If income is transferred from rich persons to poor persons the proportion in which different sorts of goods and services are provided will be changed. Expensive luxuries will give place to more necessary articles, rare wines to meat and bread, new machines and factories to clothes and improved small dwellings; and there will be other changes of a like sort. In view of this fact, it is inexact to speak of a change in the distribution of the dividend in favor of, or adverse to, the poor. There is not a single definitely constituted heap of things coming into being each year and distributed now in one way, now in another. In fact, there is no such thing as the dividend from the point of view of both of two years, and therefore, there can be no such thing as a change in its distribution. This, however, is a point of words rather than of substance. What I mean when I say that the distribution of the dividend has changed in favor of the poor is that, the general productive power of the community being given, poor people are getting more of the things they want at the expense of rich people get ting less of the things they want. It might be thought at first sight that the only way in which this could happen would be through a transference of purchasing power from the rich to the poor. That, however, is not so. It is possible for the poor to be advantaged and the rich damaged, even though the quantity of purchasing power, i.e. of command over productive resources, held by both groups remains unaltered. This might happen if the technical methods of producing something predominatingly consumed by the poor were improved and at the same time those of producing something predominatingly consumed by tile rich were worsened, and if the net result was to leave the size of the national dividend as defined in Chapter V. unchanged. It might also happen if, by a system of rationing or some other device, the rich were forced to transfer their demand away from things which are important to the poor and which are produced under such conditions that diminished demand leads to lowered prices. Per contra——and this point will be seen in Part Ⅳ. To be very important practically——the share, both proportionate and absolute of command over the country’s productive resources held by the poor may be increased, and yet, if the process by which they acquire this greater share involves an increase in the cost of things that play a large part in their own consumption, they may not really gain. Thus a change in distribution favorable to the poor may be brought about otherwise than by a transference of purchasing power, or command over productive resources, to them, and it does not mean a transference of these things to them. None the less, this sort of transference is the most important, and may be regarded as the typical, means by which changes in distribution favorable to the poor come about.