Part I, Chapter VII
ECONOMIC WELFARE AND CHANGES IN THE SIZE OF THE NATIONAL DIVIDEND

§ 1. IT is evident that, provided the dividend accruing to the poor is not diminished, increases in the size of the aggregate national dividend, if they occur in isolation without anything else whatever happening, must involve increases in economic welfare. For, though, no doubt, economic welfare as measured in money, and, therefore, the national dividend as here defined, might be increased and economic welfare in itself—not as measured in money—at the same time diminished, if an addition to the supply of rich men's goods was accompanied by a contraction in the supply of poor men's goods, this sort of double change is ruled out by the proviso that the dividend accruing to the poor shall not be diminished. But it does not follow that every cause, which, while leaving the dividend of the poor unharmed, increases the size of the aggregate dividend, must bring about an increase in economic welfare; because it is possible that a cause which increases the size of the dividend may at the same time produce other effects adverse to economic welfare. It is desirable, therefore, to inquire how far this possibility needs to be reckoned with in practice.

§ 2. Changes in consumption that come about in consequence of an increase in facilities for obtaining some of the items contained in the dividend are liable to bring about changes in taste. But, when any particular kind of commodity becomes more readily available the resultant change of taste is usually an enhancement. Thus, when machines are sent out on trial, or articles presented in sample-packets, or pictures exhibited free to the public, the popular desire for these objects tends to be augmented. When public-houses, or lotteries, or libraries are easily accessible, the taste for drink, or gambling, or literature is not merely gratified, but is also stimulated. When cleanliness, or light,^*83 or model dwellings, or model plots of agricultural land are set up, though it is only to be seen, and not owned, by the neighbours, the object lesson may still succeed and make plain superiorities hitherto unrecognised.^*84 Thus, "free libraries are engines for creating the habitual power of enjoying high-class literature," and a savings bank, if confined to the poor, is an "engine for teaching thrift."^*85 In like manner the policy of many German cities, in subsidising theatres and opera-houses and in providing symphony concerts two or three evenings a week at a very small admission fee, is an educational policy that bears fruit in increased capacity for enjoyment. It is true that an increase in taste for one thing is generally associated with a decline in taste for any other things that fulfil the same or a similar purpose, e.g. wool as against cotton or a new "best type" of motor car as against what used to be the best type, and sometimes with a decline in taste for other quite disconnected means of enjoyment. But it is reasonable, in these circumstances, to hold that the provision made for the new taste is likely to yield some excess of satisfaction over that made for the old; so that the net result of an increase in facilities for obtaining some of the items contained in the dividend will be to increase economic welfare.

§ 3. The above argument does not, however, go to the root of things. It is relevant to immediate short-period effects rather than to ultimate effects. When a group of people have passed from a state of relative poverty, to which they were accustomed and adapted, to a state of relative wealth, to which they have become adapted, will they really derive more satisfaction from the last state of their environment than they did from the first? With the changed conditions the whole scheme of their desires and habits and expectations will also be changed. If a man who had all his life slept in a soft bed was suddenly compelled to sleep on the ground under the sky, he would suffer greatly; but does a man who has always slept on a soft bed enjoy his nights more than one who has always slept under the sky? Is it certain that a hundred Rolls-Royce cars in a Rolls-Royce world would yield a greater sum of satisfaction than a hundred dog-carts in a world of dog-carts? In the chapter that follows some reasons will be given for doubting whether a substantial reduction in the real consumable income of rich people, provided it were general, would, after time had been allowed for adaptation to it, appreciably diminish their economic welfare. Analogous considerations hold good of an increase in their real consumable income. The point is a very important one. If the per capita income of this country were, say, twenty times what it actually is, it may well be that a further increase in it would not ultimately—the population being supposed constant—add anything at all to economic welfare. As things are, however, in view of low level of average real income, we may, I think, safely conclude that an increase in the dividend—apart from the fantastic hypothesis that the whole increase goes to persons already very rich—would carry with it, ultimately and not merely immediately, an increase in economic welfare. The goal of economic betterment is not a mere illusion.^*86

§ 4. There is, however, a further point to be considered. The economic welfare of a community consists in the balance of satisfactions derived from the use of the national dividend over the dissatisfactions involved in the making of it. Consequently, when an increase in the national dividend comes about in association with an increase in the quantity of work done to produce it, the question may be raised whether the increase in work done may not involve dissatisfaction in excess of the satisfaction which its product yields. Now, in so far as extra work is called out because, through inventions and so on, new and more advantageous means of employing it have been opened up, there is no fear of this. Nor is there any fear of it if the extra work is called out because obstacles, such as quarrels between employers and employed, which used to prevent people who wanted to work from doing so, have been removed. Nor again is there any fear of it if the extra work is called out because methods of remunerating work-people, which reward extra work with equivalent extra pay, have been introduced. It is possible, however, that extra work may be called out in other ways than these. Suppose, for example, that the whole community was compelled by law to work for eighteen hours a day, and—which is in fact improbable—that this policy made the national dividend larger. It is practically certain that the satisfaction yielded by the extra product would be enormously less than the dissatisfaction caused by the extra labour. There is here a cause which has increased the size of the national dividend while lessening, and not increasing, the sum of economic welfare. This type of cause is not, in the modern world, practically important, because, apart from military conscription, we have to do with voluntary, not with compulsory, labour. It is, however, conceivable that, even under a voluntary system, something analogous may emerge. From a mistaken view of their own real interest, workpeople may welcome an addition to the hours of labour of a sort which augments the dividend but damages economic welfare. Again, under the exploitation of employers, workpeople may be forced to assent to an increase of work as a less evil than reduced earnings. There are here a number of possible causes of additions to the dividend associated with damage to economic welfare. Plainly, however, among the general body of causes relevant to our discussion the part they play will be small. In general, causes which increases the size of the national dividend while involving an increase in work, as well as causes which increase it without involving this, will, the conditions of distribution being assumed, increase economic welfare.