November 2010

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30        


MBM Principal Sources

Blog powered by Typepad

« E.T. Phoned Home but Bangladeshis Can't | Main | Know Thy Knowledge Processes: The Confessions of a Negative Saver »

18 April 2007




Your point about economic forecasts/planning is well-taken.

However, I'm wondering about the specific criticism. As someone who lived through the tech boom in Silicon Valley during the 1990s, I feel intuitively that you are correct. But that may just be my own selection bias. I'd be curious to see the real data - although a lot of jobs were created in the tech sector, were there a lot of them as a percentage of total jobs created? Or can the tech sector be credited as an engine for generating a lot of subsidiary jobs?


Alastair Walling


A very good point -- there are a lot of things "we know" that data do not support. Although the idea that lots of technology jobs were created in the 1990s probably on firmer ground than lots of “known” things, it is definitely worth checking out. In response to your query (and because it is Friday), I have been absolutely “geeking out” at the Bureau of Labor Statistics (BLS) website. The website is absolutely comprehensive, which, of course, makes navigating it very difficult and time consuming, but you can find out all kinds of stuff. If you go to you have the option of building tables charting the employment of people doing all sorts of stuff. For example the number of people involved in the making of “sausage and other prepared meats” rose from 87,000 in 1990 to 103,000 in 2003 – I mean who knew? If you want to know how many people were involved in building missiles and space vehicles in the 3rd quarter of 1970, then I can tell you its there. One problem is that there are so many categories that it is difficult to find everything that would qualify as “high-tech”. However, if you keep grabbing categories that sound about right, then you very quickly get a feeling for the tech bucket. For example, the number of people employed in “computer and data processing” at the end of 1990 was about 780,000, but this figure had risen to two million by the end of 1999. Sure, this number had been rising during the 1980s, but this segment of the economy had been adding a consistent number of jobs (about 200,000 every five years in the 80s). If one just looks at the trends in the data, then the adding of about 1,200,000 jobs between 1990 and the end of 1999 would appear unlikely. To put this perspective this time period included a rather sharp recession in the early 1990s, and I think yesterday’s WSJ put the number of jobs created during the Clinton Presidency at about 11 or 12 million, so these categories likely represent a good portion of the number of jobs created in the 1990s.

Here are a few of the categories I grabbed that seemed kind of “techie.” However, I think there are some problems with antiquated categories. For example I kept looking for things like cellular communications or e-commerce but drew a blank.

Other examples – 1990-2000
• Cable TV and other pay television services 123,000 to 220,000 jobs
• Radio, Television, and computer stores: 280,000 to 500,000 jobs
• Computer programming services: 150,000 to 500,000 jobs
• Prepackaged software: 115,000 to 300,000 jobs
• Computer integrated systems design: 97,000 to 220,000 jobs
• Data processing and preparation: 200,000 to 280,000
• Information retrieval services: 48,000 to 230,000

Hope this answers your comment.




Great data - you are more than thorough :-)

Nice to see that data supports intuition occasionally :-)

I'm not surprised that a government agency would lag in keeping up-to-date as the economy changes but e-commerce and cellular technology are such enormous endeavors, and have been for years, that this lag seems odd even for government work. But your point is pretty overwhelming without it.

Loved the sausage data :-)


The comments to this entry are closed.