weblog/wEssays     archives     home

What Happens When Housing Employment Plummets?   (June 13, 2006)

UCLA's Anderson Forecast predicts 800,000 jobs will be lost as the housing bubble bursts. A look at the actual numbers suggest they've grossly under-estimated the real damage to come.

Here is what CNN's story There go 800,000 jobs out the door had to say:
The report, released Wednesday, forecast a loss of 500,000 construction jobs and 300,000 jobs in the financial services sector from the housing slowdown, but noted that job losses would not spread across many industries, other than some limited pullbacks in manufacturing.

Its research shows that nine of the previous 12 drops in spending on residential housing were followed by recessions, and the latest report forecasts a decline in spending ahead.
Before we accept the improbably low number of 800,000 jobs lost, let's take a look at the size of the construction industry and its subsidiaries. There's no better place to start than the Bureau of Labor Statistics job report:

  • total non-farm workforce: 134 million
  • total private (non-government): 112.5 million
  • direct construction: 8,053 (in thousands, e.g. 8 million)

    out of total Durable goods employment of 8,970:
  • Wood products 558.9
  • Furniture and related products 558.0
  • Miscellaneous manufacturing 655.8
  • Fabricated metal products 1,526.7

    out of total retail employment of 15,300:
  • Furniture and home furnishings stores 583.0
  • Building material and garden supply stores 1,290.9

    out of services employment:
  • Real estate and rental and leasing 2,154.5
  • Architectural and engineering services 1,335.6
  • Financial activities 8,223

    To get a picture of how many construction jobs fall into "heavy construction" such as infrastructure projects, we can guessimate that employment by ascertaining how many union workers are in the construction workforce. Since heavy construction work is largely union, while residential is largely non-union, these numbers from the Associated General Contractors job report tell the story:
    BLS reported on January 20 that Union membership fell from 1,110,000 out of 7,550,000 construction employees (14.7%) in 2004, to 1,057,000 out of 8,053,000 (13.1%) in 2005. The median weekly earnings of full-time wage and salary workers in construction rose from $893 in 2004 to $933 in 2005 (4.4%) for union members and from $588 to $590 (0.3%) for nonunion workers.
    So let's put the numbers together. It's not that easy to find hard data for every subfield, so some guesstimates have to made. The BLS doesn't break out mortgage-related employment, so we'll need to estimate the percentage of financial services jobs which will be affected by the bubble's demise. Clearly, architectural jobs in the heavy construction field are not dependent on the bubble (though as tax receipts dry up, even publicly funded construction will inevitably be slashed). It's also difficult to reach a total of housing-related manufacturing jobs, for this includes everything from air conditioning and heating systems to carpets and appliances.

    On top of that, we also have to consider how many jobs in transportation (trucking, rail and shipping) are related to moving the vast quantities of building materials and other housing-related goods around the nation. After taking all these factors into account, it seems at least 18 million jobs are directly or indirectly dependent on the housing sector. That's 16% of the entire non-government workforce.

    Given the 7 million non-union jobs in construction and another 8 million in housing-dependent fields, the UCLA report seems absurdly optimistic. the 500,000 jobs the report estimates will be lost in construction means the collapse of the housing bubble will leave fully 93% of the construction workforce fully employed? Does that sound realistic? The report forecasts a scant loss of only 800,000 jobs out of the total housing-dependent wrokforce of 18 million, a mere 4.4%. Does this sound realistic?

    Let's look at the realities of the housing-dependent industries:

  • When housing tanks, it doesn't decline; it dries up and blow away. With builders already sitting on unprecedented inventories of unsold houses, does it really make sense to think housing starts will decline by 10% and construction employment will decline by only 7%? Experience suggests housing starts will fall by half or more, and employment will track that decline. If housing starts fall in half, then we can expect 3.5 million jobs lost in residential construction, not 500,000. Frankly, you have to wonder why developrs would go to the trouble of building any new homes when the homes already for sale and in the pipeline are already so far ahead of demand.
  • The same is true of the mortgage and real estate industries. They don't decline by 7%, they dry up and blow away. Sure, there will still be realtors and mortgage brokers, but their ranks will be thinned by at least a third (a conservative estimate). That would likely total well over 1 million job losses in those industries alone, three times the absurd UCLA estimate.
  • The same is also true of sales of carpeting, appliances, furnaces, doors, windows, etc. A sharp decline in housing starts will immediately impact sales and jobs Yes, people will still be buying new doors, windows and countertops for remodeling, but with their house value declining monthly, how many people will undertake an expensive remodel once they grasp it won't reverse the decline in their house value? The obvious answer: Not many.
  • As the first chart shows, construction jobs have been a significant factor in employment throughout the nation, not just in bubbleland locales like Florida and California. If construction jobs are so important in Montana, imagine what they mean to Florida.
  • Unlike the consumer electronics industry, most construction-related manufacturing is domestic. If sales of iPods, for instance, fall in half, that will marginally hurt store sales at retailers and shippers and workers in the Asian factories where the gizmos are made, but the only real damage would occur in Apple HQ in Cupertino, which employs a few thousand people. In other words, a drop of 50% in IPod sales would have a negligible effect on U.S. employment.

    Compare that to a 50% drop in construction spending. You're not just taking out the onsite construction jobs but the folks making the carpets, the truckers hauling lumber, the millwork employees, the door manufacturers, the furniture and appliance workers--the list goes on and on. While many appliances are now made in Mexico, the majority of the products and services which go into residential housing are domestic--meaning when the drop comes, it will have a devastating effect on a huge swath of American industries and their employees.

  • Construction is the highest-paying employment in the nation, so the loss of those jobs will heavily impact consumer spending. Estimates vary regarding the "downstream" jobs created in a community from high-paying construction jobs, but many estimates are 3 to 1, meaning a well-paying construction job creates 2 or 3 jobs in the community as those wages are spent on good and services.

    If you look back at chart 1, you have to ask: will the "leisure" industry be adversely affected by housing-related job losses? Only a person in the grips of a pathological denial would claim that the "leisure" and other industries which are dependent on discretionary spending will not be affected by a sharp decline in construction-related employment.

    And then there's tax receipts. What happens to government employment when all those transfer fees, payroll and capital gains taxes decline? Clearly, the gravy train of lush taxes which has been pouring into local government coffers will dry up. And since most local government can't borrow money to make payroll (California excepted, of course), then government employment will also decline along with tax revenues.

    Put it all together and an estimate of 5.6 million jobs lost seems more realistic than a mere 800,000. That works out to 5% of total private employment in the nation. Does 5% of the total workforce losing their jobs seem extreme? If you accept that 3.5 million construction jobs out of 8 million are likely to vanish, and a million real estate and lending-related jobs will disappear, that's already 4.5 million. If you consider the millions of manufacturing and retail jobs which are housing-dependent, then it's no stretch to reckon another million jobs will be lost as housing starts and renovations plummet. And there's your 5.6 million jobs lost. As those paychecks vanish from retail and leisure outlets, doesn't it make sense that employment in those areas will fall along with sales? When you consider the entire chain of spending which flows from housing, it's not hard to imagine even more jobs falling by the wayside as consumer spending falls.

    And if 5 million people lose their jobs, what happens to the economy? Recession. The UCLA report got one thing right: Its research shows that nine of the previous 12 drops in spending on residential housing were followed by recessions, and the latest report forecasts a decline in spending ahead.

    For more on this subject and a wide array of other topics, please visit my weblog.


    copyright © 2006 Charles Hugh Smith. All rights reserved in all media.

    I would be honored if you linked this wEssay to your site, or printed a copy for your own use.


      weblog/wEssays     home