are open to grave objections. At the outset we are told that the whole purpose of economic investigation is not primarily scientific, 'if by science we intend the singleeyed search after knowledge for its own sake. It is rather practical and utilitarian, concerned chiefly to lay bare such parts of knowledge as may serve directly or indirectly to help forward the betterment of social life.' When this test is applied to Prof. Pigou's own enquiry into wealth and welfare, we may first look to the broad results, and, secondly, to the detailed treatment of some particular practical problems.

If we look for the general principles applied, and survey the broad results obtained, we find the argument rests on a simple application of the theory of utility, or, in popular phraseology, 'the greatest happiness of the greatest number.' The simplicity of the application is shown by the treatment of the common objection that the additions to the incomes of the workers may be prevented from adding to the welfare of the nation by the stimulus they afford to population. We are told that this contention, even if valid, would not suffice to prove that better fortunes for the workers fail to increase economic welfare; for, after all, if the life of an average workman contains, on the whole, more satisfaction than dissatisfaction, an increase in numbers, even though it leave economic welfare per head the same, involves an addition to economic welfare in the aggregate.' That is to say, the bigger the ant-heap, provided the average ant has a balance of the pleasure of living over the pain of keeping alive, so much greater is the aggregate pleasure of the heap. As a simple sum in political arithmetic this particular example is no doubt quite correct. But, as the foundation of a system of practical legislation, this 'obvious and simple system' of greatest happiness is as impossible as the obvious and simple system of natural liberty which it has displaced. Either principle or idea must in practice be qualified by various other political and social principles, some of which derive their effective force from historical conditions. Adam Smith showed very clearly the qualifications needed even in his day for the principle of individual liberty. It is strange that Prof. Pigou should quote the 'highly optimistic' theory of Adam Smith, that the national dividend, in given

circumstances of demand and supply, tends naturally to a maximum, and should not mention at all the serious qualifications that were introduced by Adam Smith in the immediate context, and especially the references to the limitations imposed on liberty by justice, and to the advantages of the undertaking by the state of various public works and institutions. Prof. Pigou's translation of the plain language of Adam Smith into the language of marginal net products only removes further from the facts the actual presentation of the system of natural liberty.

Similarly Sidgwick showed very clearly that the principle of utility needed serious qualifications, and in the end, in spite of his thorough-going support of utilitarianism, he considers' only very mild and gentle steps towards the realisation of the socialistic ideal as at all acceptable in the present state of our knowledge.' Prof. Pigou's economy of reference to Sidgwick is even more strange than his reticence regarding Adam Smith. For not only does the younger writer apply the same general principle of utility, but the main trend of the argument is the same. Yet Sidgwick's name is only mentioned in connexion with one or two points of minor detail. If Prof. Pigou had really appreciated the work of Sidgwick, he would have been saved from some unfortunate inconsistencies, and from some appalling lacunæ in his argument. If he had gone back to the original Adam Smith, or even to the Ricardian adaptation made by Mill, he would have been led to take some account of the institution of private property and of the laws of inheritance and bequest, and of the relations of the particular country considered to the rest of the commercial world. Instead of absorbing this old learning, Prof. Pigou quotes with approval the ingenious scheme' of M. Rignano expounded in a work entitled 'Di un socialismo in accordo colla doctrina economia liberale.' According to this plan, resources would be taxed to the extent (say) of one-third when they descended from their original accumulator to his successor, the remainder would be taxed to the extent of two-thirds when this successor handed them on, while at the next succession the whole of what was left would be absorbed. There would be, we are told, 'some technical difficulty 'in the enforcement of any plan of this kind;

but no mention is made of the possibility that the resources might be transferred to some other country, e.g. Turkey or China, where the new economy was not so rampant. The main result of Prof. Pigou's argument is that the welfare of the nation would be greatly increased by the compulsory transfer of wealth from the richer to the poorer members. Convey, the wise it call,' said Ancient Pistol; but, perhaps, compulsory transfer' is more suggestive of the attainment of the socialist ideal by way of highly progressive taxation.

6

If we test the results of this latest application of mathematical economics by particular problems, we find the same disregard of the work of other writers and the same straining after the appearance of originality. A chapter is devoted to the special case of railway rates; and, by insistence on a peculiar interpretation of jointsupply and of the value-of-service principle, Prof. Pigou tries to show that a popular confusion has vitiated the great bulk of modern discussion on the railway-rate problem. An accident of language has caused an important field of economic inquiry to be dominated by a doctrine which is essentially unsound.' This accident of language, we are told, is the misunderstanding of the term 'joint-supply.' According to Prof. Pigou, 'two products are supplied jointly when a unit of investment expended upon increasing the normal output of one necessarily increases that of the other.' In this sense, he says, the transportation of coal and the transportation of copper along a railway are not joint products; and accordingly the maxim of charging what the traffic will bear, which is supposed to depend on this theory of jointsupply, falls to the ground. It would be strange if such writers as Mr Acworth and Mr Hadley and the whole tribe of railway experts, practical and theoretical, have fallen into a gross error by not understanding the words they use. The alternative suggestion that Prof. Pigou has misunderstood his authors is at least as probable. The old idea of joint-supply was that two or more things produced together give under competition the normal rate of profit on the capital, and that the relative prices inter se depend on demand and supply. If in place of competition we introduce monopoly, we must read for 'normal' profit 'maximum' profit. If, then, a railway is

subject to competition or monopoly or to a mixture of both, the rates inter se will be adjusted to what the traffic will bear, that is, will depend on the demand for the services in connexion with the supply of means available.

Other particular problems might be cited in which there is shown the same tendency to overlook the work already done or to translate accepted theories into the new language. There are chapters on the theory of index numbers, on co-operation, on the various causes and remedies of unemployment, on the latest application of the tabular standard for lessening the variability of wages, and on a number of practical problems, all of which have been the subject of much special study. In so wide a range there are often evidences of original thinking, of apt illustration, and of vigorous criticism, and in other cases there is the expression in fresh language of old ideas. The chapter (Part II, ch. 9) on the conditions which determine the appearance of monopolistic power in modern industries is a good example of didactic presentation of accepted ideas; the examination of the influences of heredity and environment (Part 1, ch. 4) is fresh, though it begins with the strange assertion that the older economists only considered the quantity and not the quality of the people (which is directly the reverse of the truth); and other instances of exposition or criticism might be commended. But the relation of the parts to the whole, in spite of an analytical table of contents extending over twenty pages of small print, is not clear. The truth is that the thread on which the pearls or beads are hung is not strong enough to support the weight. This thread is the 'marginal equality of net products.'

The ideas concealed under this forbidding expression are difficult to express in ordinary English. Perhaps the best introduction is the lucid though lengthy attempt of Mr Wicksteed. Prof. Pigou simply takes it for granted that the expression and the curves by which it is illustrated ought to be intelligible without any elaborate explanation. And yet, if we have here presented a new idea, which is to be the supreme test of economic welfare, it must be capable of translation into ordinary thought. The nature of the idea may perhaps be understood from the applications. The main practical result seems to be

that we are to look on all the resources of a nationland, labour, and capital-as given to the nation to be used in the first place so as to produce a maximum national dividend. This we are assured will be effected if, in general, there is equality in the marginal net products of all these resources. This means that the nation should never spend a penny more in any one line of business, if it can use the penny to a little more advantage in some other line. If, then, the natural and acquired habits of individuals do not achieve this economy of production, there is a prima facie case for governmental interference. The maximum dividend being thus secured, the next step is to distribute it so that the maximum utility may be obtained for the nation. Here again we apply the principle of equality. If a little more utility is to be got by the transfer from one set of people to another, then also, if natural charity falls short, a case is made out for compulsory transfer. 'Once the things are there,' as Mill said, 'society can dispose of them in any way that it pleases.' Care must, indeed, be taken that the method of distribution does not hinder the things from getting there at all. Putting together these two main ideas of maximum production and maximum utility, the system as presented by Prof. Pigou may be described as a species of mathematical socialism. As such, like other forms of socialism, it comes in conflict with the different systems of economic thought which are based on other principles and ideas; which take more account, for example, of personal liberty and individual property and of the historical conditions under which our present methods of production and distribution have been evolved.

་

The difference between this mathematical socialism and the modified individualism of so-called orthodox writers may be realised by comparing Wealth and Welfare' with the second volume of Dr Pierson's 'Principles of Economics,' of which a translation from the Dutch has just been published. Unfortunately the long delay has taken somewhat from the freshness of some of the illustrations, but the main results are of permanent interest and value. Especially is this the case in the last part (Part IV), which treats of the revenues of the state, of public domains and taxation, of national expenditure and