Appendix. The Abuse of Shipping Statistics

APPENDIX.¹

THE ABUSE OF SHIPPING STATISTICS.

SHIPPING is one of our greatest industries. Its services, as “invisible exports,” play a large part in explaining the Balance of Trade. It is, moreover, an industry on which much, politically, depends. It is of importance, then, to ascertain how our shipping stands, and whether it is gaining or losing in comparison with that of other nations. But the comparison is so beset with technical and statistical difficulties that it may be advisable to set forth these in some detail.

As preliminary, it seems desirable to show, in untechnical language, the difference between gross and net tonnage, and their relation to cargo-carrying capacity.

Lloyd’s Register gives the gross tonnage, and, according to it, the Merchant Navy of Great Britain appears as over 16,000,000 tons, or about half the gross tonnage of the world. Statistical statements, however (e.g. the Statistical Abstract and the recent Board of Trade Memoranda), give the net tonnage, and this appears as 10,000,000 tons. What constitutes the difference between the two?

When a vessel is put into the water, her weight is, of course, the number of tons of water she displaces. As she loads and sinks deeper, she displaces so many more tons, and, if nothing but cargo were stowed within the shell, the additional tons over her own weight would represent her cargo-carrying capacity. The tons of water displaced outside would correspond with the tons of cargo she could hold inside. (It must be noted, however, that the correspondence is only approximate. The vessel’s internal capacity is calculated into weight at the ratio of 40 cubic feet of average cargo to the ton, while 40 cubic feet of water weighs something like one and a tenth tons.) But there is a good deal of space which cannot carry cargo. In all ships there is the space under the floor of the hold, the bulkheads, etc. Other spaces, again, are required for the navigation of the vessel—seamen’s quarters, chart-rooms, stores, etc.; and, in the case of a steamer, there is also the space occupied by the propelling power. When these are calculated into tons and deducted, we get the “net tonnage,” and it is on this net tonnage as, presumably, representing the productive capacity, that harbour and other dues are paid. The gross tonnage, of course, is the net, plus the deductions just mentioned.

But, on looking into the actual carrying capacity, as shown by facts, we find that, far from being represented by the registered net tonnage, it is usually, in modern ships, something like one and a half times the gross tonnage. Taking a steamer whose gross registered tonnage at Lloyd’s is 4926 tons, and her net 3147, I find that her “deadweight” (or actual lifting) capacity is 7580 tons.

The explanation is that statutory allowances are made for propelling space, for certain closed-in spaces, for seamen’s quarters, etc., and that the ingenuity of shipowners and shipbuilders is taxed to get as much as possible counted in for deduction. Engines and boilers, too, are packed as closely as practicable, and are steadily being made more powerful for the space they occupy. And I am told that it is not an uncommon practice for open spaces on steamers in certain trades to be temporarily boarded up for the reception of cargo or as bunkers.

The foregoing brings out the first difficulty in making accurate comparisons between different merchant navies. The true test is one that compares actual carrying capacity. But if ships are simply compared either by their gross or net tonnage, the more modern fleet will, ton for ton, have a greater actual carrying capacity than an older fleet. Now, it is well known that we are constantly selling our old ships, and replacing them with new, thus getting greater carrying

capacity relative to the net tonnage, and paying lighter dues. In fact, our fleet has been to a large extent rebuilt within the last 20 years, while foreign fleets embrace old and less productive vessels bought from us. To compare British tonnage e.g. with Norwegian, is obviously misleading.

A second difficulty is that, in comparing rival fleets simply by tonnage, sailing vessels rank equally with steamers. But, according to the usual calculation, steamers—presumably as doing the same work in a third of the time—are three times more effective in carrying power than sailers.

While, then, Great Britain owns only half the tonnage of the world, she owns far more than half the carrying capacity. Not only do sailers form an insignificant portion of our total fleet, but we have for years been “going out of sailers,” while foreign fleets—particularly the French—are largely composed of them. And, again, foreign fleets consist, to a considerable extent, of wooden and composite vessels, of which we have comparatively few.

Taking these two difficulties together, it seems obvious that, to compare the British fleet with foreign fleets simply by tonnage, is to put efficients and non-efficients as equal weights in the balance. It is like lumping skilled and unskilled workers together per man as “labour,” the true comparison being labour power.

A third difficulty is that Lloyd’s Register takes account only of steamers over 100 tons gross, and of sailers over 50 tons net. But Russia registers anything over 25 tons net; Germany, 17½ tons; Norway and Denmark, 4 tons; France and Italy, 2 tons. Again, therefore, on an equal comparison the British fleet would be much larger than it appears to be.

A fourth difficulty is that, in the case of passenger liners, tonnage is no indication of cargo-carrying capacity. That Germany has some 20 “leviathans” of over 10,000 tons, as against our 40 or 50—a much smaller proportion of our total—tells pro tanto against Germany.

A fifth difficulty is that, while the true international comparison should relate solely to ocean-going vessels, other registers than Lloyd’s, as we have seen, count in quite small vessels whose influence on the ocean carrying trade is negligible. And the American register includes the large but purely local traffic of ships plying on the great lakes.

These difficulties are enough to show how carefully shipping statistics must be handled, and enable us to estimate the accuracy of Mr. Chamberlain’s Liverpool speech. In it he made much of the fact that British shipping between 1890 and 1901 increased only by 1,400,000 tons, while foreign tonnage had increased by 2,200,000. Now, this is simply a comparison by tonnage, and, if there is any truth in what has been said, it raises all the questions just suggested.

Independently of this, surely the only thing that concerns us in such comparisons is the increase in shipping that does or may compete with us. Now, as a fact, over 1,500,000 tons out of the 2,200,000 are accounted for by American “enrolled and licensed,” including lake and river steamers, coasting vessels (whose trade is reserved for American owners), and cod and mackerel fishing boats. The absurdity of such a comparison requires no comment. Again, when we examine the individual increases and decreases, what strikes one at once is that, while most fleets show an increase, Norway shows a decline from 1,705,699 tons to 1,467,089, and American over-sea tonnage a decline from 946,695 to 889,129 tons. The only figures which may give us pause is the German increase from 1,433,413 tons to 2,093,033. As to this, we should require to know what kind of tons these are; but, even if they represent equal efficiency with our tons, a percentage comparison between a fleet of 10,000,000 tons and one of 2,000,000 tons is quite fallacious. We, for instance, could not double our tonnage without driving every other fleet from the sea. France could double hers without becoming an appreciably stronger competitor.

Further, the comparison is very much invalidated by the imperfection of the statistical returns referred to. Japan, e.g. in the Board of Trade Memoranda, shows no return for 1890, and the 917,879 tons of 1901 are consequently counted as all increase, while other nations give no returns for 1901.

But, if I am not mistaken, there is another fallacy still in the comparison. Mr. Chamberlain is suggesting a decline in British shipping, and he proves it by showing that the tonnage of the “whole British Empire” has increased only by 1,400,000 tons, as compared with a total increase in foreign tonnage of 2,200,000. But what we find from the returns is that British tonnage has increased by 1,629,882, and that it is a decline of 197,582 tons in the shipping of the Colonies and British possessions that brings the increase down to the lower figure. It need scarcely be said that the only question at issue is the shipping of Great Britain.

As to evidence given by statistics of clearances and of building of ships abroad, only a word need be said here. The difficulty of any comparison by clearances may be suggested by the fact that liners which cross the Atlantic or Pacific will enter and clear once only in one to four weeks, while vessels trading in and with the Baltic may enter and clear every few days. And, when we are told that more “tonnage” is being built abroad than is being built here, all the difficulties of comparison dealt with above are again suggested. So long as we can build cargo ships at about £5 10s, it does not seem probable that the competition is very serious, from the shipbuilder’s point of view at least.

Looking at the number of neglected elements which are concealed in Mr. Chamberlain’s apparently simple comparison by tonnage, I think one may use his own words, with another reference: “Serious people ought to give serious consideration to what at anyrate are signs”—of the abuse of statistics.

¹See above, p. 176.