A good friend of mine reminded me this morning that the world’s worst mega-merger in history was 10 years old. It is appropriate that there were articles today about AOL cutting over 1,000 jobs in an effort to meet certain arbitrary goals.

I was also thinking about another 10 year milestone. The census is coming up. 10 years ago was the last big census in the US. Since then, the US has invested a considerable amount of money in software and process engineers to make the census more efficient. If you read the final reports on any of these projects, you’d be impressed by the level of expected productivity gains. Unfortunately, these efforts seem to have been a complete failure, because we need to hire twice as many people per capita for this year’s census as we did for the one ten years ago. We needed 500,000 for the 2000 census, and 1.2 million for the 2010 census.

I understand that the census is a jobs program, but it doesn’t make sense to hire people to work at half efficiency. Why not hire the correct amount for the census and have the rest do something else productive? Did the magic suggestion box run out of ideas?