英语论文网

y specialise in production of the good in which they have comparative advantage, leading to gains from trade. Bleaney (1993) and Athukorla (1993) criticised Sarkar-Singer (1991) theory and published their own, separate, comments and adjustments to Sarkar-Singer hypothesis. Bleaney (1993) looks at three points of the hypothesis, firstly, “A unit value index of manufactured exports (MXUV) of all developing countries indicates a trend decline of 1.00% p.a. over 1970-87 relative to a similar index for advanced countries”. Secondly “In a sample of 29 developing countries over 1965-85, the median country showed no trend decline in the ratio of unit values of manufactured exports to manufactured imports.” Finally “Relative to the MXUV of the United States, the MXUVs of individual developing countries tended to show a decline over 1965-85, with a median rate of decrease of about 2% p.a.” Bleaney’s (1993) criticism was that he believed the results to be “somewhat ambiguous” and that there is an inconsistency between the second and third points, which may well due to the differing behaviour of the United States and the world unit value index of manufactured exports. Bleaney shows the ratio of the world unit value of manufactured exports rising from 114 (1980 = 100) in 1965 to 124 in 196, falling to 100 in 1973 and remaining at this level until 190, before rising to 146 in 1985. The particularly high level of this ration at the end of the Sarkar-Singer sample ratio means that the third result Bleaney (1993) criticises, which uses the US index as a denominator, may be subject to a downward bias and should be treated with caution. Bleaney also feels that the Sarkar-Singer results should be interpreted as an expression of real exchange rate behaviour. Bleaney compares the unweighted average real effective exchange rate of 10 developing countries from the Sarkar-Singer sample with the ratio of developing country to developed country unit value of manufactured exports over 1978-90. The result suggests that a simple regression on a time trend is likely to reveal a much slower rate of decline if the sample ended in 1985, as in the first and second points that Bleaney raises, that if it includes post-1985 data, as in the first point. Another issue that Bleaney discusses is that for the second point, the relative unit value of manufactured exports of counties classed by the International Monetary Fund as Highly Indebted Countries have a mean trend of -1.5%, compared with 0.2% for the countries not mentioned in Sarkar-Singer’s Table 4. If there is a genuine difference in the trends in the unit value manufactured exports between the two groups, it may reflect differences in real exchange rate behaviour.table 4.bmp

Sarkar and Singer (1993) reply to the criticisms proposed by Bleaney, who “seeks to undermine” the Sarkar-Singer results by pointing out that the end of the observation period is marked by the debt crisis and real devaluation of the developed countries’ currencies. Sarkar and Singer then created two separate regressions for the sub-periods, 1970-82 and 1983-89. The results of the 1970-82 period is still a 1% decline per annum, but for 1983-89, in the series of unit value ration measured in terms of US$, shows a significant acceleration. The trend decline in the series measured in special drawing right shows no statistically significant sign of acceleration of deceleration during the same period (at the 5% level). Sarkar-Singer felt t