The European Agreement of the European Union and the agreements of the European Free Trade Association (EFTA) with many East European countries are similar to free trade agreements, but they differ in some important ways from agreements such as the North American Free Trade Agreement (NAFTA). First, agriculture is essentially left out, so that in Austria it is possible to buy Dutch greenhouse tomatoes but not fresh Hungarian tomatoes. Second, “sensitive sectors” such as steel, coal, and textiles are included, but with many safeguard clauses in case these are country-specific disturbances.

Third, there is reportedly a lot of hidden protection taking place with regard to East European exports, as was illustrated by a recent Financial Times report on German garden gnomes and Polish exports. As an industrial product, Polish gnomes are, in principle, subject to a low tariff. However, since the Polish export price is only a fraction of the German price, German producers are affected by the cheaper Polish gnomes. As a result, the Germans create a trade obstacle, claiming that the design is German and thus protected by intellectual property rights, prohibiting imports from Polish producers unless the producers are willing to pay a license fee.

Last but not least, a recent Hungarian study claims that Eastern Europe, so far, has benefited little from the European Agreement. While Hungary can be expected to have a comparative advantage in the agricultural sector, Hungary’s agricultural trade balance has deteriorated over the last few years.

What conclusions can be drawn from these examples and the experience with NAFTA? First, the Commission of the European Union is aware of the current shortcomings but, as Robert Lawrence observes about NAFTA, views trade relations as a permanent process and hopes that with continuing dialogue, a move toward a less distorted trade regime will take place (Bucher 1995). Second, there appears to be little public awareness about the domestic structural adjustment that will have to take place in Europe as a result of enhanced trade relations with the East. The scope of the adjustment is not gigantic; annual gross adjustment costs are estimated at some 0.2-1.0 percent of GDP (Holzmann, Petz, and Thimann 1994). However, the low numbers disguise the distributional effects and the need to adjust in specific sectors. Last but not least, there appears to be an unsustainable inconsistency between the Commission’s approach to trade and resistance to adjustment at the national and regional levels. To paraphrase the Minister-President of Saarland, Saarland’s economy is doing fine. The only thing it needs is protection against cheap imports of East European coal and steel.

The underlying issue may serve as a link between the presentation by Mr. Lawrence and the paper by Leonardo Bartolini and Steven Symansky. What options exist for Western Europe to assist in the growth of Eastern Europe? Both NAFTA and the European Agreement can be seen as reflecting an export-led growth strategy that gives emerging market economies access to industrial markets and allows them to imitate the experience of Southeast Asia. This option contrasts with the strategy of import-led growth and the transfer of capital to Eastern Europe to enhance its capital stock.

It is the impact of the second option on Western Europe that is investigated in the paper by Mr. Symansky and Mr. Bartolini. I will discuss the paper along four lines: First, does the paper address relevant issues? Second, does it use an appropriate model and make appropriate assumptions? Third, are the results credible, and how do they compare with other findings? And finally, what is the quality of the policy conclusions?

The issue the authors are addressing is certainly a very relevant one, but perhaps more so if seen from a different angle. They investigate the impact on West and East European economies of large capital transfers from West to East—$70 billion and $250 billion annually—for a period of ten years, amounting to a multiple of the actual transfers being made. Their results indicate that the West does not need to worry too much, because even under the high-flow scenario “most macroeconomic aggregates are likely to suffer shocks significantly smaller than would be expected from a typical business cycle.” In consequence, the low current level of capital transfers cannot be justified by the argument that the shocks resulting from expanded trade financed by larger transfers would be disruptive. But there is a political problem at the very heart of the matter, even in the case of grants of $70 billion, since such a policy has to be weighed against the costs of alternative options, such as allowing Eastern Europe unrestricted access to West European markets. If loans and not grants are considered as the source of financing for the transfers, the issue arises as to what the incentives are for Western capital owners to provide funds in view of the high indebtedness of many previously centrally planned economies (PCPEs) and of the uncertainties about progress and the results of reform. If a high-risk premium is required for compensation, the loans may not be extended, or the result for Eastern Europe may not be so positive, with negative feedback for the West. However, Mr. Bartolini and Mr. Symansky treat the issue of loan versus grant financing only superficially.

This issue leads me to the question of the appropriate model and appropriate assumptions. A dynamic multicountry model (MULTIMOD), which allows for feedbacks and which is based on a forward-looking approach, is certainly welcome. However, the stylized use of the PCPEs, modeled as a capital-import developing country, casts doubt on the capacity of the model to capture convincingly the impact of financial flows on Eastern Europe. I note the following problems:

• The specificities of the PCPEs, with an assumed inefficient use of existing factors of production but with the potential for enormous efficiency gains, may drive the financial flow effects on Eastern Europe (and hence the feedback on the West) in both ways.

• The model does not capture the impact of high capital inflows on the real exchange rate. The experience of reforming countries in Western Europe (particularly Spain), and very recent indications for the Czech Republic on capital flow-induced real appreciation, are based on much more moderate capital flows than those envisaged under the model simulation.

• The demand for Western investment goods (if all transfers are used to enhance the capital stock in the East) requires non-negligible shifts in the production pattern between consumption and investment goods in the West that are also not appropriately captured in the model (Masson, Symansky, and Meredith 1990).

• The treatment of loan versus grant financing, noted above, does not allow for differentiated interest rates.

• Last but not least, there is the issue of how grants or loans will affect the West and how any grants will be funded—that is, through a tax increase or expenditure cuts. The model seemingly does not allow for such a distinction, but the economic results of and the political support for grants and loans are likely to be different.

The results are quite different from those of other investigations, which suggest a much higher estimated impact on real interest rates and hence stronger (negative) output effects on the West (Collins and Rodrik 1991; Giustiniani, Papadia, and Porciani 1993). Since the other studies use a simpler framework or do not fully account for feedbacks, differences in methodologies may explain differences in results. However, the scope of the annual capital transfers under the high-flow scenario ($250 billion) and the limited impact on the real interest rate in Eastern Europe (some additional 1.2-1.6 percentage points) are still surprising. The explanations given in the paper—investment and consumption, as specified in the model, respond to current as well as to future changes in the interest rate—are theoretically convincing, but the assumed speed of adjustment and the scope leave doubts. The figure of $250 billion is roughly the magnitude of the U.S. budget deficit, almost twice the size of the annual transfers from West to East Germany, and more than twice the value of the net capital transfers Asia and Latin America received at the peak of their capital inflows in the early 1990s.

Against this background, how are the policy conclusions of the paper to be assessed? I would consider them problematic and to some extent dangerous.

• Perhaps Western Europe has little to fear in terms of macroeconomic disturbances caused by increased interest rates, real exchange rates, and price levels, but the distributional problem of how to finance grants, accommodate the required structural changes in the West, and make sure that the flows are used not for consumption but for capital accumulation still remain. Yet Mr. Bartolini and Mr. Symansky claim that the effects on the West differ little whether the flows are grants or loans or are used for consumption or investment.

• Their conclusion about the small macroeconomic impact on the West hinges importantly on the responsiveness of Western saving to changes in the long-term real interest rate. Less short-term flexibility would substantially alter the results and conclusions.

• I concur with the authors that over the short and medium term, Western Europe is unlikely to suffer more than marginally from the projected flow of financial assistance to PCPEs, because the flow is low and unlikely to increase dramatically.

In summary, I think the paper is useful, addressing an important issue and applying a relevant methodology. Yet the model as it is currently specified can be considered only a first approach.