|
|
|
The IT Workforce Update: Further Considerations |
|
| Refiguring Losses of Core IT Jobs After the Year 2000 | |
Eva Varon of CIO Magazine contacted us recently, to draw on information from the recent
fifth report in the IT Workforce Data Project Series, The Outlook in 2003 for Information
Technology Workers in the USA (Washington, DC: CPST, September 2003). One of the things
she wanted to do was to use information in the report on the number of core IT positions
lost since the peak year of 2000. We went over the data and the first thing we discovered
was an error on our own part. Due to a mistake in transcribing numbers, we said that at
least 150,000 positions had been lost by 2000. We should have said that at least 206,000
positions were gone.
|
|
|
This was not all. Both of the numbers stated above are drawn from an older methodology
used by the Bureau of Labor Statistics to deal with occupational data up through 2002.
As was noted in a sidebar in our September report, BLS adopted new occupational measures
at the beginning of 2003, and the use of the new set of job titles appears to have
substantially raised counts of professional people in IT fields. Our judgment is that
the new approach is probably more accurate than the older one, which means that the earlier
data for 2002 and earlier years are likely to significantly understate the size of the IT
workforce. It is not easy to estimate precisely how large the underestimations might be,
but a reasonable case can be made that the earlier numbers should be increased by as much
as 45 percent. If so, the number of lost positions between 2000 and 2002 jumps to roughly
300,000 - double the 150,000 level that we originally reported. We now think this larger
estimate is the most plausible one for core IT professionals. This figure is for the net
loss of jobs; probably even more positions were cut, but at the same time some new positions
were also created. Annual estimates from BLS can be taken as applying to the middle of a
year, so the number applies to the entire two year period between mid-2000 and mid-2002.
|
|
|
Even larger estimates of total IT job losses are being made, such as those released on
November 19 in the 2003 edition of the annual Cyberstates reports issued by the American
Electronics Association (AEA), which believes that total shrinkage of jobs in the industry
exceeded half a million in 2002 alone. Such a figure may not be out of line with our
suggested level of 300,000 lost positions for core IT professionals in both 2001 and 2002,
given that the AEA numbers are based on total employment for all types of workers in a broad
range of IT-related industries. In contrast, our own estimates are for all industrial and
other employment sectors, but are limited to core IT professionals, only computer scientists
and engineers, systems analysts, network and database administrators, programmers, and computer
and information systems managers. AEA used federal data on industries; we used federal data
on occupations.
|
|
|
A remaining issue is the question of how many more positions have been lost since mid-2002.
Quarterly data from BLS, including brand new results released within the past week for the
third quarter of 2003, suggest that in this more recent period of time the overall number
of core IT jobs has stabilized, at around roughly 3.3 million positions.
|
|
|
One additional point: it will soon become possible to obtain much more precise
information on the effects of the changes in the occupational codes used by BLS.
The Bureau is currently assembling a tabulation of the entire civilian labor force
that includes occupations classified by both the old and new codes. Cross-tabulation
of these data will show how people in any one of the old-style job categories are now
distributed among the new occupational titles, and which older occupational groups are
contributing to the makeup of any one of the new ones. Such information will help bridge
gaps between old and newer data and keep time series intact. We plan to assess the detailed
effects of these methodological changes on counts of people in all of the scientific,
engineering and allied technical disciplines. Watch this space for more information.
|
|
| - Richard Ellis (November 20, 2003) | |
Sloan Foundation | Contact Us