U.S. Workplaces Have Never Been Safer but Suicides Increased During the 2008-2009 Recession

As we approach the Labor Day weekend it is interesting to examine the latest data on workplace violence and workplace safety.  The Department of Labor just released preliminary figures for fatal work injuries in 2012.  The good news is that workplaces in the U.S. have never been safer.  Workplace fatalities are down and workplace violence is lower than it has been in decades.  The most distressing pattern in recent data has been the increase in suicides in the workplace, apparently a result of the 2008-2009 recession.

The workplace fatal injury data for 2012 released last week will be revised upward when the final fatality rates are recorded.  If the upward revisions are consistent with recent years, the rate of fatal work injuries will be about 3.3 per 100,000 full-time equivalent employees, the lowest rate since the BLS began reporting these figures in 1992.  As the following chart shows, the workplace fatality rate is 36.5% lower than it was two decades ago.

Transportation accidents continue to be the most common cause of a workplace fatality and account for 41% of all fatalities on the job.  Workplace violence is the second most common cause of fatalities and account for an additional 17.5% of deaths.  Within the workplace violence category, there are more than twice as many homicides as suicides.

Workplace deaths from homicides and transportation accidents have been trending down over the past 20 years.  Transportation accidents have been declining by about 1.4% per year and homicides at work have been declining by about 4.3% per year.  As the following charts indicate, over the past decade the drop in both homicides and motor vehicle accidents has been more pronounced in the workplace than in the general population.  (I have normalized the fatality rates per adult to be 100 in 1992 for all categories of causes of death.)

Suicides in the workplace declined slightly from 1992 to 2002 following the pattern in the general population as the following chart shows.  More recently suicides in the workplace declined more sharply 2002 to 2007 but then increased rapidly during the 2008-2009 recession.

The following chart shows the relative increase in the number of suicides at work during the recession and the declining trend  in the number of transportation accidents and homicides in the workplace over the past two decades.  (I normalize the number of workplace fatalities to be 100 for each category in 1992.)

A simple statistical analysis suggests that there were about 153 more suicides in workplaces as the economy struggled from 2008 to 2010.  At the same time the drop in economic activity meant fewer fatal transportation accidents.  My rough calculation suggests that there were 749 fewer deaths from traffic and transportation accidents from 2008 to 2010 than there would have been if the economy had continued to grow.  In general the recession reduced fatal workplace accidents.

Despite the good news that workplaces in the U.S. are safer than ever, Secretary of Labor Tom Perez is not satisfied.  He issued a statement last week that “no worker should lose their life for a paycheck.”  I am sure that Secretary Perez also would like no U.S. resident to lose their life in any accident, whether they are being paid for their time or not, but that is an unrealistic goal.  Mr. Perez also finds the increase in fatalities in the oil and gas extraction industry unacceptable.  Clearly, many of the men and women who work in oil and gas extraction are risking their lives every day they go to work, but the 138 deaths in this industry last year need to be put into perspective.  Over the past two years fatalities in oil and gas extraction averaged 125 per year, or 22% higher than in the previous 8 years.   This increase is smaller than the 27% increase in employment in the industry over the same period.

As more people find jobs and shift from part-time work to full-time work, we can expect workplace fatalities to increase in many industries and sectors.  This does not mean that workplaces are becoming less safe – in fact workplaces are safer than ever.  In 2012 there were about 36.5% fewer workplace fatalities per full-time equivalent employee than just 20 years ago.  The downside is that some of the increase in safety has been achieved by outsourcing riskier jobs and replacing workers with machines.  As is the case in many economic issues labor faces a tradeoff between workplace safety and employment security.  If organized labor demands exceedingly high levels of safety in U.S. workplaces the U.S. will be more likely to lose jobs to foreign competitors with more lax safety standards and will be more likely to replace workers with machines and robots.

2.8 Million Construction Jobs Are Missing

In Saturday’s Wall Street Journal Timothy Aeppel wrote an interesting article noting that manufacturing employment has recovered only 22% of the nearly 2.3 million jobs lost during the 2008-2009 recession.  While this is true the U.S. economy and labor market can flourish despite declining manufacturing employment.  In fact, declining manufacturing employment has been the norm in the U.S. for decades.  Manufacturing employment peaked at 19.5 million jobs in 1979.  Since then the manufacturing sector has lost 7.5 million jobs as 53 million jobs were created outside the manufacturing sector.

The decline in construction employment over the past five and a half years, however, is troubling and atypical of recent economic recoveries.  Normally investment in residential housing by households and structures by businesses increases rapidly in recoveries as these investments were deferred and delayed during the recession.  A comparison of the pattern of construction employment in this recession and recovery to comparable periods during and after the 1981 and 2001 recessions reveals some clear patterns:

• Construction employment declined more than three times as much (26%) during the 2008-2009 recession as it did during the 1981 recession and more than 11 times as much as during the 2001 recession.
• Construction employment is about 1.7 million (22%) below its pre-recession peak five and one half years after the recession began.
• Five and one half years after the beginning of the 1981 and 2001 recessions construction employment had increased to about 15% above its pre-recession peak.

The following chart compares construction employment across the 1981, 2001 and 2008-2009 recessions normalizing employment to be 100 at the start of each recession.  The chart tracks construction employment in each recession/recovery for the subsequent five and one half years.

The magnitude of the decline in construction employment during the 2008-2009 recession was unprecedented.  The absence of a recovery in construction in employment during the subsequent recovery is also unprecedented.  Had construction employment rebounded over the past few years as it had in previous recoveries, there would be 8.6 million workers in the construction sector instead of the current total of 5.8 million construction employees.  The net difference represents a shortfall of 2.8 million jobs in the construction sector.  Construction employment remains well below the levels of six years ago because businesses are investing less in structures than they did in previous recoveries and the residential housing market has not recovered as strongly as it has in other recoveries.

The shortfall of 2.8 million construction jobs is equivalent to the difference between the current unemployment rate of 7.4% and a 5.6% rate.  Of course, if the economy were strong enough to generate an additional 2.8 million construction jobs there would also be more robust growth in income and employment in other sectors of the economy lowering the unemployment rate even further.

The sharp decline in employment in the construction sector since 2008 and the fact that only 8% of construction jobs have been regained during this economic recovery is extremely troubling.  Construction differs from manufacturing because jobs can’t easily be outsourced to foreign countries.  While manufacturing’s share of total employment has declined steadily for decades, construction’s share of total employment grew steadily for three decades from 1978 to 2008 until the sharp decline over the past five and one half years.  The following chart illustrates the sharp drop in construction employment since 2008.

In my view the continued weakness in the construction sector underscores a failed opportunity for those who advocate public investment in infrastructure.  The past five and one half years have seen an unprecedented slowdown in construction activity and employment.  Over two million construction workers lost their jobs in the deep recession and have remained displaced through a tepid recovery.  At the same time we have not chosen to re-build and repair our roads, bridges and infrastructure while there is an excess supply of construction labor and interest rates are low.  Government expenditures have been diverted from infrastructure to other programs such as a record extension of unemployment insurance benefits (over one billion weeks worth of UI benefits have been paid since 2008).

More importantly, the fact that the recession officially ended four years ago yet construction employment remains in a slump is a clear indication of the weakness of this economic recovery.  As the recovery officially enters its fifth year households are still unwilling or unable to purchase new houses and businesses lack the confidence and demand for their products to invest in offices, factories, warehouses and other facilities.  Until the construction sector rebounds because households and businesses are willing and able to invest in structures this recovery will continue to disappoint.

All of the Increase in Part-Time Jobs in 2013 is Due to the Changing Employment Patterns of Women

The household employment data from Friday’s BLS Jobs Report reveal a surprising and distressing pattern.  The full-time employment of women has declined by 119,000 (about one-quarter of one percent) between December 2012 and July 2013 (based on the seasonally adjusted series).  In contrast, the part-time employment of women has surged since December 2012.  The employment situation has been quite different for men, however.  Full-time employment of men has increased by about one half million since December 2012 while part-time employment has dropped slightly.

While it is true that for men and women combined more than three-quarters of employment gains in 2013 are from part-time jobs, this is due only to the changing employment patterns of women.  The surge in part-time employment in 2013 is due to women shifting from full-time to part-time jobs.

Many pundits have surmised that the shift from full-time work to part-time work in 2013 is a consequence of the Affordable Care Act (Obamacare).  If this hypothesis is true, Obamacare has had a much bigger impact on the employment patterns of women than men.  This is consistent with the notion that women’s labor supply is much more elastic (price responsive) than men’s and that women are employed disproportionately in jobs and sectors where part-employment is a viable alternative to full-time work.

The following table presents the data for employment changes in terms of number of jobs as well as percentage changes from the BLS household survey.

Group	Change in Employment December 2012 to July 2013	Percentage Change in Employment December 2012 to July 2013
Women Part-Time	+597,000	+3.43%
Women Full-Time	-119,000	-0.24%
Men Part-Time	-36,000	-0.35%
Men Full-Time	+504,000	+0.77%