rate of discount, and the leading London banks, who are now practically on an equality with the Bank of England, place their surplus cash with them. And hence the billbrokers and the joint-stock banks regulate the demand for money, and the power of the Bank of England over the rate of discount tends to decline. But it continues to be the trustee of the cash-reserve; for neither the bill-brokers nor the London banks keep reserves of their own, but are professedly dependent for cash on the Bank of England. Responsibility is thus divorced from power, and the characteristic danger of the Money Market is increased. Its recent history has illustrated this; but it has also shown that, in a large measure owing to Bagehot's forcible pleading, the consciousness of the delicacy of the English banking system is more fully and generally felt, and the directors of the Bank of England are in practice, if they are not in theory, more ready to appreciate the responsibilities of their position as the custodians of the one cashreserve. The main contentions of Lombard Street may be said to have become "part of the accepted theory of banking"; and may lead, at no great interval of time, to important changes.

CHAPTER VII.

WILLIAM STANLEY JEVONS. 1835-1882.

STATISTICS.

Jevons' Youthful Aspiration—His Combination of Qualities—The Misuse of Statistics—Owing to (1) the Recent Date of their Scientific Treatment (2) The Difficulties of the (1) Collection-And (2) Handling of Figures-Jevons' Life and Characteristics-His Power of Observation-His Shyness and Reserve-His Intellectual Training-Statistical Nature of his Books—His Investigations in Currency and Finance-The Importance of his Inquiries into Periodic Fluctuations of Prices-His Use of the Graphic Method-Curves of Prices -The Monthly Variation in the Rank Accounts-The Quarterly Variation-The Autumnal Pressure in the Money Market-The Recurrence of Commercial Crises-The Sun-spot Hypothesis-The Rise and Fall in the Value of Gold-The Index Number of the 'Economist'—Its Supposed Defects-The Effects of Changes in the Value of Gold-The Practical Bearing of Jevons' Inquiries.

THE economist, whose work we are now about to consider, refers in one of his letters to an ambition which he had felt in early life. "If there were one thing," Jevons writes, "I should wish to be, it would be a recognised statistical writer." This youthful aspiration was fulfilled. A competent judge of his statistical writings has declared that the "pure honesty" of his "mind, combined with his special intellectual fitness for the work, have made them

models for all time." 1 It is true that he is not known merely for his statistical studies; for he was a logician of a high order, and an economist of great and varied ability. But it is also true that, on the one hand, the logical character of his mind combined with his extensive knowledge of general economic subjects to produce his excellence as a statistician, and, on the other, that that part of his work seems to have presented the greatest attractions to him, which lent itself the most easily to treatment of a statistical nature.

He became a statistician of the first rank; and his qualities were such as to ensure this result. The reader of his Theory of Political Economy, in which in 1871 he explained, and applied to the theory of value, the conception of final utility,2 must be sensible that, bold as his theoretical reasoning is, he owed much of its exactitude to a keen appreciation of fact. If, on the other hand, we turn to that book on The State in Relation to Labour, which was published in 1882, or to those essays on the Methods of Social Reform, which were collected in a volume after his death, we recognise that his wise and sympathetic treatment of such practical problems as the "amusements of the people," the employment of "married women in factories," and the advantages of a "state parcel post," was due, in some measure at least, to his firm grasp of theory. PROFESSOR FOXWELL has said 3 that "in his union of high speculative ability with the greatest reserve and sagacity in the treatment of practical problems," he was a "conspicuous example of the truth" of the defence of theoretical study made by the French economist Cournot, who urged that "those who are most concerned for the

1 Professor Marshall in the preface to Jevons' Investigations in Currency and Finance, p. xliii.

2 See above, page 107.

3 In the preface mentioned above, p. xlii,

precision of their principles will be most sensible of the limits of their application, and therefore the least unpractical in their treatment of real questions."

It was this combination of qualities which rendered him so able a statistician. A reproach is frequently levelled against statistics, which is summarily expressed in the statement that "figures will prove anything"; but it is justified only by their misuse, and this may be traced in a large measure to an unnatural divorce of theory from facts, or facts from theory. The assertion "simply means," as MR. GÖSCHEN states, that "figures, which never tell untruths, may be so handled as to present untruths." There can be little doubt, however, that the view of statistics, which is commonly prevalent, is unsatisfactory; and MR. GIFFEN has remarked that in "journals of the highest standing there are the wildest blunders of the schoolboy order." This misuse and misrepresentation may perhaps be ascribed to two main causes, one of which is the comparatively recent date of any scientific treatment of the subject, and the other consists in the many serious difficulties which attend such a treatment.

It is true that in one sense, as the French statistician and economist M. Block has observed, statistics "have existed ever since there were States." The name seems to have originally referred to "inquiries into the condition of a State"; and Achenwall, who has been called the "father of modern statistics," regarded 3 them as a "survey of the customs, laws, and forms of government, by which one

1 In his inaugural address as President of the Royal Statistical Society in 1887. Mr. Göschen is the author of a work on The Theory of the Foreign Exchanges, which has passed through many editions.

2

Essays in Finance (second series), p. 133.

3 See Du Guy's paper in the Jubilee Volume of the Royal Statistical Society. Achenwall lived from 1719—1772.

nation differed from another "-as history, in short, with the uninteresting details, as they seemed to him, of the names of monarchs, the changes of dynasties, and the issues of wars, left out. He did not dwell so much on their numerical as on their descriptive character; but they are now generally understood to mean collections of facts capable of expression in a numerical and, if possible, tabular form. This was once known as 'political arithmetic;' and SIR WILLIAM PETTY,1 two centuries ago, described his essays at this 'arithmetic' as the use of a "method" "not yet very usual," which, instead of employing "only Comparative and Superlative words and Intellectual Arguments," enabled him to "express " himself in "Number, Weight, and Measure." In similar language Jevons characterised a series of statistical papers, which had been collected by him, and were published after his death under the title of Investigations in Currency and Finance, as an "attempt to substitute exact inquiries, exact numerical calculations, for guess work and groundless argument."

Even in this latter meaning of 'political arithmetic' statistics have an ancient origin. It has been said that almost the first act of a regular government would be to number its fighting men, and the next to ascertain the amount of taxation which could be raised from the rest of the community. But it is apparently 2 only within the last three hundred years that systematic scientific use has been made of such information; and it is more recently that statistics have been brought into close connection with political economy. There were 'bills of mortality' in

1 Petty also wrote on general economic subjects.

2 See the article on Statistics in the Encyclopædia Britannica (ninth edition).

M