For the purposes of this survey I define inflation as a sustained rise in prices. (There are other definitions which I shall discuss subsequently.) While the definition of inflation as a sustained rise in prices is a simple one, one encounters problems of some difficulty as soon as one tries to apply it in practice. The first such problem is that there may be price rises which are not inflationary but are merely part of the normal working of the competitive system. For example, it is obvious that if there is a crop failure or something of that kind, there will be some rise in prices to ration the reduced supply; again, if the economy is moving upwards from the bottom of a recession to a higher level of employment, there will inevitably tend to be some rise in prices due to the increased demand for goods and labour. In both cases, the rise of prices ought not to be regarded as inflationary: first, because in the nature of things it will be self-limiting, and second, because it does not in itself represent any serious policy problem.