The 1961 Revaluations and Exports of Manufactures
Author:
Erich Spitäller https://isni.org/isni/0000000404811396 International Monetary Fund

Search for other papers by Erich Spitäller in
Current site
Google Scholar
Close

It is frequently argued that small changes in exchange rates tend to have negligible effects on the balance of trade. This argument is difficult to support on theoretical grounds, and it is likely that the question could be resolved only by empirical work. An interesting illustration of a small change in the exchange rate is provided by the revaluations, by 5 per cent, of the Netherlands guilder and the deutsche mark in 1961. This paper evaluates the effects of these revaluations on the exports of manufactured goods of the two revaluing countries and estimates implicit elasticities of substitution of these exports for those of competing countries. Some indication of the reliability of these elasticity estimates is then obtained by comparing them with other estimates reported in the literature, by examining the consistency of the Dutch and German results, and by determining the extent of agreement in the estimates for different regional markets.

Abstract

It is frequently argued that small changes in exchange rates tend to have negligible effects on the balance of trade. This argument is difficult to support on theoretical grounds, and it is likely that the question could be resolved only by empirical work. An interesting illustration of a small change in the exchange rate is provided by the revaluations, by 5 per cent, of the Netherlands guilder and the deutsche mark in 1961. This paper evaluates the effects of these revaluations on the exports of manufactured goods of the two revaluing countries and estimates implicit elasticities of substitution of these exports for those of competing countries. Some indication of the reliability of these elasticity estimates is then obtained by comparing them with other estimates reported in the literature, by examining the consistency of the Dutch and German results, and by determining the extent of agreement in the estimates for different regional markets.

It is frequently argued that small changes in exchange rates tend to have negligible effects on the balance of trade. This argument is difficult to support on theoretical grounds, and it is likely that the question could be resolved only by empirical work. An interesting illustration of a small change in the exchange rate is provided by the revaluations, by 5 per cent, of the Netherlands guilder and the deutsche mark in 1961. This paper evaluates the effects of these revaluations on the exports of manufactured goods of the two revaluing countries and estimates implicit elasticities of substitution of these exports for those of competing countries. Some indication of the reliability of these elasticity estimates is then obtained by comparing them with other estimates reported in the literature, by examining the consistency of the Dutch and German results, and by determining the extent of agreement in the estimates for different regional markets.

The effects of these exchange rate changes on imports will not be examined in this paper. Price effects on imports are difficult to isolate because changes in other variables, such as income, capacity utilization, and inventory changes, tend to overwhelm the effects of small price changes.1

Similar problems of disentangling price and income effects need not arise with respect to the assessment of the effects of a revaluation on exports. These effects can be inferred from an analysis of market shares, which for a relatively homogeneous subgroup like manufactures are not very sensitive to variation in import demand caused by changes in economic activity in the various markets.2 Specifically, it was assumed that in the absence of relative-price changes Germany and the Netherlands would have maintained their market shares at the level observed for the years before revaluation, modified by any longer-run trend in these shares that was manifest over the period before the revaluation. The difference between actual exports and exports calculated on the assumption of these market shares is taken to be the effect of the observed change in relative prices. These computed effects were then used together with the changes in relative export prices to derive implied elasticities of substitution. In order to reduce the influence of random variations, two-year averages of the relevant data were employed in the analysis. Averages of exports and prices for the years 1959 and 1960 were taken as the observed values “before revaluation” and similar averages for the years 1961 and 1962 as the observations “after revaluation”; since the revaluation occurred in March 1961, averages of the data for 1962 and 1963 were used as alternative “postrevaluation” observations.

Revaluation and Relative Export Prices

Relative export prices are defined as the ratio of the export unit values of manufactures in the revaluing country to the weighted average of the export unit values of its competitors, all expressed in U.S. dollars. The competitors’ shares in any given market are used as weights.3 Variations in these relative export prices reflect changes of export prices in local currency in the revaluing countries and in countries competing with them in international markets as well as changes in the exchange rates.4

Percentage changes from 1959–60 to 1961–62 in Dutch and German relative export prices in various specified markets are shown in the third column of Tables 1 and 2. Although both countries revalued by the same amount and at the same time, the change in relative-price competitiveness of the Netherlands was smaller throughout than that of Germany. This difference arises in part because Germany is a major competitor of the Netherlands, whereas the Netherlands is a relatively minor competitor of Germany.5 The German revaluation has thus a much larger weight in Dutch relative export prices than the Dutch revaluation has in the relative export prices of Germany.

Table 1.

Netherlands: Revaluation Effects and Apparent Elasticities of Substitution of Manufactured Exports for Competitors’ Exports with Respect to Changes in Relative Export Prices

article image
article image
Sources: Based on Tables 3 and 4 of this article and Helen B. Junz and Rudolf R. Rhomberg, “Prices and Export Performance of Industrial Countries,” Staff Papers, Vol. XII (1965), Tables 8 and 9, pp. 266–69.

The revaluation effect is taken to be the difference between hypothetical and actual export volumes as a percentage of hypothetical export volumes. A positive sign of the effect implies a reduction in exports. Hypothetical export volumes recorded in Tables 3 and 4, columns 4 and 5, were computed on the basis of average shares of the revaluing country in the exports of 11 industrial countries to the specified markets over the two years preceding the revaluations and the 11 countries’ actual exports to the specified markets in 1961–62 and 1962–63. Where trends in shares exist in composite markets, adjustments for trend were made.

Relative export prices are represented by the ratio of the export unit values of manufactures in the revaluing country to the weighted average of the export unit values of its competitors. The competitors’ shares in any given market are used as weights.

The volume elasticity of substitution is obtained by dividing the percentage change in the ratio of the revaluing country’s exports to the sum of the exports of its competitors in the sample—for composite markets that contain the revaluing country, exports to the revaluing country are excluded—by the percentage change in the relative export price. The percentage changes in export ratios were computed from the revaluation effects recorded in the first two columns of the table.

Table 2.

Germany: Revaluation Effects and Apparent Elasticities of Substitution of Manufactured Exports for Competitors’ Exports with Respect to Changes in Relative Export Prices

article image
article image
Sources: Based on Tables 3 and 4 of this article and Helen B. Junz and Rudolf R. Rhomberg, “Prices and Export Performance of Industrial Countries,” Staff Papers, Vol. XII (1965), Tables 8 and 9, pp. 266–69.

The revaluation effect is taken to be the difference between hypothetical and actual export volumes as a percentage of hypothetical export volumes. A positive sign of the effect implies a reduction in exports. Hypothetical export volumes recorded in Tables 3 and 4, columns 4 and 5, were computed on the basis of average shares of the revaluing country in the exports of 11 industrial countries to the specified markets over the two years preceding the revaluations and the 11 countries’ actual exports to the specified markets in 1961–62 and 1962–63. Where trends in shares exist in composite markets, adjustments for trend were made.

Relative export prices are represented by the ratio of the export unit values of manufactures in the revaluing country to the weighted average of the export unit values of its competitors. The competitors’ shares in any given market are used as weights.

The volume elasticity of substitution is obtained by dividing the percentage change in the ratio of the revaluing country’s exports to the sum of the exports of its competitors in the sample—for composite markets that contain the revaluing country, exports to the revaluing country are excluded—by the percentage change in the relative export price. The percentage changes in export ratios were computed from the revaluation effects recorded in the first two columns of the table.

Changes in the relative prices of both Dutch and German exports are highest in the U.S. market. This can be explained in terms of the increase in the U.S. export unit values by 2 per cent between 1959–60 and 1961–626 and the decrease in the export unit values of other competitors in that period.7 The U.S. price increase carries no weight in the U.S. market but has considerable weight in markets where the United States is a major competitor—hence, the comparatively small relative-price increase in those markets.

Since 1961 is not entirely a postrevaluation year, and since price changes may take some time to become fully effective, the average level of relative export prices of the Netherlands and Germany over the years 1962–63 might be more representative of Dutch and German competitiveness after the revaluation than the average level of these prices over the years 1961–62. The change in relative export prices of the revaluing countries was therefore alternatively computed from 1959–60 to 1962–63 (column 4 of Tables 1 and 2).

Any tendency for the effect of revaluation to be increasingly offset over time by export price changes in local currency, in the revaluing country or abroad, would result in a somewhat smaller relative-price change from 1959–60 to 1962–63 than from that base period to 1961–62. However, when the change in relative export prices is measured over this somewhat longer period it could also be affected to a somewhat greater extent by price changes that are not related to the revaluation and that could make the observed price change either smaller or larger than that for the shorter period 1959–60 to 1961–62.

In the Netherlands the rise of relative export prices in composite markets between the base period (1959–60) and the average for 1962–63 was smaller by between one third and one percentage point than the increase from the base period to 1961–62. This seems to have been due primarily to the fact that, although the revaluation could not contain the rise in the general price level, Dutch export prices declined. Although a similar continuation of inflationary tendencies was experienced in Germany after the revaluation, the level of export prices during the years 1962–63 was the same as during the years 1961–62. The generally larger rise in Germany’s relative export prices over the somewhat longer period than over the shorter period cannot therefore be attributed to export price developments in the revaluing country and appears to be the result of improvements in competitiveness abroad.

Market Shares and Export Losses

In order to assess the effects of the revaluations on the export performance of the revaluing countries, it is necessary to estimate the volume of exports that would have been realized in the absence of the relative-price changes that are associated with the change in the exchange rates. These hypothetical exports are calculated on the assumption that the market shares of the two revaluing countries would have changed, from the years immediately preceding revaluation to the years following it, in accordance with the longer-run trend that these shares appear to have followed up to the time of revaluation. Such a trend, where it exists, could be the consequence of gradual changes in competitiveness resulting from factors other than recorded price changes or from a longer-run trend in the commodity composition of world demand that may be either favorable or unfavorable to the exports of a particular country.8 Trend movements in such factors were assumed to continue to affect the market shares of the revaluing countries after the revaluation.

As a first step, the market shares of the Netherlands and Germany in the aggregate volume of exports of manufactures of 11 industrial countries to selected markets were calculated for the period 1959–60 (column 1 of Tables 3 and 4). The 11 exporting countries are Belgium, the Netherlands, Germany, France, Italy, the United Kingdom, Sweden, Austria, Canada, the United States, and Japan. The markets considered are those of the 11 exporting countries themselves, Switzerland, and the rest of the world. On the basis of these shares and actual exports of the 11 countries in 1961–62 and 1962–63 (columns 2 and 3 of Tables 3 and 4), hypothetical Dutch and German exports for 1961–62 and 1962–63 were computed on the assumption that the 1959–60 shares had remained unchanged (columns 4 and 5 of Tables 3 and 4). Hypothetical exports to three composite markets are then obtained by summing hypothetical exports to the component countries. This method of computing hypothetical exports to a composite market takes account of differences in growth rates among the various component country markets. The composite markets are (1) the 12 industrial countries included as separate markets in the study; (2) the members of the European Economic Community (EEC); and (3) the 7 non-EEC countries included under (1). Moreover, the results for the composite market of the 12 industrial countries and those for the rest of the world are summed to obtain hypothetical exports to the world market. Insofar as trends in shares exist in composite markets, the assumption of constancy in shares in the absence of relative-price changes was then relaxed. Trend equations were fitted over the period 1953–60 for Dutch and German shares in the composite markets “Total World,” “EEC,” and “Industrial Non-EEC.” While the results for the Netherlands were in no case significant, Germany’s shares show in all three groups a statistically significant rising linear trend.9 Accordingly, Germany’s hypothetical exports to the composite markets were adjusted so as to reflect continuation of the observed rising trend in market shares, while no such adjustment was made for the Netherlands.

Observed exports in 1961–62 and 1962–63 to the markets specified are shown in columns 6 and 7 of Tables 3 and 4. The presumed effect of revaluation on the volume of exports is computed as the difference between hypothetical and actual export volumes, expressed as a percentage of the hypothetical export volume. These effects are shown for the Netherlands in columns 1 and 2 of Table 1 and for Germany in the corresponding columns of Table 2. A positive sign implies a reduction in exports. The negative sign found in some instances may indicate that the effect of revaluation was in particular markets more than offset by other influences operating simultaneously. However, individual markets were not examined for trends in export shares, and the assumption of constant shares in the absence of relative-price changes may well be even less appropriate for individual country markets than it would be for the composite market areas examined in this study.

The results show that the change in competitiveness following the revaluation appears to have led to losses in Dutch and German manufactured exports in most markets, including all composite markets. In the composite markets, Germany’s losses ranged from 8½ per cent to almost 15 per cent for the period 1961–62 and slightly higher, from 9½ per cent to 22½ per cent, for the period 1962–63. The smallest losses appear to have been sustained in the EEC market, the largest ones in the market of the rest of the world. The losses computed for the Netherlands were somewhat smaller, ranging from not quite 3 per cent for both 1961–62 and 1962–63 to per cent for 1961–62 and slightly less than 9 per cent for 1962–63. In contrast to Germany, the Netherlands sustained the smallest losses in the industrial non-EEC market and the largest ones in the EEC market (1961–62) and the market of the rest of the world (1962–63).

Apparent Elasticities of Substitution

Dutch and German elasticities of substitution in the various composite markets were obtained by transforming the calculated revaluation effects into percentage changes in the ratios of the exports of the revaluing countries to the sum of the exports of their competitors in the sample,10 which were then divided by the percentage changes in relative export prices. For Germany, derived elasticities in corresponding markets are consistently larger over the longer period (1959–60 to 1962–63) than over the shorter period (1959–60 to 1961–62). While this is also true for the Netherlands in the markets of the industrial non-EEC countries, the rest of the world, and total world, Dutch elasticities in the EEC market and the market of the 12 industrial countries are somewhat lower over the longer period than over the shorter period. The decline in Dutch elasticities for these two markets appears to reflect the decline in corresponding export losses, which can be attributed largely to the substantial lessening of the apparent export loss in the German market for 1962–63, compared with the apparent loss for 1961–62. With the exception of the non-EEC industrial market, Dutch elasticities over the shorter period range from – 1.6 in the market of the rest of the world to —3.0 in the EEC market and over the longer period from —1.6 in the market composed of the 12 industrial countries to —3.2 in the market of the rest of the world. The exceptionally low elasticities in the non-EEC industrial market reflect the apparent perverse reaction to the revaluation of Dutch exports to Canada, the United Kingdom, and Japan. Estimates of substitution elasticities of Germany in the composite markets show greater consistency than corresponding estimates for the Netherlands, ranging for the period 1959–60 to 1961–62 from -2.0 for the market composed of the 12 industrial countries to —2.9 for the market of the rest of the world and from —2.3 to —4.3 for the same markets over the slightly longer period. The differences between elasticity estimates for the 2 countries are fairly small in most markets. In the market of the 12 industrial countries the elasticities are —1.8 for the Netherlands and -2.0 for Germany for the period from 1959–60 to 1961–62, and -1.6 and —2.3, respectively, from 1959–60 to 1962–63. In the world market the estimates are —1.8 for the Netherlands and —2.4 for Germany for the period 1959–60 to 1961–62 and -2.0 and -3.1, respectively, for the period 1959–60 to 1962–63.

The results obtained in this study contradict the findings of a study by Sherman that attempted to examine the effects on exports of the revaluation in Germany and concluded that no such effects could be found.11 In that study, the apparent ineffectiveness of the revaluation was attributed to a number of factors: first, to the alignment of individual domestic prices for steel within the European Coal and Steel Community to the lowest prevailing level, which was not the German domestic price level; second, to the different regional orientation of competitors’ exports of machinery; third, to the concurrent lowering of German export prices in the chemical and automobile industries; and fourth, to the European consumer goods boom that affected particularly textiles and clothing. While this apparent ineffectiveness of the revaluation on Germany’s exports was also derived from an analysis of market shares, these shares were not adjusted for trend. Moreover, composite markets were treated as if they were homogeneous regional markets characterized by a uniform growth rate. In this particular instance, this treatment results in lower hypothetical exports of Germany in the absence of revaluation and thus in a smaller apparent effect of revaluation. These factors by themselves could account for the finding of that study, in contrast to the present analysis, that the German revaluation did not have an identifiable effect on German exports.

The estimates of Dutch and German elasticities recorded in the tables do not conflict with substitution elasticities estimated in time-series and cross-section studies. Fleming and Tsiang derived an elasticity of substitution of —1.7 for the manufactured exports of 10 industrial countries for the period 1948–53.12 Kreinin found an elasticity of substitution of —2.6 from cross-section analysis of manufactured exports of 10 industrial countries for the years 1955–57.13 Harberger estimated elasticities of substitution of exports of manufactures of Germany for the exports of, alternatively, the United States (—2.4), the United Kingdom (—1.6), and France (—1.7).14 From an analysis of time series covering the years 1953–63, Junz and Rhomberg obtained substitution elasticities for Germany’s manufactured exports of —2.2 in the world market, —2.0 in the industrial market, —2.6 in the EEC market, —2.4 in the non-EEC industrial market, and —2.8 for the market composed of all other countries; (corresponding results for the Netherlands were not statistically significant).15 Since the sample of exporting countries used in the present study is the same as that used by Junz and Rhomberg, and since the composite markets in the two studies are also identical, the results recorded in Tables 1 and 2 correspond most closely to the results obtained by Junz and Rhomberg.

The consistency in Dutch and German elasticity estimates and the considerable agreement in these estimates for different regional markets as well as the consistency of these estimates with substitution elasticities estimated in previous studies justifies the conclusion that the revaluations had a clearly identifiable effect, corresponding to an average elasticity of substitution of approximately —2 over an average adjustment period of two to three years, on the export competitiveness of the revaluing countries.

STATISTICAL APPENDIX

Table 3.

Netherlands: Actual and Hypothetical Exports of Manufactures in Volume Terms

article image
Sources: Based on Organization for Economic Cooperation and Development, Statistical Bulletins, Foreign Trade Series C.

The 11 countries are the individual countries under the heading “Market” with the exception of Switzerland, for which no data were available.

Table 4.

Germany: Actual and Hypothetical Exports of Manufactures in Volume Terms

article image
Sources: Based on Organization for Economic Cooperation and Development, Statistical Bulletins, Foreign Trade Series C.

The 11 countries are the individual countries under the heading “Market” with the exception of Switzerland, for which no data were available.

The volume of hypothetical German exports to composite markets is adjusted for the rising linear trend in the German shares in the volume of manufactured exports from the 11 countries to these markets.

Les réévaluations de 1961 et les exportations de produits manufacturés

Résumé

On affirme souvent que de faibles modifications des taux de change ont généralement des effets négligeables sur la balance commerciale. Il est difficile de soutenir cette thèse en se fondant sur la théorie, et il est probable que la question ne pourrait être tranchée qu’en utilisant la méthode empirique. L’auteur de cette étude fait une tentative dans cette direction en examinant les effets à moyen terme de la réévaluation de 5 pour cent du florin néerlandais et du deutsche Mark, en 1961, sur les exportations de produits manufacturés des Pays-Bas et de la République fédérale d’Allemagne à destination de certains marchés déterminés, et en évaluant les élasticités de substitution implicites entre les exportations de produits manufacturés de ces deux pays et celles des pays concurrents.

Les répercussions de la réévaluation monétaire sur les exportations ont été calculées comme étant la différence entre le volume effectif des exportations après la réévaluation (en 1961–62 et, ou en 1962–63 selon le cas) et le volume des exportations qui eût été atteint au cours de ces deux années si la réévaluation n’avait pas eu lieu. Ces exportations hypothétiques ont été évaluées en supposant que les prix relatifs n’ayant pas été modifiés, l’Allemagne et les Pays-Bas auraient, pendant la période qui a suivi la réévaluation, maintenu leur part du marché au niveau enregistré avant la réévaluation (1959–60), compte tenu des variations de plus longue durée, qui se serait manifestée de 1953 à 1960. L’élasticité apparente de substitution peut être établie en fonction d’une part des effets calculés des réévaluations sur le rapport entre le volume des exportations des pays qui ont réévalué leur monnaie et le volume des exportations des pays concurrents et d’autre part des modifications du rapport entre les prix à l’exportation (valeurs unitaires) des pays qui ont réévalué leur monnaie et la moyenne pondérée des prix des pays concurrents, les uns et les autres étant exprimés en dollars E.U.

De 1959–60 à 1962–63, ces élasticités de substitution sur le marché de tous les pays industriels se sont élevées à environ —2,3 pour l’Allemagne et à —1,6 pour les Pays-Bas; sur l’ensemble du marché mondial, ces pourcentages sont un peu supérieurs ( — 3,1 pour l’Allemagne et —2,0 pour les Pays-Bas). Les mêmes calculs appliqués à des périodes un peu plus courtes (1959–60 à 1961–62) ont donné des élasticités de 15 à 30 pour cent inférieures pour l’Allemagne, alors que pour les Pays-Bas, elles étaient d’environ 15 pour cent supérieures sur le marché des pays industriels, et inférieures de 10 pour cent sur le marché mondial.

Il semble justifié de conclure que les réévaluations ont eu un effet identifiable pendant une période moyenne d’ajustement de deux à trois ans sur la position concurrentielle des exportations des pays qui ont réévalué leur monnaie. L’ampleur de cet effet correspond aux conclusions des études de séries chronologiques sur l’élasticité de substitution des exportations de produits manufacturés.

Las revaluaciones de 1961 y las exportaciones de productos manufacturados

Resumen

Con frecuencia se aduce que las modificaciones reducidas en los tipos de cambio suelen ejercer efectos insignificantes en la balanza comercial. Este argumento es difícil de corroborar con razonamientos puramente teóricos, y es muy posible que esta cuestión se pueda resolver únicamente mediante estudios empíricos. En este trabajo se intenta hacer algo en ese sentido mediante el examen de los efectos a plazo medio que las revaluaciones del 5 por ciento del florín holandés y del marco alemán, en 1961, han ejercido en las exportaciones de productos manufacturados efectuadas por Holanda y Alemania con destino a determinados mercados y, asimismo, mediante el cálculo de las elasticidades implícitas de sustitución de dichas exportaciones por las exportaciones de los países competidores.

Los efectos que ejerce la revaluación de las monedas en las exportaciones se calcularon como la diferencia entre el volumen real de las exportaciones después de la revaluación (1961–62 y, alternativamente, 1962–63) y el volumen que las exportaciones hubieran alcanzado durante estos años en caso de que no se hubieran revaluado esas monedas. Estas exportaciones hipotéticas se calcularon partiendo del supuesto de que si los precios relativos no hubieran variado, la participación de Alemania y Holanda en los mercados durante el período posterior a la revaluación se hubiera mantenido al mismo nivel que se observó antes de la revaluación (1959–60), modificada por cualquier tendencia evidente que hubieran registrado a la larga estas participaciones durante el período de 1953–60. Los efectos calculados de las revaluaciones sobre el volumen de las exportaciones de los países que revaluaron, en comparación con el de los países competidores, pueden utilizarse conjuntamente con las variaciones en la razón entre los precios de exportación (valores unitarios) de los países que revaluaron y la media ponderada de los precios de los países competidores—expresados todos en dólares de EE.UU.—para derivar las posibles elasticidades de sustitución.

Se observó que, durante el período de 1959–60 a 1962–63, estas elasticidades de sustitución en el mercado de todos los países industriales habían sido aproximadamente de —2,3 en el caso de Alemania y de — 1,6 en el de Holanda; en el mercado mundial conjuntamente considerado fueron algo mayores ( — 3,1 en el caso de Alemania y —2,0 en el de Holanda). En cálculos semejantes correspondientes a un lapso ligeramente más breve—de 1959–60 a 1961–62—se observó que estas elasticidades eran de un 15 a un 30 por ciento menores en el caso de Alemania, mientras que en el de Holanda fueron alrededor de un 15 por ciento mayores en el mercado de los países industriales y de un 10 por ciento menores en el mercado mundial.

Parece, pues, justificarse la conclusión de que las revaluaciones produjeron un efecto perceptible durante un período medio de ajuste de dos a tres años en la capacidad competitiva de las exportaciones de los países que revaluaron sus monedas. La magnitud de este efecto está en consonancia con los resultados obtenidos a partir de estudios de series cronológicas acerca de la elasticidad de sustitución de las exportaciones de productos manufacturados.

*

Mr. Spitäller, economist in the Research Department, is a graduate of the University of Graz, Austria, and of the School of Advanced International Studies of the Johns Hopkins University, Washington, D.C.

1

See, however, an apparently reliable estimate of the price elasticity of German demand for imports of manufactures of —2.6 in a study by Edwin von Böventer, “Der Einfluss der Aufwertung auf die deutsche Zahlungsbilanz: Eine quantitative Untersuchung,” Weltwirtschaftliches Archiv, Vol. 87, No. 2, 1961, pp. 54–92. In a study on the effects of the German revaluation, Werner Gatz argues that the revaluation had little effect on imports because the reduction in the deutsche mark price of imports was not passed on. In the face of rising domestic prices, unchanged deutsche mark prices of imports were sufficient to maintain the competitive position of imports. See Werner Gatz, “Gründe und volkswirtschaftliche Wirkungen der D-Mark-Aufwertung,” Weltwirtschaftliches Archiv, Vol. 90, No. 1, 1963, pp. 379–432.

In a model of the Dutch economy constructed at the Central Planning Bureau of the Netherlands, imports are a function of total sales, the change in stocks, agricultural production, and autonomous merchandise imports. No relative price variable is included. See C.A. van den Beld, “A Macro Model for the Dutch Economy,” Section 3 in Model Building in Economics and Industry, C-E-I-R, Incorporated, and C-E-I-R, Limited (1968).

2

See J.J. Polak, “Note on the Measurement of Elasticity of Substitution in International Trade,” The Review of Economics and Statistics, Vol. XXXII (1950), pp. 16–20.

3

See Helen B. Junz and Rudolf R. Rhomberg, “Prices and Export Performance of Industrial Countries, 1953–63,” Staff Papers, Vol. XII (1965), pp. 268–69.

4

The approach then is an ex post one in that actual price changes after the revaluation are used. These price changes, in other words, are attributable partly to the revaluation and partly to autonomous factors. It is in fact difficult to separate these influences, and as a result there could be some bias in the estimation of export effects from the revaluation alone.

5

Compare, e.g., Dutch and German price changes in the Austrian market, where the difference is particularly large.

6

See Junz and Rhomberg, op. cit., Table 8, p. 267.

7

Ibid. The export unit values of Italy and Japan, for example, decreased by 4 per cent and 5 per cent, respectively.

8

For an enumeration of factors underlying competitiveness, see, for example, J.M. Fleming and S.C. Tsiang, “Changes in Competitive Strength and Export Shares of Major Industrial Countries,” Staff Papers, Vol. V (1956), pp. 218–48.

9

The equations showing linear trends over the period 1959–60 in Germany’s volume shares in three composite markets—(1) Total World, (2) EEC, and (3) Industrial non-EEC—are as follows (t ratios are shown in parentheses below the coefficients):

(1)SW=0.14(32.84)+0.0107t(13.67)R¯2=0.96
(2)SEEC=0.28(28.62)+0.0108t(5.65)R¯2=0.82
(3)SINEEC=0.14(15.50)+0.0105t(6.02)R¯2=0.83

To the extent that these rising trends in market shares could be attributed to a declining trend in prices over the period tested, they cannot be considered to have been caused by nonprice factors alone. When German relative export prices in these three composite markets are examined for downward trend, results show such a trend to be significant only in the market “Industrial non-EEC”:

PINEEC=0.99(81.14)0.006t(2.48)R¯2=0.42

On any plausible assumption with respect to the long-run elasticity of substitution, this declining trend in Germany’s relative export prices could account for only a small part of the observed rise in its market share in the industrial non-EEC area. The assumption that the rising trend in German shares could be attributed to nonprice factors was therefore retained.

10

For composite markets that contain the revaluing country, exports of competitors to the revaluing country are excluded.

11

Heidemarie Sherman, “The Effect of the Revaluation on Germany’s Exports,” Economia Internazionale, Vol. XVII (1964), pp. 721–40.

12

See Fleming and Tsiang, op. cit., pp. 218–44.

13

Mordechai E. Kreinin, ‘Price Elasticities in International Trade,” The Review of Economics and Statistics, Vol. XLIX (1967), p. 513, Table 2.

14

Arnold C. Harberger, “Some Evidence on the International Price Mechanism,” The Journal of Political Economy, Vol. LXV (1957), p. 516, Table 3.

15

Junz and Rhomberg, op. cit., Table 3, p. 240, and section B of Table 5, p. 246.

  • Collapse
  • Expand