flatly contradict each other, and two will suffice to show that the costly and tedious preparation of elaborate statistics, however imposing the result may look, offers not the slightest guarantee of their careful handling and sound analysis.[1]
It will not quickly be forgotten by specialists that Mr. Robert Giffen, an eminent public officer, occupying a high and responsible position at the Board of Trade, now President of the Statistical Society, presumedly a learned and acute body, carried out a calculation of average freight upon many millions of tons of goods as fifteen instead of one hundred and fifty shillings per cent, and thereupon built an argument (vitiated by no less an error than cutting off nine-tenths of the correct rate), to inform a select audience collected to hear him read a lengthy paper filled with figures framed and grouped to support familiar and foregone conclusions. This mistake would no doubt have passed muster had it not been that a naval officer, opposed to those conclusions, happened to be present and to catch the word which his practical business knowledge told him instantly was wrong.[2] My second
- ↑ The cases cited recall to mind an observation made many years ago by a bank manager who had been made a tool in some enterprising commercial transactions, which inflicted a loss of some £50,000 upon the banking company, and who was asked how he could possibly at his mature age have behaved so foolishly. His reply was to the effect that men in his position are daily and hourly dealing with such stupendous totals in pen and ink that the figures cease to convey to their minds those fitting and intrinsic ideas as symbols of value properly belonging to them.
- ↑ "Captain Halford Thompson asked Mr. Giffen if he considered 15s. per cent, a fair average per-centage for freight on the whole import trade of the world. "Mr. Giffen replied that that was what the figures appeared to show. "Captain Thompson said that Mr. Giffen was hard upon those who thought they could handle figures ' without previous education to the trade—an education which he evidently thought could not be complete unless it had been carried out under the supervision of the Cobden Club. (Laughter.) Mr. Giffen's primary object was to raise the bogey that there were so many deductions and allowances to be made from statistics of import and export trade that no ordinary observer could make safe use of argument based upon them. It was curious that, while denying this, he should base pages of argument and piles of figures taken in different years from returns based upon utterly different principles. Referring to Mr. Giffen's main point about the excess of imports over exports as due to the greater carrying power of the nation, every one knew that imports must always have a per-centage taken off, because imports included freight and exports did not. But while admitting that allowances must be made, it left them a long way off admitting the accuracy of the figures which Mr. Giffen had built up as to the trade of the world. What Mr. Giffen wanted to prove was that the £162,000,000 which his table showed as the excess of imports over exports in the trade of the world represented approximately the cost of conveyance after deducting a sum of £32,000,000 for miscellaneous charges and commission, which brought the amount down to £130,000,000. To prove that that was not an excessive amount Mr. Giffen informed them that that sum only amounted to a charge of 15s. per cent, on the total imports of the world. Now, so far from the result of the calculation
