Bad productivity numbers, or just bad numbers?

Last onh I blogged about the puzzling housing sector -- despite output slowing to a crawl, employment in that sector had not abated. Indeed, I made the following half-assed suggestion: This seems like a peculiar inverse of what was happening in the economy circa 2002-3 -- astounding productivity gains that were not matched by wage or employment growth. One wonders if this means that, for the next year, the U.S. economy will observe the obverse of marginal productivity increases but robust wage and employment growth. Economically, this makes little sense, but it did seem to be happening. In today's FT, Krishna Guha looks a little closer at this puzzle: A conundrum in construction lies at the heart of a US jobs market puzzle that continues to baffle economists ? including officials at the Federal Reserve. After a year of sub-par growth unemployment is a mere 4.5 per cent. With jobs growth strong but output growth weak, productivity looks very poor.... The Bureau of Labor Statistics payroll survey shows total construction employment and residential construction employment down just 2 per cent year on year in May, the latest month for which figures are available. The absence of the expected drain of net job losses in construction is the single biggest reason why overall job gains remain so strong ? 157,000 in May ? and unemployment remains so low.... There are a number of possible explanations. One is that companies are hoarding labour in expectation of a rapid rebound in the housing market. This looks increasingly implausible as the housing correction drags on. Another is that there is a time lag in construction and big job losses are just around the corner. There may be some truth to this. But the slowdown has already been under way for a long time. New home starts peaked in May 2005. The 12-month rolling average (new starts over the preceding 12 months) peaked at 2.1m in March 2006 and has since fallen to 1.6m. If it all fails to add up, the answer may be that the official statistics are not accurately capturing what is taking place in an industry that employs both a large number of small subcontractors and a large number of illegal immigrants. Specialty trade contractors ? who work for small subcontracting firms ? account for nearly two-thirds of all construction jobs. These workers tend to belong to small, often informal businesses. The payroll survey is likely to understate the extent to which these workers have switched from the residential sector to fast-growing commercial construction.... The separate BLS household survey does show a 300,000 increase in the number of people working part-time for economic reasons over the past year. The labour market statistics may also be missing a big decline in work by illegal migrants, who make up perhaps 20 per cent of the construction workforce.