The July jobs report is coming up Friday (we got a early look from ADP yesterday), so I thought I would do a piece on employment, particularly construction jobs.
Residential construction activity has come down sharply and yet the Bureau of Labor Statistics still reports next to zero job cuts in the sector. For a clearer picture of what is happening, I have indexed the number of jobs, the number of units currently under construction and the number of housing starts - see chart below (1985= 100, click to enlarge). Construction activity has fallen sharply, yet no significant job cuts are observed.
Residential construction activity has come down sharply and yet the Bureau of Labor Statistics still reports next to zero job cuts in the sector. For a clearer picture of what is happening, I have indexed the number of jobs, the number of units currently under construction and the number of housing starts - see chart below (1985= 100, click to enlarge). Construction activity has fallen sharply, yet no significant job cuts are observed.
Data: BLS, Census Dept.I have also produced a chart of two ratios: jobs to units under construction and jobs to housing starts. Reported jobs per unit being constructed are somewhat high, but it is the jobs per unit started that look abnormally high, i.e. by historical standards the BLS is reporting way too many such jobs still in existence for each house being started. If the jobs/starts ratio was the same as during the previous construction cycle low in 1991, it would translate into 806.000 jobs today vs. the June BLS report of 1.003.000 (seasonally adjusted). By this metric, the BLS is over-reporting residential construction jobs by nearly 200.000.
There have been several explanations offered for this discrepancy, most commonly that illegal immigrants are off the books and therefore their layoffs are not counted. While the home construction industry is a significant user of such workers, it does not explain the sudden jump in the ratios, i.e. if their numbers were increasing disproportionately vs. legal workers in recent years, this fact would have shown up in the ratios before 2005, as fewer workers per unit started.
I have another explanation: Builders have been holding on to their workers even when starts fell off rapidly, hoping for a turn-around in their business. They could afford to do so because the previous cycle was long and extremely profitable and they still had lots of unfinished work in progress. We see this from the jobs per unit under construction ratio which is still within "normal" parameters.
Nevertheless, given the deepening troubles in real estate, home builders must be getting very worried. If the current summer home sales season ends without any significant upturn, the charts above suggest we may see a much larger seasonal layoff come October-November - or sooner if builders decide to jump the gun. Those 200.000 "extra" jobs may disappear within 2-3 of months, putting a lot of additional pressure on consumer spending.
I believe we will be seeing this starting with Friday's jobs report.
Could it be that the numbers are distorted by the BLS using their "Birth - Death Model" which estimates how many companies started or ended in the period. It is seasonally adjusted, and since the input data for the last five years shows increasing employment, their model is as well.
ReplyDeleteAs a result, we see net jobs created in construction, that actually exceed the total in all sectors.
Cheers.
Anybody know where I can get information on what various mortgage products are trading for on the secondary market?
ReplyDeleteShorts are chattering about 60-70 cents on the dollar for plain vanilla ARMs and that the mortgage market is essentially closed right now. I'm guessing they don't have a clue.
Most of the "blow ups" seem to be concentrated in specific products (sub-prime, second lien, etc). If they are right, Wall Street might as well shut down Friday.
A lot of investment banks have AAA paper on their books that is not worth what they say its worth. I think they are trying to stave off problems and avoid marking it to market. How long can they avoid realizing the losses? What happens when they no longer can avoid it? Can they avoid it forever?
ReplyDeleteI like your chart on "Jobs to Housing Starts".From what I know from the street and being in the construction business in the Midwest that chart points to the "as you say" kafkaesque world the BLS is in.
ReplyDeleteBuilders and subs that I know and word of mouth info from others, the talk is most are seeing a drop off in work volume and a reduction in job $. A custom pond builder held out till the middle of last month for work and now is trying to sell his equipment to get out of the business. The jobs he had on the books from last year canceled and leads for this year dried up, also the wholesalers sent out notice of almost 30% price increases across the board that included the aggragret suppliers.Too bad there is a flood of equipment out there for sale now and not many buyers, he will be lucky if he can get enough to pay off his loan. In the mean time he has to squeeze out the payments some how.
The lumber yards have cut staff and parked part of there fleet of trucks.
I agree you will see a up turn in unemployment numbers of the hourly type soon but the self-employed will hit the wall this winter as they are scrambling for any work and will be slashing spending to stay alive with no savings to buffer themselves. Next year will be painful to watch IMHO.