Thursday, April 9, 2009

Recession in the Metro Dayton Construction Industry

Construction is part of the “goods producing” sector that we looked at in the previous post. While it’s common knowledge that manufacturing is declining, less is known about local construction employment, or how the recession appeared in this industry.

There are two ways to look this. One way is via the number of building permits.

The census maintains a building permits database for various localities that is quite current, with the most recent numbers from February. The database has a series for the unincorporated parts of Greene County, which is was one of the faster growing areas of the metro area.

Graphing from 2005 to the first two months of 2009 this is monthly volume of single family permits (and you might have to click on the charts to enlarge it for better reading):



Next, taking the averages of the years (but omitting those big spikes as outliers) one can see a steady drop from 2005 in annual average of permits, and how the 2008 high-low range was mostly below the 2005 & 2006 lows. But things deteriorated, with 2007 being pretty weak and 2008 weaker still:


So one can see the onset of decline perhaps in 2007 or even in late 2006.

That was Census data. Next will be Bureau of Labor Statistics data, which is by metro area. Now, looking at metro area construction employment for the same time frame, one can see a pretty obvious seasonal cycle, as on would expect in construction, especially in civil works, excavation and building framing as winter weather would impinge ability to do outside work.

As with the permits one can see some deterioration in late 2006 and 2007 already



Next, comparing the annual cycle, the summer high is pretty obvious in all the years, but one can see how things went sour in 2008, compared to 2007, which did build up to a high closer to the previous years after a weak start.


But 2008 started out weak too, but never recovered. After May employment flattned and then declined following previous years' pattern, dropping to a much lower level in January compared to previous years.


And finally, the long term trend in construction employment, which is pretty dismal for the 2000s. Employment in this sector never really reached it’s pre- 2001 recession highs, and was, in fact, deteriorating since the mid 00s:



..which seems to indicated a weak economy here even prior to the recession.

What will be interesting will be to measure the increase in construction employment due to the stimulus program. Presumably this will show up on the BLS statistics in various ways.

Perhaps a better measure will be suburban building permits (as in the Greene County example), as this type of construction won't be impacted by stimulus spending, so true recovery might be reflected in increased permits reflecting an uptick in house construction.

2 comments:

"TheDonald" said...

Jefferey, are there concise sources for this kind of data extending back 20+ years? My subjective experience is that (at least) professional employment in Dayton has been remarkably shaky and in steady decline since the late 70s. There are no excuses whatsoever for this - it is a long term, "secular" failure of local leadership. In IT things were just getting OK in the late 90s locally when the dot com boom topped out. I've literally never seen good times in this town since I was a kid in the 60s. I instinctively don't bother submitting a resume to any company with an address in Montgomery County - it's simply no use, employers expect you to beg like a dog for a job there and there's always a glut of underemployed.

Jefferey said...

Jefferey, are there concise sources for this kind of data extending back 20+ years?

The BLS has the informationg going back to 1990, by month and annual average. There was a change in classification from SIC to NAICS in 1997, I think, so not sure how comparable the time series is on either side of that change.

@@@

One of the reasons I chose to look at construction is because it isn't subject to automation the way manufacturing is, and has some relationship to the underlying economy.

Perhaps the same for retail trade? I didn't post retail trade employment stats/graphs for the metro area but did run them. Employment was dropping over this period, too.