Jfmamjjasond J Fmamjjasond

Figure 6. Cumulative Monthly Gold Movement Statistics for Pairs of Countries from Respective Sources 1900,1907, 1928

common: when differences exist, very frequently they persist for many months in succession and often are not removed. This means that there are, apparently, at best no self-correcting forces at work making themselves swiftly felt.16 In general, even if the direction of movement should be the same, the great absolute differences make it impossible to use these statistics for purposes of determining, for reasonably short intervals, the amounts of gold shipped and, possibly, of capital transfers. They can only be used for the crudest kinds of estimates and can play no major role in the fine-structured theory of international trade.

(4) Further explorations. Since most writers on international trade are particularly concerned with net movements, we insert a chart showing the net movement of gold from the United Kingdom to Germany for the periods from 1900-1913 and 1931-1936 on a monthly basis. The statistics are made up from the two possible sources: United Kingdom figures and German figures. In Figure 7, the two lines would coincide if the data were accurate, but this is the case only in rare instances. Anyone interested in the cyclical aspects of gold movements would be in a considerable quandary in trying to determine where the cycles are, what the amplitudes are, and into which particular month the peaks and troughs of the cycles fall.

The confusion of the statistics between the two European countries is much greater for the time before the first World War than it is in the early 1930's, but it is worth recalling that this latter period coincides with a time in which exchange control was applied quite strictly by Germany and so it would be expected that more accurate records would be kept.17

18 There is a lag in several cases and there may be transit movements, as already mentioned. Elaborate statistical operations on these data may be indicated in order to discover lags, especially when more than two countries are simultaneously considered.

17 A somewhat laborious extension of our procedures suggests itself: instead of forming cumulative sums (gross or net) merely for significant pairs of countries, all countries, or at least the most important, could be taken together and an overall total of imports and of exports could be formed. It might be hoped that in this way all major discrepancies would disappear. But even if a favorable result should be obtained, it is difficult to see how the all-important country-to-country data could be corrected.

1 ..l,,l,,1.,j,,t ..l,,l ,,l ,.l,,j,,l,ll..l,.j,.lt

1900 1901 1902 1903' 1904' 1905 1906 1907 1908 1909 1910 * 1311 1 1912 1913

1931 1932 1933 1934 ' 1935 ' 1936

POSITIVE NUMBERS: UNITED KINGDOM NET GOLD EXPORTS TO GERMANY

1 ..l,,l,,1.,j,,t ..l,,l ,,l ,.l,,j,,l,ll..l,.j,.lt

1900 1901 1902 1903' 1904' 1905 1906 1907 1908 1909 1910 * 1311 1 1912 1913

1931 1932 1933 1934 ' 1935 ' 1936

POSITIVE NUMBERS: UNITED KINGDOM NET GOLD EXPORTS TO GERMANY

NEGATIVE NUMBERS; GERMAN NET GOLD EXPORTS TO UNITED KINGDOM

Figure 7. Comparison of Statistics of Monthly Net Gold Movements Between United Kingdom and Germany, 1900-1913, 1931-1936

In order to facilitate the evaluation of the information contained in Figure 7, we have cumulated quarterly the discrepancies in the reported net gold movements for both cases. Figure 8 shows the quarterly cumulative values as well as the percentage of the cumulative discrepancies. Any one point of the series indicates how much gold is cumulatively unaccounted for from the beginning of this series up to that particular point of time. From January 1900 to July 1900 the cumulative net movement was from Germany to the United Kingdom. Germany then reported a net movement £310,000 less than the United Kingdom reported. From July 1900 on, the cumulative net gold movement was in the other direction (from the United Kingdom to Germany). From January 1900 to October 1907, for example, Germany reported a cumulative net gold movement to Germany of £29,000,000 in excess of that reported by the United Kingdom. This £29,000,000 discrepancy is 156 percent of the cumulative amount reported by the United Kingdom. Any other parts of the chart may be interpreted similarly.

The reader may judge for himself whether differences of these orders of magnitude mean anything. At any rate, what has happened in these arbitrarily chosen, short time intervals may easily have happened for other countries and for other intervals. For some purposes, the discrepancies may not matter, but if they do not, then it is because economic reasoning in these fields cannot be as sharp as it is often assumed to be.

The arguments used so far in evaluating the statistics were mostly qualitative because the lack of statistical theory makes the desired quantitative expression at present impossible. But in order at least to describe the data along statistically conventional lines, the significance of algebraic differences of the quarterly statistics of gold shipments between pairs of countries is shown in Table 6. The data used here include quarterly statistics of gold movements between all possible pairs, covering trade of the United States, Germany, France, the United Kingdom, and Canada for the years 1900, 1907, 1928, and 1935. Silver transactions between the United States and the United Kingdom are also included in order to provide at least one comparison of the two precious metals.

UNITED KINGDOM-* GERM ANY

UNITED KINGDOM-* GERM ANY

Figure 8. Cumulative Discrepancies in Reported Net Gold Movements Between United Kingdom and Germany, 1900-1913, 1931-1936 (Cumulative Values and Percentages)

GERMANY—* UNITED KINGDOM

Figure 8. Cumulative Discrepancies in Reported Net Gold Movements Between United Kingdom and Germany, 1900-1913, 1931-1936 (Cumulative Values and Percentages)

Was this article helpful?

0 0

Post a comment