ON BRITISH AND FOREIGN STATISTICS OF INTERNATIONAL TRADE. 311 
in the returns to account for this, for the differences in methods of tabula- 
tion necessarily cause very great discrepancies. Take the case of England 
and Belgium. We valued our exports of home produce exported to 
Belgium at 11,000,000/. sterling, and of foreign produce at 4,000,000/. 
in 1900. Belgium valued what she received at 300,000,000 francs, say 
12,000,000/., in her special trade. Of the goods we send to Belgium, 
a great part would no doubt pass on to other countries, and in spite of 
regulations much of this would be entered as going to Belgium only. 
Indeed, the English exporter might not know its destination beyond an 
agent at Antwerp. On the other hand, the Belgian importer would credit 
part of the 4,000,000/. which we described as foreign produce to the 
United Kingdom, and only part to the real country of origin. Such 
illustrations might be multiplied indefinitely. The reader is referred to 
the papers just named for analyses of special cases. 
In addition to questions of destination, the method of valuation causes 
differences, especially where official values linger behind market values. 
Again, goods leaving X for Y may not be delivered in Y in the 
same year, or they may remain in the customs warehouse of Y for many 
months ; exact correspondence year by year is therefore not to be ex- 
pected. It follows that we do not know the value of our export trade, 
even if we assume the statements to be made correctly, to any of those 
countries where transit trade is of importance, and even when we group 
them together, there is much uncertainty ;! nor can we use the statistics 
of foreign countries to check our own. Similar remarks apply to imports. 
Further, we cannot estimate the total international trade of the world, 
or of the main groups of countries, for neither the general trade nor the 
special trade is defined in the same way in different countries, and the 
totals cannot therefore correctly be added. 
In the previous paragraphs it is assumed that the returns of quantities 
and values are correctly made. Below, we show reason to think that this 
is not invariably the case in the United Kingdom. We have no means 
of investigating the accuracy of the statistics of foreign countries in 
general ; but the process of valuation seems very faulty where official 
values are employed, and declared values are subject to bias, especially as 
they are found for the most part in the case of dutiable goods. The following 
comparison of the returns of trade between Europe and the United States 
shows how far we can obtain agreement in a case where, if the returns 
were accurate as to value and destination, there should apparently be close 
correspondence. ‘This instance is taken because the basis of valuation is 
very nearly the same in the exporting and importing countries, for freight 
is not included in either. 
We put all Europe, so far as figures are available, together in order, 
to avoid confusion of origin. The visible sources of error are the inclu- 
sion of Italian exports to Canada, which cannot be separated, and possibly 
of some Asiatic produce going from Russian Asiatic ports ; some goods 
registered in Europe as going to the United States may be there treated 
as in transit for Canada or elsewhere ; while some European trade to 
Canada may find its way into the United States ; and in the table is 
included 666,000,000 (for the ten years) value of foreign goods re- 
exported by us, counted in our trade, and in many cases in continental 
‘ On this point see the warnings pretixed to the Board of Trade Accounts of 
Loreiyn Urade, and alluded to in Ca, 1761, pp. vi, 3, and 4. 
