In this paper certain statistical and theoretical problems connected with changes in the level of real wages will be examined. In the first place, information is presented which deals with recent trends in real wages in the United States and in Great Britain. Next, there follows a more or less sketchy treatment of the conventional theory of distribution—this, to provide the analytical tools for the interpretation of these movements. Finally, the significance of these various determinants of the level of real wages is discussed, in the light of statistical and descriptive material to be presented in this paper.

It must be noted at the outset that no very careful treatment is accorded the British material. For the purposes of this survey, the British experience—which was perhaps “normal” in some vague sense—is used as a basis for comparison and contrast with the distinctly “abnormal” American experience. The statistical data presented deal only with the six-year period beginning with 1932 and ending in January, 1938. Prior to the earlier date (1932) adequate information about wages and hours of work in the United States was not available. In any case, even six years, if studied by months, provide quite a wealth of statistical data—enough at least for a preliminary survey such as this.

The subject of the investigation has been the changes in real hourly earnings. This was decided upon for reasons of convenience and not because it was supposed that such an index offered the most significant measure of change in the workers' income. Very probably, most of us would feel that changes in the purchasing power of weekly or annual earnings, or even of working-life earnings, would more satisfactorily measure such alterations. But it is difficult to establish such indices, and the concepts they measure are not so amenable to theoretical treatment.