Chapter

IV Effects of Integration

Author(s):
International Monetary Fund
Published Date:
December 1990
Share
  • ShareShare
Show Summary Details

The EC internal market program has potential effects on the allocation of resources not only within the EC but also in countries with close ties to the Community, such as the EFTA countries. Increased integration in the EC is widely expected to stimulate the EC economies and to contribute to a better use of their resources. For the EFTA countries, the principal issue is whether participation in the EC integration process—assuming that the EC would agree—would yield substantive economic gains; and whether the decision not to join would entail substantive losses.

The implications for the EFTA countries will depend on both the form of the internal market and the future relations between the EC and EFTA countries. Although many institutional arrangements may be possible, it is convenient for analytical purposes to concentrate on two highly stylized cases: a nonintegration case, in which the EFTA countries do not integrate further with the EC; and an integration case, where EFTA members fully participate in the economic dimension of EC integration (which does not necessarily imply that they must become full Community members). Arrangements in between, which appear most likely, may be viewed as a combination of these two cases.

Structure of EFTA Foreign Trade

The EC has long been the EFTA countries’ most important trading partner (Chart 3). This is particularly true of Austria and Switzerland, which traditionally have been more closely linked to the EC core than the Nordic EFTA countries. Among the Nordic countries, Finland still trades relatively little with the EC, a reflection of traditionally extensive commercial relations with the U.S.S.R. and Sweden. Norway is the only country whose trade with the EC has generally decreased in recent years. However, in relative terms, it still exports more to the EC than any other EFTA country. In both Iceland and Sweden—the smallest and largest exporters, respectively, within the EFTA—the Community accounts for more than half of both exports and imports.

Chart 3.EFTA: Direction of Trade

(in percent of total)
(in percent of total)

Source: IMF, Direction of Trade Statistics

The trade of the EFTA countries is heavily concentrated in manufactures, which account for over four fifths of total EFTA exports (Table 1). Reflecting the dominant role of industrial products, EFTA trade is of the intra-industry rather than the inter-industry type. There are, however, important differences within the group of EFTA countries. Iceland and Norway are highly dependent on primary products. While the bulk of Iceland’s exports consist of fish products, Norway is a major exporter of crude oil and natural gas. Although these are the most obvious cases of inter-industry specialization, there are other examples as well. Largely because of the availability of low-cost hydroelectric power, a significant part of Norway’s non-oil exports consists of such energy-intensive products as aluminum. Similarly, reflecting an abundance of raw materials, Finland and Sweden are large exporters of forestry products, which account for a considerable portion of exports. However, with the exceptions of Iceland and Norway, the EFTA countries are essentially exporters of differentiated industrial products. In this respect, they are similar to the most advanced EC countries.

Table 1.EFTA: Merchandise Exports by Commodities1(in percent)
19771987
Austria
Food and beverages3.63.4
Raw materials8.65.4
Fuels1.91.8
Manufactures85.989.4
Finland
Food and beverages4.02.2
Raw materials16.412.7
Of which:
Wood8.65.4
Paper pulp5.14.6
Fuels2.12.2
Manufactures77.582.8
Of which:
Paper products22.825.8
Iceland
Food and beverages70.977.9
Of which:
Fish56.472.5
Raw materials8.03.1
Fuels
Manufactures21.119.1
Norway
Food and beverages10.07.7
Raw materials5.94.4
Fuels20.140.5
Of which:
Crude oil18.528.3
Natural gas11.9
Manufactures64.047.3
Of which:
Nonferrous metals10.69.1
Sweden
Food and beverages2.11.8
Raw materials13.99.4
Of which:
Wood4.73.5
Paper pulp5.83.7
Fuels1.82.9
Manufactures82.285.9
Of which:
Paper products9.011.7
Switzerland
Food and beverages4.03.0
Raw materials1.71.2
Fuels0.10.1
Manufactures94.295.6
EFTA
Food and beverages4.73.9
Raw materials8.96.1
Fuels3.96.9
Manufactures82.583.1
Source: OECD, Foreign Trade by Commodities.

According to SITC Rev. 2. Product classification as follows: food and beverages: 0–1; fish: 03; raw materials: 2–4; wood: 24; paper pulp: 25; fuels: 3; crude oil: 33; natural gas: 34; manufactures: 5–9; paper products: 64; and nonferrous metals: 67.

Source: OECD, Foreign Trade by Commodities.

According to SITC Rev. 2. Product classification as follows: food and beverages: 0–1; fish: 03; raw materials: 2–4; wood: 24; paper pulp: 25; fuels: 3; crude oil: 33; natural gas: 34; manufactures: 5–9; paper products: 64; and nonferrous metals: 67.

Nature of Remaining Barriers in Western Europe

European markets appear to remain fragmented despite the fact that trade in manufactures within the EC and EFTA is free from all tariffs and quantitative restrictions. The existence of large price differences for most commodities among national markets clearly supports this view (Wieser, 1989). Judging from these price differences, in general, the original six EC countries appear to constitute the most integrated area in Western Europe (Table 2). The degree of price dispersion is higher when Denmark, Ireland, and the United Kingdom are included. As perhaps can be expected, the area that includes all the current EC members and the EFTA countries appears to have achieved a much lower degree of integration than the six original EC members. The Nordic EFTA members stand out as the high-cost countries of Europe (Table 3). Of course, this may partly reflect such variables as higher real per capita incomes, higher indirect taxes, higher transport costs, and (perhaps to some extent) currency valuation. However, smaller home markets—implying both less competition and scale disadvantages—may also be important.

Table 2.Price Dispersion in the EFTA and EC in 1985(standard deviations from EC average)
EC-6EC-9EC-12EC-EFTA1
Food4.910.712.730.7
Beverages9.732.034.684.2
Tobacco21.941.543.947.2
Nontraded goods and services
Gross rent37.933.242.946.3
Medical and health services17.022.226.724.1
Public medical and health care19.020.825.523.4
Operation of transport equipment7.810.516.717.3
Transport services23.931.637.746.5
Communication35.739.138.241.6
Education, recreation, and culture15.014.126.826.9
Government consumption17.712.626.929.3
Construction7.612.517.618.5
Traded goods and services
Clothing and footwear10.814.011.623.1
Fuel and power10.813.617.619.1
Household equipment and operation5.39.513.515.3
Recreation equipment and repairs10.013.112.518.1
Medical and pharmaceutical products29.328.330.731.5
Transport and communication10.516.421.123.7
Transport equipment12.023.726.132.4
Books, magazines, and newspapers21.425.732.653.4
Business fixed investment6.212.011.513.3
Machinery and equipment9.914.68.212.9
Transport equipment19.623.523.331.9
Non-electrical equipment5.311.54.97.9
Electrical equipment6.214.09.319.1
Source: Wieser (1989).

All current EC members plus Austria, Finland, Norway, and Sweden. Data for Iceland and Switzerland are not available.

Source: Wieser (1989).

All current EC members plus Austria, Finland, Norway, and Sweden. Data for Iceland and Switzerland are not available.

Table 3.Price Levels in the EFTA in 1985(EC =100)
AustriaFinlandNorwaySweden
Household consumption112137143127
Food, beverages, and tobacco116172183172
Food116162169165
Beverages109291296255
Tobacco132133191145
Clothing116146143152
Gross rent119118128135
Household equipment and operation98134120105
Medical and health care98116114100
Medical and pharmaceutical products145132139102
Medical and health services81939881
Public medical and health care111122120104
Transport and communication115139143107
Personal transport equipment100168163102
Operation of transport equipment126124103112
Transport services105155202164
Communication11612118961
Education, recreation, and culture116128137122
Miscellaneous105141162127
Government consumption113126149143
Gross fixed capital formation98112119127
Construction96109108140
Machinery and equipment98120137108
Transport equipment106169164129
Non-electrical equipment9511412193
Electrical equipment100110121165
Gross domestic product108130135128
Source: Wieser (1989).
Source: Wieser (1989).

Although tariffs and quantitative restrictions on trade within Western Europe have largely been abolished, other obstacles remain. These can be divided roughly into two different groups. The first group includes various administrative procedures such as border controls. These give rise to trade costs in much the same way as transportation does. Unlike tariffs and quotas, which either generate government revenues or rents to private companies, border controls impose net resource costs both on governments and companies. The second group includes such government practices as public procurement rules, technical standards and regulations, barriers to entry, and public monopolies. These act as import quotas in that they tend to generate rents both to foreign and domestic firms. As argued by Baldwin (1988), the benefits of eliminating factor market barriers may well be as important as other trade liberalization measures.

Although all of these obstacles do not generate any government revenues or perhaps are not as direct as quotas, they may still be as harmful as conventional barriers. First, they entail trade costs and are therefore trade-decreasing. Second, by restraining foreign competition and preventing arbitrage between different national markets, they encourage market segmentation and effectively serve to create local markets. While barriers resulting from government regulations may be the most obvious obstacles to less-than-full integration of European markets, other elements may also contribute. The most important are private anti-competitive strategies aiming at exploiting market power and differences in buyer preferences among different markets. It is unclear whether or not the internal market program will eliminate these more subtle obstacles.

Economics of the Internal Market

The EC is seeking to remove all artificial barriers that inhibit trade within the Community. To better understand the likely effects of the internal market, it is useful to review the determinants of the pattern of international trade. Economic theory identifies three benefits of trade: it allows specialization according to comparative advantage; it results in a better use of scale economies; and it increases competition.

According to the theory of comparative advantage, gains from trade arise because countries differ in some respect. The most common explanation is that countries differ in factor endowments as in the so-called factor proportions model (Helpman, 1984). 19 However, the fact that the Western European countries are at similar levels of economic development and that trade costs are already quite low suggests that the theory of comparative advantage can explain only the relatively small fraction of total trade among these countries, which comprises products based on natural resources.20

Greater integration in Western Europe may yield important gains from further specialization based on comparative advantage. Are there any costs involved? As far as trade arrangements are concerned, the EC is at present essentially a customs union. This means that internal tariffs and quotas within the Community have been eliminated while each country has to conform to a common external trade policy. When union members reduce barriers among themselves and coordinate their external trade barriers, there are two conflicting effects: the first tends to increase welfare of the participating countries while the second tends to decrease it.21

As internal barriers within a customs union are eliminated in each country, domestic goods are substituted for by goods from its union partners. This is the trade-creation effect that tends to increase welfare. However, another effect works in the opposite direction. As goods produced within the customs union become cheaper relative to goods from the rest of the world, there is a tendency to switch demand to relatively more expensive union goods from more efficiently produced goods from outside the union.

It is in principle not possible to exclude the possibility that a customs union could cause trade-diversion effects sufficiently large to reduce welfare in the union as a whole. Also, even if the union as a whole would gain, some members may be hurt. In the case of Western Europe, the fact that conventional tariff-type barriers are already low suggests that the trade-diversion effects of further integration may be rather limited. However, a harmonization of nontariff barriers may clearly have some damaging effects.

The recognition that most trade among industrial countries is in similar but differentiated products has in recent years resulted in alternative theories that explain trade mainly in terms of scale economies and imperfect competition (Helpman and Krugman, 1985). With differentiated goods subject to increasing returns to scale and with consumer preferences for diversity, gains from trade arise that have nothing to do with comparative advantage. The basic proposition is that if each country concentrates its resources on a limited range of goods, it can produce them on a larger and more efficient scale while the variety available for domestic use is increased through imports.

The traditional analysis of trade liberalization is based on a framework similar to the neoclassical growth model in which the equilibrium growth rate is determined by the growth of labor supply and technical progress. The latter is usually assumed to be invariant to policy changes. In conventional models, because of the assumption of constant returns to scale, trade liberalization has no permanent effects on growth. If barriers to trade are lowered, output will grow faster for some time but will eventually return to its steady state growth path. This result does not necessarily hold in models with increasing returns to scale.22 In such models, trade policy may permanently affect the rate of growth of output through its effects on investment. For example, Baldwin (1989) argues that the internal market may significantly raise the equilibrium growth rate in the EC. It is thus important to distinguish between the static and the dynamic effects of trade liberalization.

With increasing returns to scale, markets will be imperfectly competitive. A common assumption is that of monopolistic competition, implying that there are many firms producing differentiated products of the same variety. However, despite their monopoly power, firms will earn only normal rates of return. Although the model of monopolistic competition is attractive in its simplicity and has provided valuable insights, most markets are probably oligopolistic in that firms take each other’s actions into account and consistently extract monopoly rents. In this case, the possibility of international trade may induce a more competitive behavior and make oligopolistic pricing strategies less likely.23

In conventional models based on perfect competition, measures that distort free trade generally tend to decrease welfare. This is, however, not necessarily the case once oligopolistic behavior is present. Beginning with Spencer and Brander (1983) and Dixit and Kyle (1985), recent theoretical studies have shown how discriminatory actions by governments—“strategic trade policy”—could be beneficial by helping firms exploit market imperfections. It remains to be seen how important the arguments of strategic trade policy really are. Nevertheless, the internal market program has some elements reminiscent of strategic trade policy. One of the most important motives behind the new initiatives has apparently been to improve the ability of EC firms to compete with firms in other countries, notably in Japan and the United States.

The existence of scale economies and imperfectly competitive industries argues for greater integration. Nonetheless, although the creation of the internal market in general can be expected to be beneficial, a formal analysis is complicated by second-best problems. In a typical imperfectly competitive industry, the elimination of internal barriers is likely to stimulate firms both to reduce prices and to raise production. This tends to increase welfare. From a general equilibrium perspective, however, the net effect on welfare may be smaller. For example, if factor prices are distorted, an expansion in one particular industry may adversely affect resource allocation elsewhere in the economy.

The expectation that the internal market will lead to improvements in economic conditions within the EC is based on all three trade-benefit principles. Trade will result in some further specialization based on comparative advantage; some increase in the scale of production in certain industries; and some increase in competition. However, although comparative advantage is important, most of the potential gains will probably derive from a better use of scale economies and increased competition in the larger united market. This is illustrated in several model simulations, for example, in Venables and Smith (1988), who studied the effects of a reduction in trade costs and an elimination of market segmentation within the framework of a partial equilibrium model in ten EC industries. In a similar exercise for Norway and Sweden, Norman (1989) reached the same conclusions.

There seems to be little doubt that the internal market program, if implemented successfully, will yield substantial gains for the Community. However, an accurate calculation of the magnitude of these gains is difficult if not impossible to achieve.24 Several theoretical and empirical considerations underlie this pessimism. In principle, the effects of the internal market could be simulated with the help of a computable general equilibrium model. The basic problem with this approach is the lack of a general theory of imperfect competition. The model builder is thus forced to make a series of arbitrary choices among a number of more or less equally plausible assumptions concerning in particular the interaction of firms in oligopolistic markets. Unfortunately, these assumptions typically prove to be crucial for the results. In addition, as soon as imperfect competition is present, second-best effects make any comparisons of alternative nonoptimal situations difficult.

There are also empirical problems. The observed price differences among Western European markets may not only reflect barriers imposed by governments but also an important element of price discrimination. It is not clear to what extent the observed price differences are the result of obstacles that the internal market program is supposed to remove. As a consequence, it is difficult to predict the degree of price convergence. Although a number of plausible arguments may explain why firms will be less able to discriminate between different national markets—in particular, it may be easier to engage in cross border arbitrage—it is unlikely that the internal market program will end all such practices. This is crucial since most of the gains from the internal market program are expected to come from an elimination of market segmentation rather than from a reduction in trade costs (Venables and Smith, 1988; and Norman, 1989). Also, to take advantage of scale economies, the number of domestic competitors in each country must decrease. This can only be compensated for by an increase in competition at the European level. There is, of course, a risk that the current stream of mergers and acquisitions in Europe may prevent such an increase from materializing.

Benefits and Costs for EFTA Countries

The EFTA countries would share in the gains from the internal market program, if they decided to and were permitted to participate fully in the EC integration process. First of all, there would be scope for further specialization along the lines of comparative advantage. A de facto integration would also neutralize the possible negative effects on the EFTA countries from trade diversion in the EC.

Since this strategy would amount to nothing less than the formation of a customs union, it would not be costless. The creation of an internal market encompassing both the EC and EFTA would require a complete coordination of trade policies. As trade between the EFTA and EC would be liberalized while barriers to trade with the rest of the world would remain, trade diversion effects could arise, which would tend to decrease welfare in the EFTA countries. The importance of these effects depends to a great extent on the future course of EC trade policy, which may be hard to predict. With notable exceptions, tariffs do not seem to differ substantially between the EC and EFTA countries (Herin, 1986; and Pintado and others, 1988). However, quantitative restrictions may be a more serious issue.25 In any case, the EC’s willingness to be more restrictive with respect to third countries—or to be more restrictive in the future—suggests that the problem of trade diversion cannot be completely dismissed.

Although a further inter-industry specialization may yield some benefits, most gains would probably derive from a better use of scale economies and increased competition. In a typical noncompetitive industry producing a differentiated good subject to increasing returns to scale, there would be both a downward pressure on prices and an expansion of production. The elimination of internal barriers within the EC and EFTA would tend to shift demand in each country from domestic sources to imports. Since total domestic demand and production would be likely to increase, this would not necessarily imply that the demand for domestic goods in domestic markets would need to decrease. On balance, however, the expansion would likely take place mainly in export markets.

How would welfare be affected? Transfers from domestic producers to domestic consumers that result from the fall in prices charged in domestic markets are welfare-neutral from the point of view of the economy as a whole. Any gains must therefore come from the effects of increased competition on domestic resource allocation, which are normally positive, or from the effects on terms of trade, which are unclear. Although it seems reasonable to assume that the positive effects would dominate, on a theoretical basis, the effects on welfare are ambiguous. Furthermore, in a general equilibrium perspective, it is even more difficult to assess the effects on welfare. As already indicated, if factor prices are distorted, an expansion of certain sectors may come at the expense of other sectors where resources could be used more efficiently.

While the effects on the EFTA countries are qualitatively the same as those on the EC, the EFTA countries may have more to gain from participating in the internal market than the EC on average. The EFTA countries have considerably smaller home markets than most countries in the EC. As is formally illustrated in model simulations by Norman (1989), this generally implies a more severe trade-off between scale advantages and competition.

If the EFTA countries would fail to integrate further with the EC, they would not be able to reap the gains accruing to EC members. Are there additional costs? It is reasonable to argue that some economic losses are involved, although their magnitude is uncertain. To determine the effects on the EFTA countries, consider first the current situation. As far as manufactures are concerned, the EC and EFTA constitute a free trade area. No tariffs or quantitative restrictions restrict the free flow of these goods within Western Europe. The remaining barriers to trade affect the EC and EFTA countries more or less equally.

The internal market program aims at the complete elimination of existing trade barriers within the EC. As a result, each country within the Community will experience a fall in import prices and an improvement in terms of trade. This will tend to shift resources away from the production of import substitutes into the production of exportables while the composition of demand will undergo a similar shift but in the opposite direction. The exploitation of scale economies and increased competition will further contribute to the expansion of demand and output.

In the nonintegration case, as trade barriers are eliminated within the EC, demand will tend to shift from EFTA products to EC products both within the EC and EFTA. As EFTA terms of trade fall, a reallocation of resources from export industries to import-competing industries will take place, resulting in a less efficient use of available resources. However, to the extent that prices in the EC would fall, possibly reflecting a better use of scale economies and a decrease in monopoly power, there would also be an offsetting tendency for EFTA terms of trade to improve. Without a detailed knowledge of how these price changes are distributed over different industries, it is difficult to calculate the net effect.

The adverse effects stemming from the reduction in demand for EFTA products are likely to be reinforced by a less efficient use of scale economies. In a typical industry operating under increasing returns to scale and imperfect competition, the more competitive environment in the EC would force EFTA firms to make both price and quantity adjustments. The reaction of EFTA firms depends, among other things, on their competitive position on foreign and domestic markets and the relative importance of these markets.26 EFTA firms probably have a stronger position in their domestic markets than in foreign markets. They may therefore partly resist lowering their prices on domestic markets or in any case lowering their prices less than their EC competitors.27 Nevertheless, in contrast to the integration case, prices and volumes of EFTA firms are both likely to decrease as barriers within the EC are eliminated. As in the integration case, the fall in prices on EFTA and EC products will boost the welfare of consumers while decreasing that of producers. However, on balance, the negative effects are likely to dominate.

Reflecting the adverse shift in demand for EFTA products, the EFTA countries’ trade balance is likely to deteriorate. With less demand for EFTA products, firms would cut back production, resulting in reduced demand for labor and higher unemployment. The only way that production and employment would be unaffected is through a decrease in production costs. This can be achieved only through an offset in labor costs. In order to maintain, employment, wages in EFTA countries would need to decline relative to wages in the EC. With such a decline, EFTA employment and production would again rise, thereby restoring demand for EFTA products through a decline in EFTA relative prices. While general equilibrium considerations may complicate the picture, we can conclude that with unchanged relative wages, the deterioration in the EFTA countries’ relative position will show up in higher unemployment and reduced competitiveness. On the other hand, if relative wages are allowed to decline, EFTA terms of trade will fall further. Both possibilities imply a reduction in EFTA purchasing power.

A failure to integrate further with the EC may also have important consequences for firms’ locational decisions. In order to avoid the deterioration in external competitiveness that is likely to take place in the nonintegration case, EFTA firms may be tempted to move some of their production facilities into the Community. This possibility may be attractive for several reasons. First, with increasing returns to scale and transport costs, it makes sense to concentrate production facilities as close as possible to the most important market (Krugman, 1988). This trade-off is particularly stringent in the case of the EFTA countries, which all possess small domestic markets. Second, establishment in the EC may help EFTA firms compete for public procurement under the same conditions as EC firms and also avoid restrictions related to rules of origin and technical standards. Third, the internal market may give rise to external effects that make establishment there more profitable. All three factors suggest that EFTA terms of trade would decline and that relative wages would have to compensate for the shortfall. Most disadvantages obviously disappear if the EFTA countries would become fully integrated with the Community. However, a complete integration is likely to reduce the costs of establishment in the EC. If the external effects from the concentration of economic activities in the EC are sufficiently strong, establishment in the EC is likely to become more attractive even with full integration.

19

Another, perhaps less likely, possibility is that countries have access to different technologies. For a survey of tests of different trade models, see Deardorff (1984).

20

As already indicated by the structure of EFTA foreign trade, products in which the EFTA countries are likely to have a comparative advantage include forestry products in the case of Finland and Sweden, fish in the case of Iceland, and energy-intensive products for Norway. For a study of comparative advantage between the EC and EFTA, see Haaland and Norman (1987).

21

The traditional literature on customs unions has concentrated on the case of trade under perfect competition and constant returns to scale. Surveys of this literature are included in Corden (1984) and Pomfret (1986). For more recent theoretical studies, see Ethier and Horn (1984); Venables (1985); and Winters (1988).

22

The existence of economy-wide increasing returns to scale may not only reflect scale advantages at the level of the individual firm but also externalities. Technological advances in one industry may, for example, have spillover effects in other industries. In addition, profit-motivated technological improvements may be as important as investment in real capital. For a discussion of the link between growth and increasing returns, see Romer (1987).

23

It is theoretically possible that a reduction of trade barriers in this case could have adverse effects on welfare. According to the so-called reciprocal dumping model by Brander and Krugman (1983), oligopolistic firms may have an incentive to discriminate between different markets and charge higher prices at home than abroad. This implies a waste of resources both because prices are different in different markets and because of the transportation costs. Although not likely, lower trade barriers may further reinforce this tendency.

24

The EC Commission (1988b) has tried to do this through a combination of judgmental methods and model simulations. The calculations were made in two steps. The first step was to estimate the direct cost savings resulting from the elimination of the existing barriers. The second step was to calculate the relationship between direct effects on the basis of a partial equilibrium model with imperfect competition and economies of scale developed by Venables and Smith (1988). This model was calibrated to data on only ten industries, which is why the exercise may provide little guidance on the overall effects. See also Flam and Horn (1989).

25

The automobile industry is a good example. Several EC countries impose quantitative restrictions on imports of Japanese cars in the form of voluntary export restraints (VERs). As a result, market shares of Japanese car producers differ substantially both within the Community and within Europe as a whole.

26

Assuming that firms are unable to discriminate between different markets, the same price will be charged in each market. This price will then be a weighted average of the hypothetical prices that would prevail in each market if price discrimination were possible.

27

If firms were able to discriminate between different markets, prices in the EFTA countries might rise while prices in the EC would decline. Since the completion of the internal market may result in reduced production of EFTA firms, average costs could rise, forcing these firms to increase their prices in EFTA markets. This may also allow EC firms to charge higher prices in EFTA markets than in the EC. The implications of different assumptions about segmentation for the effects of trade policy are discussed in Markusen and Venables (1988).

    Other Resources Citing This Publication