by Calculated Risk on 6/04/2008 08:00:00 PM

An article in the WSJ today reminded me of the residential construction employment conundrum. From the WSJ: Housing Slump Hits Hispanic Workers, But Most Immigrants Remain in U.S.

The housing slump has disproportionately hurt Hispanic workers, provoking a jump in unemployment that has hit the immigrants among them the hardest, according to a new study.
...
Many undocumented workers don't appear on employment rosters because they work as independent contractors or are hired indirectly by big developers through subcontractors or labor brokers who don't officially hire every worker. "They were ghosts to begin with," says Rose Quint, an economist at the National Association of Homebuilders. Thus, she says, "the decline in employment is probably bigger than numbers are showing."

Here is the report: Latino Labor Report, 2008: Construction Reverses Job Growth for Latinos

This graph shows the construction employment conundrum: why have starts and completions declined about 50% from the peak in 2006, and yet residential construction employment is off only 14%?

Click on graph for larger image in new window.

Note that starts are shifted 6 months into the future since it takes a little over 6 months to complete a typical residential unit.

Last year Greg Ip at the WSJ reviewed an analysis from Deutsche Bank economists suggesting that the illegal immigrant explanation accounts for most of the missing job losses:

[E]conomists at Deutsche Bank estimate construction employment should have fallen about 900,000 since early 2006 when in fact it’s only down 150,000. They conclude 500,000 of the unexplained gap is attributable to layoffs of illegal Hispanic workers.

To update the numbers, residential construction employment is off 477,000 from the peak, but there are still close to 1 million too many jobs based on starts and completions.

The uncounted illegal immigrant argument is important for the impact on the economy, but it doesn't seem to explain why the BLS employment numbers haven't fallen more. Although the BLS is missing the job losses for illegal workers on the way down, they also didn't count them on the way up either.

Although miscounted illegal workers probably explains some of the fewer than expected BLS reported job losses, there are two other explanations that make sense:

Some construction employees have moved from residential to commercial work, but they are still being reported as residential construction employees to the BLS.

Many workers are still employed, but they are working far fewer hours.

The answer is probably a combination of all of the above.