Setting priorities for services trade reform

Background

Seminal work to measure impediments to services trade was supported by two previous ARC grants.

The first project (Findlay and Warren 2000) made significant intellectual headway by recognising that services are highly differentiated, so it was not appropriate to use the price comparisons often used to quantify non-tariff barriers in goods trade, since these assume homogeneity across borders. Strictly speaking, it may not even be appropriate to talk about a ‘tariff equivalent’, since this (a) assumes that services are primarily traded cross-border, and (b) often assumes that the domestic and foreign services are perfect substitutes.

Because services trade barriers operate behind the border, the research quantified the effects of those barriers on a behind-the-border measure of economic performance. In generic terms, econometric methods were used to estimate the effects of some measure of services trade barriers (TB) on some measure of economic performance (Y) in a market, controlling for all the other factors (Z) that affect economic performance in that market. The estimated model was then used to construct the value Y’ that would obtain in that market if there were no services trade barriers (normally, if TB took a zero value). The actual models varied from one sector to the other, but were typically drawn from the literature on structure, conduct and performance for the sector in question.

The methodology required a quantitative measure of services trade barriers (TB). Qualitative information about regulatory restrictions was converted into a quantitative index (or indexes), using a priori judgements about the relative restrictiveness of different barriers. It was sometimes possible to use the econometric stage to test the weighs that were assigned a priori to different categories of restrictions in the first stage, essentially by reestimating them. Barring econometric problems, this was done by entering the index scores for the different categories of restrictions separately into the estimating equation.

The second ARC-funded project (Sidorenko and Findlay 2003; Stephenson, Findlay and Yi 2002) explored in more detail the links between services trade barriers and domestic regulation. A key methodological issue was to distinguish between regulatory measures that were protective and those that were designed to meet legitimate economic or social policy objectives. One approach was to decide a priori which policy measures were designed to meet legitimate objectives, and to exclude them from the analysis at the outset. Another approach was to treat regulation on a continuum, and to ask the question whether the current level of regulation was ‘too little’ or ‘too much’. Conceptually, this was accomplished by allowing for a non-linear relationship between regulation and performance, and identifying at what point the degree of regulation had the least adverse incidental effect on economic performance.

There are two recent critiques of the services trade barrier measurement methodology, by Whalley (2004) and Deardorff and Stern (2005).

Whalley (2004) first notes that with multiple restraints on trade it is not clear which restrictions are binding and which are not, and that different restrictions will have different effects on trade. This argues for continuing to separate out the different dimensions of trade barriers and entering them separately into the econometric models of sectoral performance. Second, he notes that there may be country discrimination in the application of barriers, even though both de jure and de facto discrimination are breaches of national treatment under the GATS. This has been dealt with by explicitly identifying both discriminatory and non-discriminatory measures, using policy information sources that go well beyond GATS schedules. Third, he states that quantity-based measures are ‘typically based on model-generated residuals given by observations relative to econometric model predictions’ (p. 1239). This describes Francois and Hoekman (1999), but not the work in previous ARC-funded projects. Finally, he presumes that international price comparisons are the appropriate way to measure barriers to services trade, but notes that ‘price differences across countries for services need not be related to barriers, even if they could be measured’ (p. 1239). As noted, the highly differentiated nature of services makes direct international price comparisons an inappropriate tool, so the research has instead undertaken behind-the-border studies of market structure, conduct and performance to generate price or cost impacts, controlling for a range of other factors.

Deardorff and Stern (2005) assert first that the appropriate way to model services trade barriers is as a tariff equivalent, but all the examples they give to justify this are of discriminatory trade barriers. This pays insufficient attention to the critical importance of non-discriminatory barriers in services trade. Second, they note that if the econometric models are not structural enough to control for the effects of services quantities, then estimates of the elasticities of supply and demand are required to convert the price impacts from econometrics into vertical shifts in supply or demand curves. A solution to this issue (discussed in more detail below) is to begin to control for quantities in the econometric work. Third, they discuss the issue of assigning weights in a restrictiveness index by judgement or by the statistical technique of factor analysis. They state that ‘[Judgement] may well be the best approach if the investigator really is knowledgeable, as in the case when an index is being constructed for a specific narrowly defined industry. … [Factor analysis] is a purely statistical technique that is not, in our view, necessarily an improvement on the use of judgement weights’. (p. 569) Finally, they provide guiding principles for measuring barriers to services trade. They note that no single methodology is sufficient, and investigators need to draw on all available information, including both direct observation of particular barriers and indirect inference using data on prices and quantities. These are the approaches used in the two previous ARC-funded projects.

Subsequent papers that have adopted the above methodology for measuring the direct effects of services trade barriers include Barth, Caprio and Levine (2004), Fink, Mattoo and Neagu (2001), Clark, Dollar and Micco (2004) and OECD (2005). Papers that have used the estimates of barriers to services trade to evaluate the economy-wide effects of services trade liberalisation include Dee and Hanslow (2001), Dee, Hanslow and Phamduc (2003) and Dee (2007).

The applications have demonstrated one key shortcoming of the research to date. The single biggest determinant of the projected gains from services trade liberalisation (and the policy priorities that flow from that) is whether the trade barriers are modelled as affecting markups or affecting real resource costs. This ‘treatment’ effect often dominates the estimated ‘height’ of the trade barrier, and accounts for some of the variation in modelling outcomes observed by Whalley (2004). But whether barriers create rents or add to resource costs is severely under-researched currently. In some cases, the empirical evidence is suggestive, but not conclusive, because only one performance measure has been used. In other cases, a price impact is estimated, and then it is simply asserted whether the effect operates through price-cost margins or through real resource costs.

The first aim of this project is to produce new empirical estimates of barriers to services trade that measure explicitly whether they add to markups or to real resource costs. The second aim is to expand existing cross-sectional datasets on regulatory barriers to trade in services by looking at how regulatory structure have evolved over time. This will allow panel estimation, so that country-specific influences can be controlled for, and the separate economic effects of different regulations can be identified.

Barth, J., Caprio, G. and Levine, R. 2004, ‘Bank Regulation and Supervision: What Works Best?’, Journal of Financial Intermediation, 13, pp. 205–48.

Clark, X., Dollar, D. and Micco, A. 2004, ‘Port efficiency, maritime transport costs, and bilateral trade’, Journal of Development Economics, 75, pp. 417–50.

Deardorff, A. and Stern, R. 2005, ‘Empirical analysis of barriers to international services transactions and the consequences of liberalisation’, in P.Dee and M. Ferrantino (eds), Quantitative Methods for Assessing the Effects of Non-tariff measures and Trade Facilitation, APEC Secretariat and World Scientific Publishing, Singapore, pp. 549–609.

Dee, P. 2007, ‘East Asian economic integration and its impact on future growth’, World Economy, 30(3), pp. 405–23.

Dee, P. and Hanslow, K. 2001, ‘Multilateral liberalisation of services trade’, in Stern, R. (ed.), Services in the International Economy, University of Michigan Press, Ann Arbor, pp. 117–39.

Dee, P., Hanslow, K. and Phamduc, T. 2003, ‘Measuring the cost of barriers to trade in services’, in Ito, T. and Krueger, A. (eds), Services Trade in the Asia-Pacific Region, NBER-East Asia Seminar on Economics, Volume 11, University of Chicago Press, Chicago, pp. 11–43.

Findlay, C. and Warren, T. 2000, Impediments to Trade in Services: Measurement and Policy Implications, Routledge, London and New York.

Fink, C., Mattoo, A. and Neagu, C. 2001, Trade in International Maritime Services: How Much Does Policy Matter?, Working Paper No. 2522, World Bank, Washington DC.

Francois, J. and Hoekman, B. 1999, ‘Market access in the service sectors’, Tinbergen Institute, manuscript, cited in B. Hoekman 2000, ‘The next round of services negotiations: identifying priorities and options’, Federal Reserve Bank of St Louis Review, 82(4), pp. 31–47.

OECD 2005, ‘Modal Estimates of Services Barriers’, TD/TC/WP(2005)36, OECD, Paris.

Sidorenko, A. and Findlay, C. 2003, Regulation and Market Access, Asia Pacific Press, 2003.

Stephenson, S., Findlay, C. and Yi, S. 2002, Services Trade Liberalisation and Facilitation, Asia Pacific Press, Canberra 2002.

Whalley, J. 2004, ‘Assessing the benefits to developing countries of liberalisation in services trade’, World Economy, 27(8), pp. 1123–53.

Links:

Home
Background
Aims
Approach
Industry partner organisation
National benefit
Personnel
Conference program and papers — 2009
Conference program and papers — 2008
Conference program and papers — 2007

[back]