EVERY GOOD theoretical or econometric study must be based on a reasonably
accurate empirical foundation. If the basic magnitudes of the subject
are misperceived, the theoretical model or econometric specification
will lead the research astray.
In recent years, research on the central macroeconomic questions of unemployment
and wage inflation has been advanced by the empirical studies
of Hall, Holt, Parnes, Perry, Wachter, and others. Meanwhile, the U.S.
Bureau of Labor Statistics has benefited the profession by expanding the
data base with detailed monthly summaries of household and establishment
data and through the provision of complete data from the Current
All of this microeconomic evidence has greatly enriched understanding
of the nature of unemployment. The traditional view, based on the experience
of the depression, pictured the unemployed as an inactive pool of job
losers who had to wait for a general business upturn before they could find
new jobs. Modern research has shown that this picture is distorted. The
majority of the unemployed do not become unemployed by losing their
previous jobs; they quit voluntarily or are new entrants or reentrants into the labor force. Moreover, the typical duration of unemployment is very
short; more than half of unemployment spells end in four weeks or less.