Chapter 3. The benefits of open services markets

Chapter 3 presents evidence that services trade impediments severely affect trade, which has wide-ranging consequences for other areas of economic performance. It reviews the findings of recent OECD analysis of the effect of restrictions on services sectors themselves, as well as in downstream industries that use services as intermediate inputs. The chapter also presents initial insights from on-going OECD analysis to develop trade cost equivalents for services barriers. Finally, it reports the preliminary findings of work estimating the value of commitments made in international trade agreements, particularly in terms of enhancing transparency and predictability in trade and investment.


Services generate more than two-thirds of global GDP, attract over three quarters of foreign direct investment (FDI) in advanced economies, employ most workers and create most new jobs globally. While the shift towards a service economy happens at earlier and earlier stages of development, trade policy has not always followed suit and impediments to global services trade remain more pervasive than in the manufacturing sector. Services trade restrictions create losses for firms that have a tougher time penetrating foreign markets, for household and business customers that are shut out from access to more diversified services at the best price, and for workers that lose the opportunities that would come from more exports and FDI. Conversely, integration in global networks as an exporter or as an FDI recipient can bring about streams of local value creation, innovation and quality jobs that are forgone every time restrictive policies deter a firm from expanding abroad.

Trade in services involves different business models through which firms best serve the needs of their overseas customers. A services provider can directly export abroad, for instance if it sells software through digital means (Mode 1: cross-border trade). Firms can also set up local affiliates when proximity to customers matters, as when a commercial bank opens a network of retail branches in a promising market (Mode 3: commercial presence). Service-exporting firms can send professionals abroad on a short-term basis, for example engineers co-designing projects locally with the client or technicians travelling to install and repair equipment (Mode 4: movement of natural persons). Lastly, international trade also comprises services provided in the firm’s own country to non-residents, primarily in the tourism industry (Mode 2: consumption abroad). These distinct business models often complement each other for firms wishing to offer a full suite of services to customers. In so doing, services exporters may have to overcome trade costs induced by the general policy environment as well as by policies that target each mode of supply.

To reap the full benefits of trade openness and regulatory reforms, policy makers need to identify regulatory bottlenecks and assess the trade-promoting or trade-restricting impact of policies in place. There are several dimensions through which regulations can depress trade and broader economic outcomes: being restrictive, inconsistent or unpredictable. The Services Trade Restrictiveness Index’s primary intent is to focus on the restrictiveness of trade and investment policies towards foreign services providers. At the same time, it can also be used to assess the costs of complying with different regulatory frameworks in different jurisdictions, as well as those of policy uncertainty arising from the gap between applied policies and international commitments.

A common way of quantifying the effects of trade policies is to convert indicators such as the STRI into ad valorem trade cost equivalents, or in other words to estimate how high a tariff-like instrument would need to be in order to produce a similar trade-depressing effect. Ad valorem equivalents are expressed as a percentage of the value of services provided abroad and provide an easy way to understand quantification of restrictiveness. It should, however, be kept in mind that these estimates necessarily provide a simplified view of a complex and highly diverse range of regulations, creating costs that may not always be proportional to the amount traded and that may not fall equally on all exporters and investors.

Empirical analysis1 reveals that the costs of services trade and investment barriers are high, largely exceeding the average tariff on traded goods, and that these costs apply to all modes of supplying services abroad. For instance, the average level of restrictions in force in the telecommunications sector amounts to trade costs of up to 150% on cross-border exports and 73% on foreign affiliate sales for highly customised services.

Those trade costs arise both from policies that explicitly target foreign suppliers, and more generally from domestic regulation falling short of best practice in the area of competition and rule-making. Further opening up of services markets through negotiations remains important for reducing trade costs, but attention should also be paid to reviewing sectoral regulations, as well as the efficiency of administrative and licensing procedures, to ensure they do not place an undue burden on new competitors.

Differences in how countries regulate the provision of the same service create additional costs for exporters that need to adapt to new sets of rules in each new market. Regulatory co‐operation has taken centre stage in trade and investment negotiations, but regulatory convergence in practice remains scant. When markets are relatively open, trade costs imposed by the average degree of regulatory differences range from 20% to 80%. However, lifting existing restrictions where they are still high should come first. Harmonising the rules of the game brings by far the largest gains when the countries undertaking regulatory co‐operation have already brought down their trade barriers.

As services trade restrictions are eased and regulatory co-operation makes tangible progress, small and medium-sized enterprises (SMEs) are the first to gain. The costs of dealing with regulatory hurdles and complying with diverging regulations in every new market fall more heavily on smaller exporters. Often, SMEs find that identifying the regulatory requirements of each country, adapting production methods and documenting compliance is beyond their capacity and give up seeking new customers in new markets. Reducing the costs of market entry would help improve the inclusiveness of services trade by allowing more SMEs to take up global opportunities.

With a strong potential for improving business dynamism, reviving competition and promoting FDI attractiveness, open markets and pro-competitive reforms in services can be an engine for productivity growth and competitiveness throughout the economy. In some segments of the transport and logistics chain, the trade restrictions in place raise prices by an average of about 20% for business users of services inputs, which ultimately falls on consumers. More intense competitive pressure could go a long way to secure better access to world-class services at attractive prices, thereby helping manufacturing firms compete on price and quality.

This chapter first presents the overall trade costs induced by the STRI for the different modes of supply. In all instances, the effects presented below should be understood as expected long-run outcomes of services trade liberalisation after economic actors adjust to the new policy environment, rather than as immediate responses.2 It then delves into the distribution of these costs among firms of different size, maturity and activities, and the overall benefits of better services regulation for the broader economy. Lastly, the main policy implications are outlined.

The costs of services trade restrictions

It is a well-established fact that international trade is more costly than domestic sales. These costs involve monetary factors, such as shipping, insurance and tariffs. Non-monetary factors, such as transport time or the challenge of enforcing contracts in a foreign jurisdiction, are also quantifiable. Information on the size of trade costs from policy barriers is particularly important for well-informed policy decisions. Services regulation can make trade more costly in a variety of ways, two of which are explored in this section. Most evidently, it can be expected that more stringent services trade restrictions in a country represent higher trade costs. In addition, there is evidence that bilateral trade costs also arise from the heterogeneity of regulation between importing and exporting countries. A third source of trade costs is uncertainty due to the fact that countries often apply policies that are more liberal than their commitments under international agreements, but which therefore cannot be counted on to prevail in the future.

Services trade barriers create large costs for exporters

Empirical analysis shows that the cost of cross-border trade in services resulting from protectionist and anti-competitive regulation is an order of magnitude larger than remaining tariffs for trade in goods. The average STRI3 in each sector represents trade costs that lie between 142% and 1800% for courier services, between 115% and 1191% for commercial banking, between 31% and 149% for telecommunication services and between 32% and 154% for construction services.4 Figure 3.1 shows the patterns of trade costs over a larger range of restrictiveness.

Figure 3.1. Ad valorem trade cost equivalents for cross-border exports

Note: Specialised services correspond to an import demand elasticity of -1.5, and standardised services to an import demand elasticity of -5.

Source: OECD calculations; see Benz (2017).

Sales of foreign affiliates seem to be slightly less affected by restrictive services regulation in ad valorem terms.5 However, since foreign affiliates usually sell larger volumes in a foreign market than the average cross-border exporter, total trade costs may be comparable across the two strategies. In transport services, the average level of services trade restrictiveness generates costs of commercial establishment that correspond to a mark-up between 51% and 292% on all sales of foreign affiliates (Figure 3.2). For foreign affiliates in the computer services sector, the average STRI represents trade barriers that raise the cost of supplying services by an amount equivalent to a tariff of 31% for standard services, and of 147% for specialised services. For telecommunication services, these figures are 18% and 73%, respectively. In the distribution sector, which accounts for a major share of trade via affiliate sales in most countries, trade costs are still far from negligible. On average, trade restrictions in that sector generate trade costs equivalent to a tariff of 10% (standard services) and 37% (specialised services).6

Figure 3.2. Foreign affiliate sales trade cost

Note: Specialised services correspond to an import demand elasticity of -1.5, and standardised services to an import demand elasticity of -5.

Source: OECD calculations; see Rouzet et al. (2017).

Trade cost equivalents are typically lower where services are relatively standardised. One example is courier services for standard letters for which quality differences in the provision of the service are small. For such services, consumers react strongly to even small changes in price by shifting demand towards cheaper suppliers. Conversely, suppliers of specialised services have more market power and can raise prices as alternatives are not easily found. This is often because they provide a quality that cannot be matched by competitors or they tailor the services to match the needs of each customer. Evidence suggests that banking and professional services are relatively more specialised while distribution is a relatively standardised sector. This means that trade costs for banking services are more likely to be at the upper end of the indicated range while they tend towards the middle of that range for most other services categories and they are at the lower end for distribution services.

Typically these trade costs do not increase linearly with the level of services trade restrictions. For cross-border trade there is some evidence that initial restrictions in relatively liberal environments tend to be more costly than additional restrictions introduced in more restrictive regimes. A potential explanation for this pattern is that above a certain threshold, trade costs might already be prohibitively high for exporting firms. Alternatively, higher degrees of restrictiveness may result in a situation in which a regulatory tightening does not further restrict the company’s business model because only a limited range of functions is still supplied to the foreign market.

Services are often supplied via a combination of different modes

At first glance, exporting or setting up a local affiliate are alternative means of reaching consumers abroad. Firms may choose between these modes of delivery by weighing the additional cost of a local establishment and the risk of a less agile model against the benefits of proximity to local consumers. However, there is some evidence that the modes of supply complement each other. In addition, cross-border exports as well as foreign affiliate sales seem to depend on the free movement of natural persons. Potential explanations are that large projects require a combination of different modes or that different segments of the customer base are served best through different channels. For example, in the construction sector, barriers to commercial presence (Mode 3) account for the lion’s share of the costs affecting cross-border exports in the sector. Barriers to Mode 3 also contribute to cross-border trade costs in courier services and telecommunication services.

Restrictions behind the border represent important barriers to cross-border trade and to affiliate sales. Especially in the telecommunications and maritime transport services sectors, cross-border exports are affected by weak enforcement of competition such as shipping agreements being exempt from competition law, inadequate regulation of dominant telecommunications providers, or lack of transparency with respect to procedures and regulation. Similarly, most important barriers to sales of foreign affiliates seem to arise behind the border, as summarised in Figure 3.3 by restrictions to “All Modes”. While weak pro-competitive regulation or a heavy regulatory burden are non-discriminatory in the sense that both domestic suppliers and their foreign competitors are affected, foreign suppliers are often more severely hampered due to their unfamiliarity with regulation and their being required to cope with different types of restrictions in different countries. This result highlights the importance of extending best-practice regulation and liberalisation beyond market access and national treatment restrictions. In fact, discriminatory barriers seem to be relatively less important for the decision to establish a commercial presence in a foreign market than the overall business environment.

Figure 3.3. Breakdown of foreign affiliate sales trade cost

Note: Specialised services correspond to an import demand elasticity of -1.5, and standardised services to an import demand elasticity of -5.

Source: OECD calculations; see Rouzet et al. (2017).

Commercial presence abroad also seems to rely heavily on the accompanying movement of people from headquarters to the affiliate, as shown by the contribution of barriers to Mode 4 to the costs of selling abroad through affiliates (Figure 3.3). Barriers to Mode 4 also explain a major share of trade costs for cross-border trade in courier services and commercial banking services, even though their contribution to the STRI score is relatively small in most countries. This indicates that many firms with a commercial presence abroad actually export via a combination of different modes and that affiliates may be more successful where they are also able to trade services back and forth with their headquarters. Policy should therefore aim towards liberalisation jointly in the areas of trade and investment in order not to distort the choices of exporters and to maximise the resulting benefits.

The benefits of liberalisation depend on an industry’s predominant organisational model

The business models for value creation in services are diverse and are not always organised in the form of a “value chain”.7 In fact, the most prevalent organisational model is the “value shop”, accounting for 50% of employment in major economies, in which value is created by experts and professionals using tailored solutions to solve specific customer problems. Another model is the “value network”, which creates value by linking customers with different needs, as in banking and insurance services; networks of physical infrastructure and “virtual platforms” like Uber or Airbnb are types of value networks.

Specific trade policy measures do not necessarily affect all organisational models. Services in value chains are often bundled with goods, making them sensitive to tariffs and non-tariff barriers to trade in goods, custom procedures or the lack of efficient infrastructure at ports and airports. By contrast, barriers to the movement of people are most relevant for value shops, which are largely traded via Mode 4 and rely on the access to skills and the access to customers.

Barriers to competition typically affect physical value networks where market imperfections need to be addressed through pro-competitive regulations within the country. A requirement of commercial presence is often the main restriction for virtual networks. It may arise in the form of restrictions to data flows, such as requirements of data localisation, which restricts trade even when designed to protect consumer rights.

Regulatory co-operation can reduce trade costs

Above and beyond the costs of dealing with restrictive policies, merely having to comply with regulations in target markets that differ notably from those of the home country is also costly for exporters. The heterogeneity of services regulation across countries is substantial and appears to be significantly receding in only a minority of sectors. For relatively liberalised countries with a very low STRI score of 0.1, the average degree of regulatory heterogeneity between two countries can account for ad valorem costs of cross-border trade between 20% and 80%, on average over all sectors (Figure 3.4). In more restrictive countries with an STRI score of 0.25 such heterogeneity still represents trade costs of between 12% and 45%.8 In other words, as countries open up their markets, regulatory differences become more costly. This effect does not outweigh the benefits of liberalisation, but it may slightly reduce them. The results show that regulatory co-operation can substantially stimulate services trade and that such cooperation becomes even more crucial when a sector has already reached an advanced stage of liberalisation.

Figure 3.4. The cost of regulatory heterogeneity for cross-border exports

Note: Specialised services correspond to an import demand elasticity of -1.5, and standardised services to an import demand elasticity of -5. Average regulatory heterogeneity among the 44 countries in the STRI database is 0.28. Regression lines are based on a pooled sample of cross-border exports between countries in the STRI database.

Source: OECD calculations; see Nordås (2016).

A predictable regulatory environment supports trade in services

While the heterogeneity in regulations creates costs for exporters, another source of potential costs comes from the unpredictability of the trade regime and the risk of policy reversals. Both exports and establishment abroad involve costs that are often sunk in the sense that once incurred, they cannot be recovered when a change in regulation no longer makes the foreign market profitable. Services providers are therefore more likely to engage in trade when they receive some guarantee that the regulatory regime will not become more restrictive in the future. This is precisely the role assigned to trade agreements covering services trade. These agreements include legal bindings that in some cases offer better market access to foreign providers, but that also have value when they simply bind the existing regime by offering a more predictable regulatory environment.

This positive impact of services trade agreements has been discussed for a long time in the trade literature following the entry into force of the General Agreement on Trade in Services (GATS) and the rapid increase in regional trade agreements (RTAs) covering services. However, it is only with the development of STRI indices providing a detailed account of all regulatory measures affecting trade in services that it has become possible to measure to what extent trade agreements bind the existing regime. Calculations done at the OECD have shown that there is some heterogeneity across countries and sectors (Box 3.1).

Box 3.1. To what extent are trade agreements binding in current services trade regimes?

Services trade agreements, such as the GATS, include market access and national treatment commitments that provide some guarantee to services exporters that the regime they face will not become more restrictive in the future than the level specified in the agreement. The difference between the bound level of restrictiveness permitted by the agreement and the actual trade regime is called the “water”. The higher the level of this water, the less predictable the trade regime is for potential exporters since countries can arbitrarily raise barriers within this interval without breaching their international agreements .

Figure 3.5 illustrates the heterogeneity across sectors in the level of the “water” in GATS commitments. For sectors like telecoms, distribution, computer services or construction, a predictable trade environment is offered to exporters at the multilateral level with market access and national treatment commitments being very close to the existing applied regime. For other sectors, such as audio-visual or transport services, there is more uncertainty. Regional trade agreements remove some of this uncertainty by going further than the GATS at the bilateral or regional level.

Figure 3.5. Average water in the GATS for selected STRI sectors

Source: OECD calculations; see Miroudot and Pertel (2015).

Binding the existing regime has a direct impact on the volume of bilateral trade, confirming that a predictable regulatory environment supports trade. By combining the information on the water in the GATS (Box 3.1) with further analysis of commitments made in RTAs, one can estimate the implied response of trade to the reduction in the uncertainty of the trade regime. For most services sectors in the STRI database, positive and significant effects are found. They are often in the same range as the ones observed for an actual reduction in the level of trade restrictiveness (i.e. a decrease in STRI indices). These results suggest that even when they do not actually remove trade barriers, services trade agreements still play a positive role through their legal bindings and the reduction in the uncertainty for services exporters.

Small countries suffer more from services trade barriers and may gain more from liberalisation

Country specificities play a role in how laws and regulations came to be and how the policy environment has a bearing on economic and social outcomes. Because of limitations in data and methodology, individual country-level differences with respect to the cost of a given policy stance cannot be robustly identified. However, a general pattern is that the trade effect of restrictions decreases with market size. In other words, a given element of regulatory liberalisation yields a larger increase of services trade in countries with a smaller GDP than in countries with a higher GDP. This is consistent with the hypothesis that larger internal markets make an economy less dependent on international trade, which also manifests itself in higher ratios of exports and imports to GDP in smaller economies.

So far, all trade costs have been measured in ad valorem terms, representing a mark-up over the price that is paid by consumers in the importing country. Such analysis shows that the benefits of services trade liberalisation are not spread equally across countries and sectors. Hence, an in-depth analysis of reform should be conducted in order to gain information on target sectors and policy barriers in order to maximise the economic gains from a services regulation reform agenda. However, while the economic effectiveness of reform is paramount for its success, no less important is the distribution of gains across the economy. The distribution of these gains is imminently linked to the structural composition of trade barriers into variable costs, fixed costs and sunk costs, requiring an analysis that goes beyond the calculation of aggregate ad valorem trade cost equivalents. The following section will therefore present more information on the decomposition of trade costs into different structural components and on how the gains of services trade liberalisation are distributed across firms.

A more inclusive globalisation of services?

While the previous section highlighted the aggregate costs of regulatory restrictions, it bears noting that these costs are not equally spread among services traders and investors. As the international community strives to design trade rules for a more inclusive globalisation, a key question is identifying which businesses lose out as a result of existing trade barriers, and in turn which firms would benefit most from further services liberalisation. It emerges that SMEs and new entrants are the hardest hit by policy impediments to trade in services, such that regulatory reform would help diversify export and growth opportunities for some of the most dynamic and job-creating actors of the services economy.

Services liberalisation spurs new entry into export markets by less experienced firms

Services trade barriers not only depress trade in services in aggregate, but also narrow the prospects of pioneer exporters by favouring experienced incumbents over firms that try to penetrate a new market for the first time. Every time a country opens up moderately to services trade so that its STRI falls by 0.1, the probability that a firm with no established trade relationship there start exporting to that country increases by between 2 and 12 percentage points. In contrast, liberalisation does not significantly boost the likelihood that incumbent exporters will remain in the market.

Even when they do manage to enter, the same restrictions are more costly for new entrants than for repeat exporters. For example, in a relatively restrictive market characterised by an STRI score of 0.4, new exporters have to incur costs equivalent to an additional 14 to 53 ad valorem percentage points on top of the regulatory costs faced by incumbent exporters (Figure 3.6). As newly created firms are primarily engaged in the service economy and, by definition, start doing business without export experience, co-ordinated services trade liberalisation could therefore emerge as a tool to boost start-up dynamism at the global level.

Figure 3.6. Additional trade cost of regulatory restrictions for new exporters
Estimated additional tariff equivalent of the STRI index compared to experienced exporters

Note: The numbers indicate the additional ad valorem tariff equivalent for firms without export experience, where export experience is defined as having exported the same service to the same country in the previous year. Specialised services correspond to an import demand elasticity of -1.5, and standardised services to an import demand elasticity of -5.

Source: OECD calculations; see Rouzet et al. (2017).

The penalty for new entrants reflects the importance of initial barriers in the overall burden of restrictive regulations. One can distinguish one-off entry costs, recurrent fixed costs and ad valorem variable costs (Box 3.2). Evidence from firm-level data suggests that while policy-induced variable costs are far from trivial, a substantive share of the trade-restricting impact of regulations is accounted for by fixed and sunk costs, which are borne irrespective of how much is actually sold in a foreign market. Such costs come for instance in the form of identifying and understanding the regulatory framework in place, adapting business processes to the requirements of the local regulators, or dealing with the administrative hurdles of acquiring and renewing a license to operate. Their primary effect is to reduce the number of firms that find it worthwhile to do business in a foreign market. Indeed, the number of foreign firms that establish trade relationships with partners in a given country is reduced where this country imposes more severe restrictions in the sector.

Box 3.2. Entry costs, fixed costs and variable costs

Trade costs, whether they arise from actionable policies or more structural geographical, historical or preference factors, come in various shapes and forms. A main distinction can be made between variable trade costs and fixed trade costs, with different implications for the extent to which trade restrictions affect the decision to enter a market (called the extensive margin) and the volume of exports conditional on entry (called the intensive margin).

Fixed trade costs may either recur each period or they may take the form of a one-time payment. Entry costs or sunk costs are incurred once and for all the first time a firm penetrates a foreign market. Such costs arise, for instance, from screening procedures for foreign investment, or demanding processes for the recognition of professional qualifications acquired abroad by foreign engineers, accountants or lawyers. They reduce the probability that firms that have not previously sold in a given market will decide to do so, but they have no impact on the probability of remaining in the market for repeat exporters or on the amount actually sold by active exporters.

Another type of fixed trade cost must be paid on a recurring basis, also regardless of how much activity is undertaken in the host country. For instance, burdensome renewal processes for licences to operate in a foreign market, data localisation requirements, or economic-needs tests for key personnel are likely to comprise a strong fixed cost component. High fixed costs discourage both new and repeat exporters from trading in a country altogether, but do not influence the amount exported once a firm has decided to serve a market.

Once a firm has paid all fixed trade costs, the volume of exports is only determined by variable or ad valorem trade costs, proportional to the value of services sold. Discriminatory regulations affecting foreign firms’ on-going operations in an overseas market, such as higher taxes on sales or profits generated by foreign suppliers or local sourcing requirements, tend to be of this nature. The effect of variable trade costs is the most similar effect to that of tariffs: they discourage entry into foreign markets and also reduce the amount exported by firms that do sell abroad.

SMEs bear the highest cost from restrictive trade policies

Smaller firms often expand internationally by pursuing a niche strategy or providing a suite of customised services to few clients. The digital economy has raised hopes that it would breed countless such opportunities for small and medium enterprises to go global in a borderless world. Yet borders are still highly relevant in services regulation and where high entry and fixed costs mean that scaling up and attaining critical mass rapidly is a pre-condition for success, SMEs are the most harmed by restrictions. Fixed trade costs are all the more costly to absorb for firms that export modest amounts as those costs cannot be spread over a large volume of foreign sales. Larger firms with deep financial resources and extensive networks of business partners are also better equipped to overcome the compliance cost of exporting in complex regulatory environments, and may have sufficient market power to pass the cost of regulation on to consumers. As a result, smaller would-be exporters are the first to be discouraged from entering more restrictive markets at all, and they realise lower sales in those markets if they do manage to recoup the cost of entry.

In other words, a given level of regulatory restrictions represents larger trade costs for smaller firms. For instance, an average level of service trade restrictiveness corresponding to a score of 0.2 on the STRI is as costly as an additional tariff of up to 14 percentage points on small firms’ exports, relative to large firms with turnover of EUR 400 million and over (Figure 3.7a). In a more restrictive environment represented by an STRI index of 0.3, the extra burden on small firms is compounded and is equivalent to an extra tariff of over 20 percentage points for the smallest firms trading customised services.

Figure 3.7. Additional trade cost of regulatory restrictions for SMEs
Estimated additional tariff equivalent of an STRI of 0.2 compared to large firms of EUR 400 million or more

Note: The numbers indicate the ad valorem tariff equivalent of an STRI score of 0.2 for small and medium-sized enterprises on top of what is paid by firms of 400 million and more in turnover. Specialised services correspond to an import demand elasticity of -1.5, and standardised services to an import demand elasticity of -5.

Source: OECD calculations; see Rouzet et al. (2017).

Setting up and operating an affiliate abroad involves a wider range of sunk and fixed costs, regulatory or otherwise, than trading on an arm’s length basis: acquiring or leasing real estate, tying up capital in the new venture, obtaining the necessary approvals and licenses, recruiting local employees and bringing in highly qualified personnel from headquarters, entail procedures and expenses that come before any service is actually rendered in the host market. It therefore comes as no surprise that regulatory restrictions create even more of an obstacle to invest abroad than to export cross-border for firms that do not enjoy the benefits of scale. For instance, the extra burden of an STRI of 0.2, for a medium-sized firm of EUR 5 million in turnover selling specialised services through foreign affiliates is estimated to be equivalent to an additional 19% tariff compared to large firms, whereas the extra equivalent tariff is 9% on cross-border exports of the same services (Figure 3.7b).

In a similar fashion, differences in regulatory approaches dealing with the same policy issue are more of a deterrent for SMEs. Smaller exporters lack the scale that may make it profitable to invest upfront in adapting to unfamiliar sets of rules. More often than not, SMEs also lack the legal and administrative capacity to gather information about the regulatory requirements in new markets, adapt their processes and document compliance where those requirements differ widely from regulation at home.

Firms with multiple ties to the host market are better able to deal with regulatory hurdles

Which other factors, apart from size and experience, increase firms’ immunity to the effects of restrictive policies in their target markets? Some firms do not only aim to expand their customer base abroad, but appear to have an overarching reason to sell services in a specific country rather than seeking more favourable conditions elsewhere; these firms are less sensitive to the services trade policies they encounter. Two such motives stand out: providing services to the exporter’s foreign headquarters or as complement to its parent’s local activity, and adding value to manufacturing exports though services activities.

First, firms that are subsidiaries of a foreign corporate group have a particular advantage when dealing with restrictive regulations in the country of their parent. When a country has an STRI score of 0.4, the ad valorem equivalent of the cost of services trade restrictions is between 4 and 14 percentage points lower for firms that are subsidiaries of a parent firm in that country than for firms with no ownership ties to the target market (Figure 3.8a). This suggests that affiliates of foreign firms benefit from their parent company’s knowledge of how to navigate complex requirements in the headquarter country. Moreover, exports back to the home country include services provided to the parent firm. Intra-firm transactions are likely to be driven by an established division of tasks within the firm and less sensitive to regulatory hurdles than arm’s length dealings.

Figure 3.8. Estimated discount on the tariff equivalent of the STRI index for specific firms

Note: The numbers indicate the amount by which the ad valorem tariff equivalent of the STRI is reduced for firms that have headquarters or ultimate owners in the destination markets, compared to firms with no ownership links in the country (left); and for firms also exporting goods to the same country, compared to pure services exporters (right). Specialised services correspond to an import demand elasticity of -1.5, and standardised services to an import demand elasticity of -5.

Source: OECD calculations; see Rouzet et al. (2017).

Second, services exported jointly with goods are even less sensitive to policy conditions in the destination market. Firms exporting both goods and services to the same country tend to be present in more restrictive destinations than firms selling services alone. In other words, goods exporters perceive a restrictive policy regime to be less costly for their services packages than exclusive services exporters. For a relatively high STRI score of 0.4, the ad valorem tariff equivalent of this discount can be as high as 35 percentage points for customised services (Figure 3.8b). This is probably because, when offering a joint supply of services, goods exporters add value to their manufactured product exports, a phenomenon often referred to as bundling of goods and services (Box 3.3). Where the primary driver of a firm’s services exports is demand for exports of its physical products, complying with costly services regulations may be viewed as a necessary means for reaping the gains from manufacturing sales, rather than a prohibitive factor.

Box 3.3. How do manufacturing firms participate in services trade?

Evolving models of how firms do business and create value mean that the distinction between goods and services trade has become increasingly blurred. Manufacturing firms both import and export a range of services. Services trade can be driven by manufacturing activities through several channels:

Services inputs imported in market-based transactions. Many services procured from external suppliers are essential inputs in local and global value chains. They enable firms to manage operations seamlessly across facilities and production stages (e.g. transport, logistics, communications, financing, outsourced back-office functions), add value to the products and improve processes (e.g. technical testing, engineering, design when not conducted in-house). The Trade in Value-Added database has revealed that such services account for about a third of value-added in global manufacturing exports, 40% of which are services imported directly or indirectly by the manufacturers (OECD-WTO, 2017).

Services inputs provided within a corporate group. Some key services are provided in-house, either by the headquarters to affiliates or by a firm’s local and offshore subsidiaries to headquarters. Examples are R&D centres, marketing and sales teams, legal counsel, internal audit or human resources. For instance, over 80% of US imports of R&D, business and management consulting and IT services are purchased from within the importer’s corporate group, while telecommunication and engineering inputs tend to be outsourced (Figure 3.9).

Figure 3.9. US services imports by affiliation, 2015

Source: OECD calculations based on data from the US Bureau of Economic Analysis.

Services sold jointly with goods. Firms increasingly offer integrated solutions to their clients, comprising bundles of goods and complementary services. Car manufacturers may operate car dealerships themselves and provide credit and leasing to their customers; high-tech manufacturers sell embedded software, feature customisation, after-sales repair and maintenance services packaged with physical goods. In Finland, where high-tech equipment and electronics dominate industrial exports, manufacturers realised more than four-fifths of total export sales in computer services and a quarter in architecture and engineering services between 2008 and 2014. In Belgium, more than half of exports of architecture, engineering and telecommunications services in 2014 accompanied sales of goods by the same firm to the same country, as did a quarter of computer services sold abroad (Figure 3.10).

Figure 3.10. Share of bundled and non-bundled services exports in Belgium, 2014

Source: OECD calculations based on data from the National Bank of Belgium.

Services liberalisation

Evidence from the previous sections makes a clear case that impediments to trade and investment weigh on foreign firms’ decision to enter new markets via cross-border trade and/or the establishment of an affiliate. Beyond trade itself, inadequate regulation on market entry, competition rules or administrative procedures place a heavy burden on services firms and their customers, impairing overall business dynamism and competitiveness.

Consumers and firms pay the cost of trade restrictions

As well as those restrictions that specifically target foreign firms, it is clear that the role of domestic regulation related to barriers to competition and lack of regulatory transparency should also be examined. When competitive pressures are weakened, incumbent firms tend to consolidate and expand their market power, reducing the chance for disruptive innovation by new players. Restrictive domestic regulation creates a policy environment that discourages new entrants, whether they be potential domestic businesses or foreign competitors. For this reason, the cost of weak competition policy is also covered in the analysis of trade restrictiveness.

The final cost of a policy environment that reduces competition from new entrants, whether domestic or foreign, is ultimately borne by consumers and downstream business customers, which pay higher prices and enjoy less choice than they would in more competitive markets. The resulting price premium for domestic users of services can be quantified as a sales tax equivalent on their purchases.9 On average across 42 countries, estimates of this tax equivalent range from about 3% in road freight transport to almost 40% in broadcasting, with considerable variation across countries in all sectors (Figure 3.11, Panel A). In some segments of transport and logistics, as well as in construction, restrictive regulations raise the price of services by about 20% on average, and in some countries by almost 80%, imposing substantial additional costs on manufacturing enterprises and eventually on final customers.

Figure 3.11. Average estimated tax-equivalent of services trade restrictions, by sector

Note: The estimates are simple averages of the tax equivalents observed in 42 countries. Restrictions to foreign entry, to movement of people and other discriminatory measures mostly reflect discriminatory policies, while barriers to competition and lack of regulatory transparency are mainly associated with domestic regulation.

Source: OECD calculations based on the BvD Orbis and STRI databases; see Rouzet and Spinelli (2016).

Inadequacy or lack of enforcement of competition law and opaque rule-making (together with lengthy or costly customs procedures) could be among the main factors explaining high costs of services in telecommunications, transport and logistics, where such impediments allow firms to charge between 10% and 20% more than they would have been able to in more competitive markets (Figure 3.11, Panel B).

While behind-the-border regulation applies equally to domestic and foreign companies established in a given market, it therefore stands out for being just as relevant as discriminatory measures listed in trade and investment negotiations, since ultimately it contributes to raising the cost of services across the board. It follows that substantial pro‐competitive gains could be reaped from trade and investment liberalisation, but also from policy reforms targeting inefficient domestic regulation in services markets.

Furthermore, evidence discussed in the previous section has shown that restrictions to services trade introduce substantial entry costs favouring experienced incumbents over exporters venturing into new markets for the first time. Further openness of services trade would thus not only bring overall gains from more intense competitive pressure, but also encourage the entry of young and small firms that start exporting only when market conditions are more favourable, thereby creating a more dynamic and diversified economy.

Efficient intermediate services help manufacturing businesses compete

Intermediate services inputs provide essential links in value chains, from product development to production, marketing and sales. They move parts and components to the assembly line and final products to end-users; they design, engineer and innovate; they monitor demand and consumer tastes, and transmit such information to the product developers; they provide funding and insurance; and they facilitate compliance with laws and regulations in all the markets where the product is sold. Depending on the position in the value chain, intermediate services reduce costs, upgrade quality and match suppliers and customers around the world. Where services regulations depart from best practice, the cost surcharge highlighted above hampers manufacturers competing for market shares abroad: they are less competitive on price and quality, export lower volumes, and are less diversified across markets and products.

A rich and cost-effective services supplier base is a driver of comparative advantage in industries that are intensive users of intermediate services, which are often high-end manufacturing sectors. Taking into account in-house service functions as well as outsourced services used by manufacturing firms, the total contribution of services activities to the value of manufacturing exports ranges across countries between 40% and 60% (Figure 3.12). While offshored services are the most impeded by trade restrictions, the cost and quality of services provided in-house or procured on the domestic market are also likely to be affected by the prevailing competitive conditions.

Figure 3.12. In-house, outsourced and offshored service value-added in manufacturing exports, 2011
As a % of gross exports

Source: OECD ICIO and occupational data; see Miroudot and Cadestin (2017).

Freight transport directly underpins global trade in components and final goods. It should, therefore, be no surprise that countries with higher restrictions on air, maritime or road transport and courier services are less successful exporters of manufactured goods, particularly in time-sensitive products where speedy and reliable delivery can make or break a sale. Not only do restrictive transport conditions at home translate into lower export volumes, but they also limit the ability of manufacturing industries to reach a large number of foreign markets and to offer a diversified set of products. Figure 3.13 shows that restrictions on road freight (relative to other transport modes) most strongly inhibit exports of various key industries (automotive, electrical equipment and chemicals) (Figure 3.13).

Figure 3.13. Estimated impact of halving the distance to best-practice STRI on manufacturing exports by sector, average of 44 countries, 2014

Note: The methodology is based on Arbache et al. (2016). Best-practice STRI means the country with the lowest score for the sector in the sample.

Source: OECD calculations based on the CEPII BACI database, 2014 data.

In the same manner, access to credit on competitive conditions is a prerequisite if firms are to compete in international markets. Open markets encourage widespread access to bank funding and compressed interest rate spreads, which in turn help firms get financing for innovation, working capital and trade. Where the regulatory environment is conducive to trade and FDI in commercial banking, manufacturing exporters in industries that rely on external finance perform better, entering more markets and developing more products for overseas customers. For instance, reforming the regulatory environment in commercial banking so as to halve the distance from the most liberal market could, for the average country, boost exports of electrical machinery by almost 8%, and automobile exports by about 3% (Figure 3.13). Intermediate inputs in fragmented global supply chains are more sensitive to financial conditions than are final goods, and exports in volatile industries such as fast fashion, and to countries with higher commercial and political risks, also depend critically on well-developed and well-functioning credit markets at home.

Knowledge-intensive services like software, communication services and R&D add value to products and processes. Fast transmission of information across production locations, efficient systems to monitor markets and to adjust supply continuously to customer tastes, computer-assisted design and manufacturing, and cutting-edge product development are all vital determinants of competitiveness in high-tech products. Innovative sectors are less successful on export markets when the regulatory environment limits competitive pressure at home in computer services, telecommunications and engineering. Under such conditions, R&D-intensive industries such as chemicals and transport equipment tend to export lower volumes, with sales thinly spread across markets in a more homogeneous set of varieties.

Policies with respect to services influence manufacturing export performance mainly because services help reduce production and transaction costs. For example, best-practice regulation supporting open and efficient transport, IT and finance is associated with more competitive export prices in downstream sectors, particularly so in final goods. Conversely, where essential services are not available at the best price due to regulation, the more expensive input costs and less efficient production processes tend to be reflected in less competitive export prices, which limit firms’ ability to penetrate foreign markets.

Competitive telecommunications, in particular broadband internet, play a particularly decisive role in supporting quality improvements and vertical differentiation. Best-practice regulation in the sector is associated with a higher unit-value of manufacturing exports, suggesting that they are value-adding services which contribute to tilt exports towards the higher end of the market. This can be explained by the positive effect of pro-competitive regulation on the deployment of high-speed internet services, which provide firms with real-time tools to track and match ever shifting consumer preferences and to access information and innovation. Furthermore, internet services open up new possibilities to command a premium by bundling goods and services, including after-sales services, establishing customer communities on social media and the like.

Box 3.4. Bringing services from a bottleneck to a springboard for manufacturing in Brazil and India

India and Brazil are two emerging markets which have promoted the development of a stronger manufacturing base as prime policy priorities, through recent initiatives such as “Make in India” or Plano Brasil Maior. Detailed analysis of the experience of these economies highlights how upgrading services, including through better regulation, can be critical for the successful emergence of high-end industries. Improving the physical and virtual connectivity on which manufacturing supply chains rely is particularly stressed.

Brazil has established a solid comparative advantage in natural resources sectors, and its exports have largely relied on the strong competitiveness achieved in commodities. Brazil’s manufacturing faces the challenge of diversifying and upgrading its export base. Less than 0.5% of all formal-sector firms sell abroad; the top 10 products account for 45% of total goods exports, and the share of high-tech exports has been falling over time. Export diversification would be strongly boosted by improving services that reduce production and trade costs, and that are currently underperforming, in particular finance, transport and logistics services. Brazil’s top manufacturing exports, such as food products and iron and steel products, do not rely heavily on external credit, but industries targeted as priorities for development (automobiles, semi-conductors and medical equipment) require effective credit intermediation to grow. Some products like food and petroleum products benefit from Brazil’s strength in natural resources to remain internationally competitive despite the high cost of transport, but improvements in supply chain infrastructure may be a necessary condition for sustainable expansion in the chemicals and automobile parts industries. Brazil would then be in a favourable position to leverage its highly-performing engineering sector to further develop innovative industries.

In contrast to Brazil, India’s exports in priority sectors (automobile and components, chemicals, electrical machinery, pharmaceuticals and apparel) already exhibit an advanced degree of diversification across products and markets, but they are thinly distributed and face challenges in scaling-up. Indian manufacturers also obtain export prices that are 6-15% lower than average in these products, indicating that they remain confined to low-end segments of the global market. To address the constraints on quality competitiveness and to scale up existing exports, analysis shows that a lack of high-speed internet services is creating a major bottleneck, especially regarding exports to high-income economies like the Unites States and the European Union, which are India’s primary customers. While Indian consumers and businesses enjoy a vibrant and competitive market for mobile telecommunications, the country lags a long way behind the world average broadband penetration and access to secure servers. Pro-competitive regulatory reforms to improve fixed broadband services could significantly help extend the reach of Indian industrial businesses.

Source: Arbache et al. (2016); Benz et al. (2017).

Well-designed regulatory systems make markets more attractive for foreign direct investment

In addition to the welfare cost of services trade and investment barriers to consumers and businesses, they also reduce the attractiveness of a market for foreign direct investment. Not only are tight regulatory environments less appealing for multinational enterprises (MNEs), but even those foreign multinationals that are able to absorb higher entry costs and succeed in establishing an affiliate in more restrictive markets have to face disproportionately high operational costs, which jeopardise their contribution to knowledge diffusion and value creation in the host economy.

Firms establishing abroad may face outright discriminatory regulations, such as nationality requirements for board of directors and managers, restrictions to the recruitment on foreign highly qualified personnel, prohibition from hiring locally licensed professionals, limited access to government contracts, and so on. All these factors increase their operational expenses. Multinational companies also bear higher compliance costs when facing new, unfamiliar and often different regulatory frameworks in every new market they enter. Consequently, countries seeking to improve their attractiveness for foreign investment could consider easing trade and investment barriers through unilateral or negotiated liberalisation, but also pursuing efforts to align aspects of their domestic regulation with regulatory environments observed in more open and FDi-engiendly economies.

Foreign investment provides a number of benefits to the host country, by acting on economic, social and political levers. Multinational companies create job opportunities by recruiting local employees. Furthermore, multinational activity is often associated with skill and technology transfers. Foreign affiliates may employ more advanced production techniques that can be diffused to the domestic economy, thereby contributing both to the skill upgrading of local workforce and to increased innovation incentives in the whole industry. The host country could also benefit from a wider choice of services products, of better quality and, under favourable regulatory conditions, at more attractive prices than local or imported substitutes.

Finally, multinational companies tend to expand their activities well beyond the geographical boundaries of the host country, hence contributing to boosting the host country’s own exporting activity. Foreign affiliates act to a notable extent as export platforms, in particular supplying financial, transport, computer and distribution services not only to their local host markets but also to third countries (Box 3.5).

Box 3.5. FDI-export platforms: The case of US and Japanese affiliates

One of the channels linking FDI attractiveness and trade performance is the FDI export platform phenomenon. In this operating model, MNEs establish affiliates abroad not only to supply services to local customers, but also to serve as regional hubs or as specialised service centres for their corporate group in neighbouring markets.

Foreign affiliates of US services MNEs are largely set up to serve their host countries, with on average more than 70% of their total sales addressing local demand (Figure 3.14). However, foreign affiliates of US non-services parents are more strongly engaged in exports and intra-firm trade, which together comprise more than half of their total activity. This suggests that foreign services subsidiaries of US manufacturing MNEs are in large part established to support the production of the parent company or to reach out to further markets, acting as distribution hubs.

Figure 3.14. Decomposition of total foreign affiliate sales, by final destination

Note: The shares show respectively the percentage of sales of foreign affiliates that are destined to the local markets (local), to the parent firm (intra-firm) and to third countries or to unrelated parties in the home country (exports). Average shares across sectors over the periods: 2008-12 for the United States and 2008-13 for Japan.

Source: OECD calculations based on data from the US Bureau of Economic Analysis and the Japanese Ministry of Economy, Trade and Industry.

Similarly, the activity of Japanese foreign affiliates directed back to the headquarters or to third markets ranges between 40% and 50% of total foreign affiliate sales, depending on the main activity of the parent company. Hence Japanese MNEs, regardless of their core activity, appear to follow a strategy of establishing regional bridgeheads to sell more broadly than in their host market.

Nevertheless, the purpose of establishing foreign services affiliates differs considerably across sectors (Figure 3.15). In sectors such as construction, logistics and telecommunications, Japanese foreign affiliates are established mainly to achieve a better penetration of the domestic market. However, in some other sectors Japanese MNEs establish affiliates abroad not just to address local demand but especially to act as a springboard for exports. This export-platform phenomenon is most prevalent in finance and transport services. In transport, exports back to the headquarters account for almost one fourth of total affiliate sales. Together with exports to third countries, they represent over 70% of affiliate sales. In financial services, exports to third countries command two thirds of foreign affiliate sales.

Figure 3.15. Decomposition of Japanese total foreign affiliate sales, by sector, 2008-13

Note: The shares represent the percentage of sales of Japanese foreign affiliates that are destined to the local markets (local), back to the parent firm (intra-firm) and to third countries or unrelated parties in the home country (exports).

Source: OECD calculations based on data from the Japanese Ministry of Economy, Trade and Industry.

Export-platform sales are also relevant in computer and distribution services, with 37% and 38% of affiliate sales respectively. This evidence indicates that Japanese MNEs tend to develop a web of local establishments abroad to expand their client base both locally and in neighbouring countries, but also as providers of support functions to sustain the activity of the corporate group.

Services trade liberalisation boosts the competitiveness of the domestic economy

Open services are essential elements of domestic competitiveness and productivity growth. This section has highlighted the two main channels through which open markets bring wider benefits to the economy. Services trade liberalisation has the potential to boost a country’s export performance by attracting foreign investment. Foreign firms bring about further advantages for the host market, in the form of new job opportunities, and skill and technology transfers.

Moreover, reducing services trade restrictions is also likely to influence exports by enhancing the competitiveness of existing firms both in services and in downstream manufacturing. New entrants push local services providers to become more efficient and more innovative in order to preserve their market shares. In becoming more productive, domestic firms can be encouraged to venture into new foreign markets. At the same time, the least competitive and inefficient companies might also be driven out of the market once they are no longer sheltered by protectionist measures. While this may require costly adjustments in the short run, it has the potential to unlock a far more efficient reallocation of resources towards high-performing firms with strong growth prospects.

Implications for trade policy-making

This chapter has examined the magnitude, nature and impact of the costs entailed by restrictive services trade policies. The new evidence provided is meant to inform trade policy makers about the likely effects of unilateral or concerted regulatory reforms and help prioritise policy action. Among the main implications emerging from the analysis are that:

  • The existing stock of services restrictions has a clear dampening effect on cross-border trade and foreign investment activity. There is substantial scope for reducing trade costs in major services sectors by scaling back discriminatory measures towards foreign providers, but even larger gains would occur if more general domestic regulation regarding competition and transparency were concurrently improved.

  • As explicit trade barriers come down, the costs of having to comply with different regulations in each jurisdiction become more significant for internationalised firms. There is ample opportunity for regulatory cooperation to make doing business easier for exporters while preserving adequate consumer protection and other public policy objectives. However, where high explicit restrictions to services trade still prevail, it is essential for them to be reduced before regulatory cooperation can make a substantial difference.

  • Smaller and less experienced exporters face a heavier cost burden in more restrictive regulatory environments. Yet more and more often, small and young firms need to seek opportunities abroad to deal with stagnant domestic demand and rising scale economies in the digital world. Opening up services markets would primarily benefit SMEs, which are responsible for the greater part of new job creation.

  • The business models for supplying services are complex, often involving a combination of modes, a bundle of goods and services, or a mix of digital products and face-to-face interaction. Where possible, policies should avoid, distorting those business models, specifically by adopting a balanced approach towards the different modes of supply through trade, investment and competition policies.

  • There is not one service industry, but many services with different business models, competition challenges and best-practice regulatory frameworks. To maximise the benefits from reforms, targeted policy advice needs to identify the main bottlenecks by taking into account the level of preventable trade costs in each sector, how far sectoral regulation diverges from that of key trade partners as well as flow-on effects on downstream segments of the economy.

  • Reforming services brings benefits for consumers but also strengthens productivity and performance in the domestic economy. Under competitive pressure, services firms are pushed to innovate and offer the most cost-effective solutions to their customers. Modern manufacturing is a heavy user of services inputs and its competitiveness relies on access to state-of-the-art suppliers at the best price. Countries with more favourable and transparent regulatory environments are also more attractive for FDI, stimulating economic activity, jobs and exports.

  • The benefits of services trade liberalisation are, however, unlikely to come without temporary disruptions to labour markets. For gains from trade to materialise, some firms need to close down while better performing ones expand. Reforms towards more competition in services therefore work best for society if accompanied by effective safety-nets and active labour market policies. Such policies should be designed to mitigate the short-term effects on workers who may be displaced, to enhance their opportunities to find new jobs, and to make it more attractive and less risky to start new ventures.


Arbache, J., D. Rouzet and F. Spinelli (2016), “The role of services for economic performance in Brazil”, OECD Trade Policy Papers, No. 193, OECD Publishing, Paris,

Benz, S. (2017), “Services trade costs: Tariff equivalents of services trade restrictions using gravity estimation”, OECD Trade Policy Papers, No. 200, OECD Publishing, Paris,

Benz, S., A. Khanna and H. Nordås (2017), “Services and performance of the Indian economy: Analysis and policy options”, OECD Trade Policy Papers, No. 196, OECD Publishing, Paris,

Lodefalk, M. and H.K. Nordås (2017), “Trading firms and trading costs in services: The case of Sweden”, Örebro University, School of Business Working Paper (forthcoming).

Miroudot, S. and C. Cadestin (2017), “Services in global value chains: From inputs to value-creating activities”, OECD Trade Policy Papers, No. 197, OECD Publishing, Paris,

Miroudot, S. and K. Pertel (2015), “Water in the GATS: Methodology and results”, OECD Trade Policy Papers, No. 185, OECD Publishing, Paris,

Nordås, H. (2016), “Services Trade Restrictiveness Index (STRI): The trade effect of regulatory differences”, OECD Trade Policy Papers, No. 189, OECD Publishing, Paris,

Nordås, H. and D. Rouzet (2015), “The impact of services trade restrictiveness on trade flows: First estimates”, OECD Trade Policy Papers, No. 178, OECD Publishing, Paris,

OECD-WTO (2017), “Trade in value added”, OECD-WTO: Statistics on trade in value added (database),

Rouzet, D., S. Benz and F. Spinelli (2017), “Trading firms and trading costs in services”, OECD Trade Policy Papers, forthcoming.

Rouzet, D. and F. Spinelli (2016), “Services trade restrictiveness, mark-ups and competition”, OECD Trade Policy Papers, No. 194, OECD Publishing, Paris,

Stabell, C. and Ø. Fjeldstad (1998), “Configuring value for competitive advantage: On chains, shops, and networks”, Strategic Management Journal, Vol. 19, pp. 413-437.


← 1. This chapter builds on the findings of Arbache et al. (2016), Benz (2017), Benz et al. (2017), Miroudot and Pertel (2015), Nordås (2016), Nordås and Rouzet (2015), Rouzet et al. (2016), and Rouzet and Spinelli (2016).

← 2. The time period covered by the STRI and of the available trade and investment data precludes the identification of trade costs from policy changes within countries over time. Estimates presented in this chapter are derived from cross-country differences in policies, and are therefore better suited to represent the long-run potential of regulatory reforms.

← 3. The average STRI in the sectors considered is 0.29 for courier services, 0.24 for commercial banking, 0.23 for telecommunications and 0.23 for construction.

← 4. In other sectors, data are insufficient to conduct an analysis with satisfactory robustness or the resulting coefficients are not significant. Statistical uncertainty and the lack of precise information on the sensitivity of import demand to price changes mean that only a range of trade cost estimates can be reported.

← 5. These results are based on confidential firm-level data that cover foreign affiliates of corporate parent companies in Finland, Germany, Japan and the United States.

← 6. The average STRI in the sectors considered is 0.30 for transport services, 0.23 for computer services, 0.23 for telecommunications, 0.19 for distribution, 0.23 for financial services, and 0.23 for construction.

← 7. The discussion in this subsection is based on Miroudot and Cadestin (2017). The concept of value shopsand value networks was first introduced by Stabell and Fjelstad (1998).

← 8. These estimates describe the average level of trade costs across all sectors and may hide significant variation at the sector level. Hence, it may well be that trade costs due to regulatory differences are substantially higher in some sectors.

← 9. There are two ways through which trade restrictions raise prices: barriers to entry create rents for existing firms, which can charge higher prices as long as they are shielded from competition from imports and from the potential entry of new suppliers; and excessive regulation increases the cost of compliance for established firms, which can be passed on in higher prices. Either way, these restrictions bring about a cost that is similar to a sales tax on consumers and downstream companies.