Workers are constantly leaving industry and going into busi ness for themselves. Some are retiring and living on their past savings, or on the income of others. Some leave industry and go to schools, colleges and universities. Theoretically, unless Ameri can employment figures cover every branch of production and dis tribution it would be quite possible for employment in the major industries of the country to be declining, and at the same time have the number of unemployed grow less. On the other hand, with approximately one-third of a million immigrants entering the United States each year, with about 200,000 people coming into the cities from the farms, and with anywhere from 1,500,000 to 2,000,000 young people reaching the working age each year, a very marked growth in our employment may at the same time be accompanied by a serious increase in the number of people unem ployed.
The most comprehensive employment reports are to be found in the monthly survey of employment for the United States as a whole which is published in the Monthly Labor Review of the United States Bureau of Labor Statistics. These reports, based upon returns from 12,000 plants employing over 6,000,000 work ers with a weekly payroll of over $170,000,000, present monthly changes in both employment and in payrolls for 12 groups of industries classified as follows: The statistics of unemployment compiled and issued by the different countries exhibit the widest variety in completeness and value. The only statistics yielding trustworthy percentages of unemployment are those concerning workers insured against un employment, for only in such cases is the "field" of persons registered as unemployed accurately known and proof of unem ployment rigorously required. Trade unions paying unemploy ment benefit to their members come within this category, and for these, more or less extensive and trustworthy statistics exist for many countries. The numbers covered are, however, usually only a small proportion of the whole wage-earning population, and the data require very careful examination before the degree in which the percentages reflect unemployment among workers generally can be estimated. In no country other than the United Kingdom and Germany does there exist a national system of un employment insurance covering the bulk of wage-earners so or ganized and used as to produce unemployment statistics of national range and practical trustworthiness. In some countries, notably the United States, statistics of unemployment are virtually non existent, but in that country, as also in Canada, effort has been directed to statistics showing variations in the numbers employed from one date to another in representative establishments in certain industries. These employment statistics have their own
especial value, but it is not possible to deduce from them with any assurance, figures of unemployment. In many countries there are national or local employment agencies, where workers desiring employment may register and the statistics of registrations and placings by these agencies give some indication of changes in intensity of unemployment in the areas where they operate but in no case, except where, as in the United Kingdom and Germany, the administration of an unemployment insurance fund is added to their employment-finding functions, do these administrative statistics give any measure of the gross volume or rate of unem ployment. For these reasons the figures quoted below cannot, in any case, be taken as measures of the gross amount of unem ployment in the country in question nor as reliable indications of the rate of unemployment throughout the country as a whole; and they should consequently on no account be used to compare either the level or the rate of unemployment as between country and country.