tics' projections for 2005 (Silvestri, 1995). All of the figures in Table 2.6 are based on the updated occupational classification system used by the Census Bureau. The projections shown here use the Census Bureau's "moderate" series estimates that assume moderate (versus higher or lower) assumptions about economic growth and industry employment projections. Managerial occupations, after peaking in the 1990s, are expected to decline in their relative share of the workforce, and the projection for professional specialty occupations shows steady growth in relative share over the period, which is expected to continue into the next century. Again, the demographic changes mentioned earlier may be among the underlying forces driving this change. Technical, sales, and administrative support occupations, although expected to grow in their absolute numbers, have and will retain about equal proportions of the workforce over the period from 1985 to 2005. Service occupations, although fairly constant in relative share in the 1980s and 1990s, are expected to employ about 1 in 6 workers by 2005, a trend that is likely in part to reflect underlying demographic forces. Finally, occupations at the lower end of the socioeconomic spectrum—production, craft, repair, operators, fabricators, laborers, and farm, forestry, and fishing occupations—will all shrink in their relative shares of the workforce. The latter downward shifts are more likely to reflect technological and productivity changes rather than demographic shifts.
The long-term industry shifts are equally dramatic and generally reflect the well-known shift from jobs in goods-producing to service-producing categories. Service-producing industries contained 3 in 10 jobs at the turn of the century but had grown to nearly 7 in 10 jobs by 1980, and nearly three-quarters of all jobs in 1995.
A final set of general demographic trends that are important for understanding the evolution of occupational classification systems involves the gender composition of occupational employment. Long-term comparisons of trends in the sex segregation of occupations are difficult because of changes in occupational classifications systems and problems in comparability of categories over time (England, 1981; Beller, 1984; Reskin, 1993; Wootton, 1997). Occupational sex segregation is usually defined and measured as the percentage of women (or men) who would have to