The concept of trade in services and how did it develop?

While the economies of developed countries are dominated by services, the treatment of international trade continues to focus on goods. The most extreme version sees services as non-tradable even though some goods producing industries are non-tradable (construction and mining are location specific) and transportation services have always been at the heart of trade in goods. The nearest equivalent to trade in goods for services is direct cross-border transactions in which the purchaser of the service is located in one country and the provider in another. Services also cross borders when the consumer moves to the country of the provider to obtain a service or when the producer temporarily moves to the country of a client to provide a service. In the terminology of the General Agreement in Services (GATS), the three different types of service provision are known as modes 1, 2 and 4. Collectively they constitute cross-border trade in services.

Since there are no custom posts for such services, surveys provide the source of data for these transactions and are published as part of the balance of payments statistics. Statistics on trade in services cover modes 1 and 2 with very limited coverage of mode 4 (Bensidoun and Ünal-Kesenci 2007). Overall, services account for 20% of world exports, a share that has been stable since the mid 1990s. Only one tenth of world services output enters international trade, compared to over half of the production of goods (UNCTAD 2004).

The seemingly rather modest contribution of cross-border trade in services compared to its share in output needs to be qualified in a number of ways. First, it measures trade as opposed to tradability. If services can be traded but face non-tariff and other barriers that effectively diminish the extent to which services are traded, there may be considerable untapped potential for trade in services. Second, services enter into trade as inputs to the production process of traded goods. They can do this either directly as intermediate consumption to an industry producing traded goods or indirectly embodied in inputs produced in other manufacturing industries that supply inputs to the final traded industry. According to the OECD, in the mid-1990s services accounted directly or indirectly for about 22% of manufacturing production (Pilat and Wölfl 2005). Furthermore, it found that the importance of services embodied in manufacturing has increased substantially since the 1970s. Even within manufacturing, services contribute a considerable share to the output that is finally traded. In the OECD in 2002, occupations that can be considered as services made up about 40% of all persons employed in the manufacturing sector.

Conceptually, the share of service occupations in manufacturing covers three different types of situations. Certain of these occupations could be outsourced but the firm prefers to produce their services in-house. Others are integral to the production process and could not realistically be outsourced (for instance production engineers, sales staff). Finally, manufacturing firms also produce services for sale either as joint products (for instance an operating system with a computer) or as a separate add-on service (a maintenance contract or car finance). Available data on turnover for manufacturing firms broken down by activity is scarce. For the few available countries, service activities make up 10–15% of total turnover of manufacturing firms. When such services are supplied by manufacturing firms across borders, they may or may not be recorded separately from goods.

The belief that cross-border trade in services may be under-reported and that in any case there exists potential for greatly increased trade either for specific services or for specific occupations has given rise to a vigorous debate on the extent and possibility of off-shoring for services (Blinder 2005; McKinsey Global Institute 2005; Liu and Trefler 2008). As yet it is accepted that such effects have been modest but views differ sharply about the extent to which the phenomena will grow and the nature of the impact should it do so.

All services are tradable through a permanent commercial presence in the host country (mode 3 of the GATS terminology). Indeed on different measures this constitutes the dominant mode of supply of services abroad today. It is therefore surprising that commercial presence has generated relatively modest interest compared to that generated by cross-border trade in services. This in spite of the fact that until recently many services remained closed to investment from abroad either as a result of direct supply through a public administration or through public ownership of network industries or through various restrictions based on the nationality of the individual supplying the service.

Data on foreign direct investments (FDI) provide the most comprehensive source for studying trade through a permanent presence. Under the impulse of globalisation both cross border trade and FDI increased significantly during the last 25 years. Contrary to cross-border trade, a clear shift has taken place in the share of services in FDI. By 2005, services represented three fifths of the global FDI stock up from a half in 1990 (UNCTAD 2007). This can be taken in part as an illustration of the difficulties of cross-border trade for services and in part of the need for a direct and on-going presence to supply many services.

The figures for FDI tend to underestimate the significance of permanent presence for the provision of services because non-equity forms of participation constitute an important means for foreign companies to penetrate the markets of third countries. They include such techniques as franchising, management contracts, concessions and public–private partnerships and are prevalent in industries as diverse as hotels, restaurants and fast-food, car rentals, retailing and construction. For many professional services, partnerships composed of locally established professionals are often the dominant or even required form of operation. Ownership restrictions of airlines have led to the growth of alliances of major carriers.

Data on cross-border mergers and acquisitions (M&As) is more up-to-date than that for overall FDI. It confirms the growing importance of services with world-wide cross-border M&As rising from 37% in 1987–1990 to 58% in 2002–2006. Unlike green field investment which tends to increase competition on the domestic market, mergers can pose the danger of powerful foreign suppliers dominating the domestic market. Perhaps the data of greatest interest is that of the local sales of foreign affiliates that can be compared directly with cross-border sales of services and also with sales of local companies. Data on Foreign Affiliates Trade in Services (FATS) is however very limited. According to the CEPII analysis of FATS data from four major investor countries (US, France, Germany and Japan), commercial presence was the dominant form of exchange varying from 70% for the US and France, 59% for Japan and 49% for Germany.Footnote 1

To explain these different developments we have to dig a little bit into the concept of trade in services and the content of services. First of all, services are a broad concept covering very diverse activities from hairdressers to IT services, from accountants to retail. The papers in this issue focus on producer services. Producer services cover business services, trade, communication, transport, financial, and energy services (see Rubalcaba and Kox 2007 for an overview). Government services and consumer related services such as personal services and tourism are not included in the analysis. Within the group of producer services most papers concentrate on the globalisation of business services, in some cases to a specific type of business services like knowledge-intensive business services (KIBS).

Next, different approaches are possible. One is largely empirical, sometimes unconventional, based on the growing availability of data on services and the possibility of using large sets of micro data. The other is more theoretical, based on models of how trade in services should perform under certain assumptions, using a general equilibrium framework. Examples of both these approaches are provided in this volume.

Tradability of services

To what extent are services tradable in theory compared to the actual, rather modest, share of services in cross-border trade? This is the issue addressed by two contributions, those of J. Bradford Jensen and of J. Francois and J. Woerz. Using an original approach, the geographical concentration of service activities, Bradford Jensen identifies what activities are traded within the US. He then infers that activities that are traded within the US are likely candidates for trade with other countries. This procedure leads to the conclusion that there is a significant share of US employment in the service sector that appears to be in tradable activities and that therefore there is the potential for significant trade in services. Indeed, there is more employment in professional and business services in industries classified as tradable than in the manufacturing sector in tradable industries.

Looking at the characteristics of workers in tradable high-tech service activities, Bradford Jensen finds that they are better educated and have higher earnings than workers in non-tradable high-tech services. He concludes that these activities are less likely to be vulnerable to competition from low-wage, labour abundant countries. In addition some worker inputs into service production might be tradable even though the service industry is not, for instance back office operations for retail banking. Overall the typical white-collar occupation involves an activity that could be traded. This conclusion evidently provides support for the argument that high skilled occupations could be vulnerable to off-shoring. Bradford Jensen however rejects such a conclusion and considers rather that increased trade in services is likely to benefit US service firms and workers.

On the basis of newly developed establishment level data on the service sector, business service producers have similar characteristics to manufacturing in terms of scale, wages and even the use of other inputs. Service establishments that engage in exporting tend to be larger, pay higher wages and are more productive than non-exporters. Bradford Jensen concludes that the increase in services trade is likely to increase productivity, as it did in manufacturing. It remains an open research question to what extent technological barriers inhibit services trade and to what extent it should be attributed to other factors such as regulation and local standards, language and culture.

Francois and Woerz examine the role of services as inputs in manufactured exports as well as the relationship between service sector openness and manufacturing performance in different sectors. Using a combination of data from input–output tables and from trade in services, they observe a number of important relationships. To begin with, exports of producer services embodied as inputs in manufacturing are strong, confirming the OECD findings, and are irrespective of the level of income by country. The relative service content of exports nevertheless is significantly correlated with income levels—especially for business services. Service linkages to manufacturing have become increasingly important in the 1990s. They find that labour intensive industries involve lower multiplier effects in higher income countries. In contrast, multiplier effects within service sectors grow stronger with increasing activity in technology intensive industries in more advanced countries.

On the issue of the openness of the producer service sectors and the relative efficiency of manufacturing industries, Francois and Woerz have a number of significant findings. Exports become increasingly producer service intensive with a rising level of development. This is what one would expect from the overall link with development. In addition, inward FDI and trade openness in the service sector benefits the export competitiveness of manufacturing sectors with stronger service linkages and hurts those with weaker upstream linkages to services. Increased service sector openness produces a shift in value added and employment to service-input intensive manufacturing sectors. As a result, service sector openness has boosted the competitiveness of more technology and skill intensive industry in the OECD at the expense of sectors like textiles and clothing, thereby reinforcing comparative advantages. Again business services play a leading role in boosting competitiveness supporting the attention paid to business services in older studies (Rubalcaba-Bermejo 1999; Ochel and Wegner 1987).

Trade through permanent presence

What are the effects of trade through a permanent presence compared to cross-border provision of services? This issue is addressed by J. R. Markusen and B. Strand who use a general-equilibrium model to analyze off-shoring of business services in small open economies. They address the questions of what sort of activities can be off-shored and what activities would a firm in a high-income country choose to off-shore among those that are feasible to do so. Codifiability, routinization and the need for face-to-face contact are viewed as essential elements to off-shoring and not merely the skill intensity of an activity.

Faced with the complexity of trade in services comprising different types of skill as well as both investment and trade, a simple two-good, two-factor Heckscher–Ohlin is insufficient to explain why relatively skilled-labour intensive services are being off-shored to skilled labour scarce countries. By extending their model to include both headquarters and office functions for services employing two types of skills, one routine and one non-routine, and both unskilled and skilled final goods production, Markusen and Strand are able to provide a much richer interpretation of the possible effects of trade in services.

According to the model, trade and investment in services tends to benefit a small, skilled-labour abundant economy. There are several sources to these gains. Off-shoring allows service firms to source from abroad the downstream part of service that is costly at home. Access to foreign service providers increases the range of services available to domestic manufacturing producers. Access to foreign markets or the domestic service firms makes them more competitive, allowing them to spread their fixed costs over a larger output. The effects under the different scenarios differ according to skill level. The highest skill level in services (so called entrepreneurs) reaps the highest rewards. On most scenarios, the high skilled also gain, not least because the efficiency gains dominate the relative wage changes. Only in some scenarios do the high skilled actually fall (but the low skilled even more), because they have to compete with off-shore workers of the same skill level.

P.-J. Norbäck and L. Persson examine the welfare effects of cross-border mergers & acquisitions and the implications for policy. They employ a model with entry barriers and oligopolistic competition. According to the policy framework, the size of entry barriers and the level of complementarities between the domestic firm being acquired and the foreign multinational acquiring, the results in terms of the price paid to shareholders, the prices for consumers and profits for the acquirer can differ substantially. Complementarities are important because they indicate the size of the efficiency gains to be reaped from an acquisition.

Discriminatory cross-border merger laws risk being counterproductive by blocking M&As associated with complementarities that would benefit both domestic consumers and domestic owners. Restrictions on competition from the takeover of a domestic firm are limited by the possibility of green field entry and expansion by rival firms. General restrictions on cross-border acquisitions would not therefore seem to be well founded with the possibility to block those which would unduly restrict competition on the domestic market to be preferred.

Specific takeover regulations that increase the cost of foreign acquisitions could be counter-productive because these policies imply that there will be few potential buyers of domestic firms. Under optimal conditions shareholders will gain from a higher bid price and consumers from lower output prices when there are more bidders. However, when complementarities are very important, more domestic firms are willing to sell and the bid price declines. Under these conditions the gains accrue to the foreign acquirer leaving domestic firms worse off and may reduce welfare. Overall, having more potential buyers seems to provide the best outcome with harmonization of EU takeover regulations beneficial for potential acquiring firms but foremost for owners of target firms.

Intellectual property and services share the characteristic of being intangibles. What is the relationship between the two and how should policy evolve to ensure a supportive framework for the development of both? These are the questions addressed by K. E. Maskus in his contribution on the globalization of intellectual property rights and innovation in services.

The extent to which services produce and protect intellectual property varies greatly from service to service. Software, recorded entertainment, broadcasts, research services and databases are all industries that generate intellectual property and are protected by rules for IP. Although the situation differs somewhat between the US and Europe, service enterprises that protect intellectual property formally tend to rely on copyright rather than patents. However, nearly all services use trade mark protection to build brand awareness. Rather than formal protection, many service enterprises use trade secrecy, lead-time advantages, market loyalty and specific enterprise to protect or exploit their intellectual property rather than patents or copyright.

In terms of policy, it is important to establish whether existing forms of protection for intellectual property are appropriate for the development of services, whether such protection needs to be reinforced or extended and the degree to which services suffer from an unequal application of existing rules among different countries. In general, Maskus takes a restrictive view of these issues. He is opposed for instance to the patenting of business processes or of applying patents rather than copyright as the appropriate form of protection for software. Of crucial importance is the fact that services are a major user of intellectual property as well as producers and that increasing transaction costs can slow the take-up of innovation. Inadequate protection of trademarks and the possibility to copy recorded music and video products through digital media certainly represent problems for the service sector.

Policies to liberalise trade in services

Two contributions address the issue of why efforts to liberalise trade in services appear to have such a limited impact. B. Hoekman looks at the GATS under the WTO and the framework it provides for world-wide liberalisation. M. Canoy and P. Smith examine services and the single market and the degree to which the EU's internal market has been able to effectively promote continent-wide integration for services.

Hoekman points out that trade agreements are an instrument to promote market access and liberalization through quid pro quo bargaining on trade policies. Countries negotiate away negative terms-of-trade externalities created by partner countries imposing trade restrictions. Alternatively, trade agreements may offer a commitment mechanism to governments for desirable domestic policy reforms. However, neither of these motivations appears to be sufficient to ensure substantial progress towards the liberalization of trade and investment in services under the GATS.

Poor progress is ascribed to a number of possible causes. Governments are implementing liberalization autonomously, thereby undermining the need for commitment. Since there are no tariff barriers, direct cross border trade may already be very open. The same is true for FDI. Indeed the openness to FDI may undermine the desire of governments to make further progress on cross-border trade. Nevertheless there remain very substantial barriers to both trade and investment in services. Not all of these are discriminatory but may stem from barriers to entry that apply equally to domestic and foreign suppliers. Hoekman also examines institutional causes, such as the fact that the negotiating set may be too small or that the pro-GATS forces among exporters are insufficiently strong. Incentives for governments may be weak when the relevant policies are under their own control, as in the case of tourism. Finally and most importantly, many barriers to trade in services stem from national regulations, which are not easily amenable to negotiations under the WTO.

In many ways, the contribution of Canoy and Smith is complementary to that of Hoekman. Unlike the GATS or regional trade agreements such as NAFTA, the European Union has a legal order that enables it to assume regulatory responsibilities required to liberalise markets. Contrary to received wisdom, much of the single market programme has been directed to services and the design of an adequate European regulatory regime. This for instance has been the case in areas as diverse as financial services, different modes of transport, telecommunications, postal services and to a certain extent energy. It is true that in areas covered by the services directive, considerable latitude is provided for the regulatory regimes of Member States and that this is the area where integration has progressed least far.

In spite of the possibility to regulate services at EU level, integration for services has remained slow outside a few specific sectors (telecommunications, road and air transport, wholesale financial services). Many, but by no means all, services are characterised by their bespoke nature and need for customisation that limit the possibility for standardisation and economies of scale. Issues of asymmetric information are rife. Many services have potentially long-lasting consequences that will only show up after the service has been supplied. Quality is an overriding consideration for the purchaser but one that is difficult to measure. All of these characteristics are particularly problematic when it comes to cross-border provision even when such provision is physically possible. As a result, a permanent presence remains the preferred form of integration for most services. Free movement of labour has also proved a very important factor of market integration which is generally not available outside the EU.

In large part, therefore, non-regulatory barriers can explain the difficulty of market integration. Oligopolistic market structures can hinder integration and require strong competition policy to complement the regulatory structure. In general, a range of policy instruments both at national and at EU level are required in order to effectively liberalise trade in services. The European experience with deep integration shows that obstacles to trade in services are often indicative of broader and deeper problems with the functioning of national markets for services, which must first be addressed for successful international integration.

What have we learned from these seven papers on the globalization of services trade and how can these papers be embedded in the broader literature on trade in services? We try to answer these questions by addressing six issues.

  1. 1.

    The tradability of producer services

The papers of Bradford Jensen and Francois and Woerz address the tradability of services. The facts presented by Hoekman suggest that business services are becoming more tradable. Are producer services different than goods with respect to tradability? The answer is affirmative. Quite often the proximity between producer and consumer is needed which limits the opportunities for services trade. In addition, many services are intangible, services are often more customized and less standardized, in particular business services, and services are often more regulated than goods, among others to ensure quality of the services and service provider. From this perspective, the degree of tradability of producer services seems to be an intrinsic characteristic of services, like intangibility and lack of standardization. However, these characteristics are not fixed forever. In particular, the IT revolution led to the codifiability of services, and the possibility to transmit services output via IT. On the other hand, tradability seems to be limited by other (policy) factors like regulation. These factors can in principle be affected by policy intervention.

Based on the geographical concentration of services industries Bradford Jensen concludes that much services are internationally tradable because these are tradable within countries. That is an interesting thought, but surpasses underlying factors determining tradability. Francois and Woerz stress the intangibility of business services and their trade potential attached to goods trade. The manufacturing good is the vehicle by which the producer service is traded. It does show that services are a bigger part of cross-border trade than they normally account for in statistics. However, the paper does not address the potential and limitations for trade in services.

Some literature addresses this topic. Gravity equations are used to explain bilateral trade in services with the level of regulation in the exporting and importing country as explanatory variables. High levels of regulation seem to hamper services trade (see among others Kimura and Lee 2006; Mirza and Nicoletti 2004; Kox and Lejour 2006).Footnote 2 These high levels of regulation are often an indicator for the functioning of national services markets. However, the precise mechanisms between business services trade and regulation are not clear. The indicators are quite rough, covering large parts of the economy and the indicator used for trade in services is often total services. Total services includes also tourism (mode 2), government services, producer services closely linked to goods trade (like transport and wholesale and retail) and business services less linked to goods trade. Ideally, one would try to explain the relation between trade in a specific service and the regulation related to that service to do justice to the diversity in service activities and the subtle effects on regulation on services activity. Now that more data are becoming available at least some part of their diversity in services trade and regulation can be covered. Hopefully, this kind of research can teach us more on factors hampering and stimulating trade in services. It could also give some indication on the tradability of specific services.

  1. 2.

    The complementarity between services and goods trade

Francois and Woerz focus on indirect exports of producer services in manufacturing exports. Goods and services exports are also directly linked through such services as transport and trade. In an extensive survey of the literature, Anderson and van Wincoop (2004) divide trade costs into three broad categories: transport costs, border costs (which include policy barriers, but also language and currency barriers), and retail and wholesale distribution costs. They find a 170% ad-valorem tax equivalent of all (goods) trade costs for a developed country. This can be roughly divided into 21% transport costs, 44% border costs and 55% retail and wholesale distribution costs. These results emphasize the significance of services in trade. Not only transport and trade services are important but also business services related to security, information, language and currency (as part of the border costs). These services come on top of the production of a good, and should also be modelled in that way, not as intermediate input in production but as a margin in the trade relation. Because the good is traded, the service is also often traded. In particular, transport and wholesale services could be complementarity to goods trade. From this perspective the tradability of the goods affects the tradability of the service.

The Francois and Woerz results also suggest that availability of foreign producer services (imports) increase the skill level and technology mix of manufacturing exports. Kimura and Lee (2006) also found a complementarity between goods exports and services imports. These results indicate that bigger tradability of producer services benefits also productivity and trade in manufacturing. Both papers establish an empirical relation, but the theoretical relation is not specifically modelled. It points to the importance of services in manufacturing production and the importance of a range of available services inputs as is suggested by the endogenous growth literature (love of variety). Whatever the precise mechanism foreign services provided by cross-border trade or foreign establishments seem to increase productivity and pinpoint the importance of the tradability of services for economic growth.

  1. 3.

    Complementarity between cross-border trade and foreign affiliates

Foreign markets can be served through cross-border trade or local sales of a foreign affiliate. Markusen and Strand model the interaction between cross-border trade and establishing or foreign affiliate. For manufacturing many theoretical and empirical papers have studied this relation. The literature seems to suggest that the case of complementarity between FDI and trade is stronger than the case of substitution, which is at least associated with vertical FDI. Proximity between producers and consumers could also be a determining element for choosing the mode to serve the foreign market in manufacturing but for services it is often decisive. For example, in banking and insurance cross-border-trade is relatively modest and foreign establishments are more important. Some empirical papers have investigated this relation. Buch and Lipponer (2004) examine the relationship between FDI in the German banking sector and cross-border financial services by German banks and conclude that in this sector FDI and trade are complements. Fillat-Castejón et al. (2008) also conclude that services cross-border trade and FDI are complements for a sample of OECD countries between 1994 and 2002. In the longer term the complementarity even increases through the services imports by the foreign affiliates. According to their analysis business services, communication and financial services have the biggest cross-border trade potential if regulation is reduced and FDI is increased.

Given that cross-border trade in services and FDI are complements, the tradability of services also depends on the freedom to conduct FDI. FDI restrictions and other barriers to establish a foreign affiliate will have a negative impact on trade in services.

  1. 4.

    Heterogeneity of service firms in productivity and trade performance

Bradford Jensen refers to the recent literature on the heterogeneity of firms in international trade in a section ‘lessons from manufacturing’. Many papers in this area use firm level data for a specific country and study the relation between productivity and the international orientation of firms (Mayer and Ottoviano 2007 and ISGEP 2007 present some overviews).

Most of these data focus on manufacturing firms, but sometimes services firms are also included. Studies show that exporting firms are more productive than non-exporting firms. Moreover, most firms only export to a few destinations and a few products, but some high-productive multinationals are responsible for a large share of a country exports. They also export to many destinations. These results also suggest that exporting firms are already more productive before entering the export market. A related result is that (fixed) market entry costs seem to matter more for trade than tariffs. The question is whether these results also hold for services. If this is the case, the tradability of services is determined by market-entry costs, and depends on the extent to which service providers can overcome these market entry costs. The results of the heterogeneous firms’ literature also suggest that the tradability of services can be improved if productivity increases; this is an interesting hypothesis to explore.

  1. 5.

    Knowledge-intensive business services inputs and productivity

Knowledge intensive business services (KIBS) produce often a tailor made service with a high knowledge content such as software and computer services and professional services. KIBS could also contain some part of financial services and telecommunications although overall the latter sectors are on average less knowledge intensive. KIBS are important in disseminating tacit knowledge on production and markets they learned from their clients (Rubalcaba and Kox 2007). These knowledge spillovers are difficult to control. Sometimes exclusive arrangements on intellectual property rights (IPR) are signed between KIBS and clients. Maskus discusses at greater length whether these type of arrangements contribute to innovation and the knowledge dissemination.

These spillovers could be important for productivity. Guerrieri et al. (2005) explore the role of business services in knowledge accumulation and growth and the determinants of knowledge diffusion by modelling and estimating the imports of business services in production growth. It has to be said they estimate large productivity effects from the knowledge spillovers compared to other studies focusing on manufacturing imports. In their view increasing trade in business services is important for growth as is also concluded by Francois and Woerz. Trade in business services disseminates more easily knowledge. The size of these knowledge spillovers due to KIBS is still unclear. From manufacturing we know that receiving firms and countries need at least to have the absorptive capacity, i.e. a minimum level of skills and R&D to benefit from these spillovers. There are no automatic gains from knowledge spillovers. It is still unclear to what extent this holds for services.

  1. 6.

    Outsourcing trade in tasks

François (1990) models producer services coordinating complex production processes, so that freer trade enhances gains from specialization. His model could explain how firms can take advantage of the globalization process by an appropriate establishment of foreign affiliates and contractual relations with suppliers. This is further developed by Grossman and Rossi-Hansberg (2006) who consider that outsourcing strategies can deeply change the nature of international economic relationships. In their model producing final goods involves the performance of many tasks that requires a different set of production factors. Then, firms can break-up their production process, offshoring tasks in different countries, depending on comparative advantages and transport costs. Grossman and Rossi-Hansberg’s model leads to conclusions that differ greatly from those derived from traditional trade models. Traditionally, lowering the cost of task trade improves the productivity in final good industries. In this model the productivity effect favours also the factor used in imported tasks, so that trade liberalization may benefit all factor-owners in the economy. The model is not specific to trade in services, but is a promising framework to study the consequence of freer trade in services. According to their ideas, firms have to focus on trade in services tasks not on services sectors, a promising avenue to explore. In this literature the question of the legal relationship between the final producer and its suppliers is not addressed. Having the “right” legal institutions could be an important fundament for services trade.

Conclusions

What brings the tradability of services for small open economies? The answer depends on the perspective you take. It is nothing new if you consider cross-border trade in services as a similar development to trade in goods, i.e. services trade is nothing special. However, it is hard to defend that customerization, proximity, imperfect competition, and intangibility are less important or just as important for services than for goods. Although some characteristics of services, like intangibility and proximity, limit the tradability, trade in services is affected by many other factors. Many of these factors are discussed in this special issue like the complementarity between goods and services trade, the complementarity with foreign establishments (FDI), productivity of services to overcome market entry costs, the knowledge intensity of business services and the wish to protect innovation by IPR. Also other policy factors are discussed, like discrimination of foreign service providers, limited market access and the level of regulation. From theory we know that many of these mechanisms affect the tradability of services, but we are less sure about the impact of them. We also know that the importance of these factors depend on the type of services. A single overall approach clearly does not fit. However, data limitations so far hamper extensive empirical research, although progressing is being made.

What is in it for small advanced countries? To some extent comparative advantages play an important role, but economies of scale are probably much more important. Lack of standardization and imperfect competition limit the market size for small countries. Improved tradability of services could free these countries. Their service exporters could benefit from large export markets. Domestic manufacturing and service firms could benefit from a richer variety of business services inputs. From this point of view small advanced economies could benefit more from the tradability of services than large economies. In general, advanced small economies have the skills and innovative capacity to absorb and benefit from knowledge spillovers. The skilled workers seem also benefit the most from globalization of services.

The policy recommendation would be to stimulate services trade by reducing market entry costs for the home and export markets. It creates opportunities, but it is not only gold that is shining. Less competitive service firms have to leave the market or have to become more competitive. Firms become more vulnerable for merger and acquisition by the pools of foreign capital. This could bring in more productive capital and more knowledge, but also changes ownership relations and creates uncertainty. However, autarky has not proved to be a good strategy to generate wealth in the past. Most of the papers in this issue point convincingly at the positive productivity effects of trade in producer services.