group. Non-tariff barriers calculations are from estimates in Kee et al. (2008, 2009). Estimates in column 4 are for a 50% reduction in tariffs. Number of countries by income group in parenthesis. Income groups and abbreviation using 2011 GNI per capita, cut-offs in  parenthesis: high-income (>$12,476), upper-middle Income ($4,036-$12,475), lower-middle income ($1,026-$4,035), and low-income countries (<$1,025). 
The Trade Restrictiveness Index (TRI) is the uniform tariff which, if applied to imports instead of the current structure, would leave welfare unchanged. If tariffs of barriers were uniform and equal to the average values in cols 1,5,7 and 9 the TRI would have the same value as the average tariff.

During the Doha negotiations, developing countries refrained from submitting lists because they feared an invasion of imports from OECD countries. Their average tariffs are indeed higher, but by how much would imports increase if they participated in tariff reduction negotiations? Columns 3 and 4 give proximate answers based on import demand elasticities and applied tariffs. Since these first-order partial equilibrium estimates are just obtained as the sum across products and countries aggregated at the country-group level, these estimates for a 50% across-the-board reduction in tariffs can be read independently. Interestingly, average import price elasticities increase as one goes up the income group categories confirming that low-income countries have few domestic substitutes. The estimated percentage increase in imports in column 4 is the product of tariff heights and the import elasticities. The largest average increase is for the low-income group because the tariff height dominates the elasticity effect. Yet, the increase is less than 4%. Aggregated over the 21 countries in that group, the estimated increase in imports would amount to $42 million, hardly a flooding of imports (estimates with the WTO list of 411 products would still only result in an increase of $1.2 billion).

Measuring the equivalent of non-tariff barriers is difficult. Available estimates at the product level cover only 70 countries and are based on data collected in 2003-04. These barriers include price control measures, quantity restrictions, monopolistic measures and technical regulations, but they do not cover others like government procurement, burdensome custom procedures and local content requirements that are likely to be more important for environmental policies. More problematically, unlike tariffs, not all NTBs are welfare-reducing. However, in spite of the smaller sample and the above caveats, as for non-environmental goods, the ad valorem equivalents of non-tariff barriers suggest much higher barriers to trade for environmental goods than for non-environmental goods. For environmental goods, overall protection ranges from 6% for the high-income group to 45% for the low-income group. While the agenda for the negotiations is yet to be finalised, it is expected that non-tariff barriers will not be on the agenda as they are likely to be referred to the WTO’s Technical Barriers to Trade Committee (as was the case during the negotiations for the plurilateral Information Technology Agreement).

North-south regional trade agreements: The route for liberalising trade in environmental services

A reduction in tariffs and in welfare-reducing non-tariff barriers should help diffuse products and technologies necessary to reduce environmental damage (e.g. pollution at source or at end-of-pipe). However, very often these products and technologies form part of environmental projects that include services (e.g. wastewater management services, water collection and purification, recycling). Thus environmental projects have a great degree of ‘jointness’ or complementarity between the services provided by environmental goods and those provided by services, especially in developing countries where case studies show that the services included in environmental projects incorporate an increasingly large array of services that extend beyond those that are classified as environmental services (OECD 2005). Hence, it is necessary to go beyond the standard UN CPC definition to get an approximation of the environmental services that are most intensively used in environmental projects (see Figure 1b).

Regional trade agreements in services have grown rapidly, but estimates of trade costs for services indicate that they could be an order of magnitude higher than trade costs for goods (Miroudot and Shepherd 2013). Furthermore, the extra reduction in trade costs for agreement members are minimal, suggesting that it is difficult to give preferences in services as regulatory reform occurs de facto on a most-favoured nation basis so that commitments at the General Agreement on Trade and Services (GATS) (for those who made any since countries were not obliged to table offers) just consolidated members’ existing services policies. Measuring commitments (e.g. national treatment, restrictions on foreign ownership, restrictions on foreign service suppliers) is the best one can do to approximate commitments that – as for goods – are not a measure of actual policies. Figure 1 shows the values for an environmental-services-liberalisation index by income group adapted from Miroudot et al. (2013), which aggregates commitments by mode of supply for 155 services subsectors at the national and then at the income group levels. The index values show higher commitments for the high-income group where commitments are also higher for environmental services, perhaps a reflection that the environment is a normal good. Carrying out the same estimates for a data base of 57 bilateral trade agreements for which an OECD country, India or China is a party shows that “liberalisation” (as measured by the index values) goes further in regional trade agreements than multilaterally as most of the world market, particularly for infrastructure services, is in the hands of firms in high-income countries that have strong interests in prying open developing countries’ markets. Indeed, in these regional trade agreements, developing countries made substantial commitments almost opening entirely their environmental-service sector which they had kept closed in the GATS.

Figure 1. GATS score commitments for environmental services and other services

Source: Melo and Vijil (figure 3).
Notes: Index adapted from Miroudot et al. (2013). A score of 20 is no commitments. Income categories as in table 1. No data for low-income in the service commitments database due to no commitments. The narrow definition only considers environmental services as defined by the W/120 list; the wide definition adds to these ES the following W/120 sectors: professional services, research and development services, other business services, and construction and related engineering services.

It is likely that negotiations on environmental services will also be off the agenda, as negotiators will hold off taking them on board, pending the outcome of the Trade in Services Agreement negotiations. Should this agreement result in participants exchanging the best commitments they have so far undertaken in their preferential trade agreements (Marchetti and Roy 2013), environmental services could be substantially liberalised, as most of the opening has occurred on a north-south preferential basis. Because complementarities between trade in environmental goods and in services are especially strong in low-income countries, they are likely to lose the most if the agenda is not extended to tackle barriers and services. Should environmental services be on the agenda, negotiators are likely to stumble when it comes to agreeing on a more appropriate list than the current UN CPC. And even with a more appropriate list of environmental services, because it is far harder to monitor the fulfilment of commitments to liberalisation, disincentives to negotiate on services are strong especially when negotiating partners lack trust in each other. Reflecting on the lack of success with liberalisation of Services, Messerlin (2013) argues that ‘mutual equivalence’ rather than mutual recognition or harmonisation is the better way to go and that this route, which was followed by the EU Services Directive, might be best implemented on a regional basis where the trust necessary to agree on the regulations to be covered by mutual agreement is more likely.


Balineau, G. and J. de Melo (2013), “The Stalemate at the Negotiations on Environmental Goods: Can it be Broken?”,, 5 May.

Davos (2014), “Joint Statement regarding trade in environmental Goods”

Evenett, S. and A. Jara (eds) (2013), Building on Bali. A Work Programme for the WTO, A ebook.

Frankel, J. (2013), “Protectionist clouds darken sunny forecast for solar power”,, 7 August.

Kee, H., A. Nicita and M. Olarreaga (2008), “Import Demand Elasticities and Trade Distortions”, The Review of Economics and Statistics, 90(4), 666-682.

Kee, H., A. Nicita and M. Olarreaga (2009), “Estimating Trade Restrictiveness Indices”, The Economic Journal 90(4), 666-682.

Marchetti, J. and M. Roy (2013), “The TISA initiative: an overview of market access issues”, WTO Staff working paper # ERSD-2013-11.

Melo, J. de and M. Vijil (2014) “Barriers to Trade in Environmental Goods and Services: How Important are they? How much progress at reducing them?”, CEPR Discussion Paper No. 9869.

Messerlin, P. (2013) “The Quest for an Efficient Instrument in Services Negotiations”, in S. Evenett and A. Jara (eds), Building on Bali. A Work Programme for the WTO, A ebook.

Miroudot, S. and B. Shepherd (2013) “The Paradox of ‘Preferences’: Regional Trade Agreements and Trade costs in Services”, MPRA #41090, World Economy, forthcoming.

Miroudot, S., J. Sauvage and B. Shepherd (2013) “Measuring the Cost of International Trade in Services”, World Trade Review 12(4), 719-39.

OECD (2005) Trade that benefits the Environment and Development, OECD Trade Policy studies.


Topics: Environment, International trade
Tags: environmental goods, environmental services, trade agreements

Senior Fellow, FERDI; Professor, University of Geneva; CEPR Research Fellow
Research Associate, Foundation for International Development Study and Research