President Barack Obama has repeatedly claimed that his economic
stimulus bill will "create or save" 3.5 million new jobs by 2011,
including 600,000 jobs "created or saved" by the end of this
summer. It is impossible to hold the President accountable to these
promises because there is no way of measuring "jobs saved."
However, the Bureau of Labor Statistics (BLS) does closely monitor
the state of the labor market, including job creation, job-loss
rates, and the total number of hours worked.
Job losses will slow and unemployment will stabilize at
some point in the business cycle--the economy will not continue to
lose hundreds of thousands of jobs a month indefinitely. For the
Obama Administration to claim credit for creating and saving
jobs, the labor market should have already started to show signs of
real improvement. This has not happened. Since President Obama
signed the stimulus bill, job creation has hit new lows and the
total number of hours that employees work has continued to fall.
The data do not support the President's claim that the stimulus has
"created or saved" jobs.
The economy has continued to deteriorate since President Obama
took office. In response, President Obama campaigned for, and won
enactment of, a massive $787 economic stimulus bill, primarily
composed of new spending on traditional liberal
priorities. The President has promised that this stimulus will
"create or save" 3.5 million new jobs by the end of 2010. The
President recently reiterated that promise, claiming that the
stimulus bill had already created or saved 150,000 new jobs and
would create or save another 600,000 jobs by the end of the
The President's calculation of 150,000 new jobs comes from
estimates by his economic advisers that assume that every dollar of
new government spending will boost the gross domestic product
and create new jobs. These estimates are unreliable at best, as
demonstrated by the fact that the authors would not vouch for their
accuracy. The President's advisers note that:
It should be understood that all of the estimates presented
in this memo are subject to significant margins of error .... Our
estimates of economic relationships and rules of thumb are
derived from historical experience and so will not apply
exactly in any given episode. Furthermore, the uncertainty is
surely higher than normal now because the current recession is
unusual both in its fundamental causes and its severity.
Such a tepid defense hardly inspires great confidence--let
alone supports the claim that the stimulus will "create or
save" 600,000 jobs by the end of the summer.
Analysts have also criticized the promise to create or save
millions of jobs as being deliberately amorphous. While
the Bureau of Labor Statistics measures job creation in great
detail, no government statistics measure "jobs saved." In order to
measure how many jobs a policy has saved, the BLS would have to
know how many jobs would have been lost had that policy not been
enacted. This involves assumptions about alternative futures that
are not knowable. No matter how many jobs are lost, the President
can claim that the recession would have cost 3.5 million
more jobs without his policies. No one can prove otherwise.
The President has knowingly chosen a metric that cannot be
verified and-- for which he, therefore, cannot be held
While it is impossible to measure how many jobs the stimulus
bill "saved," the BLS does measure job-creation rates, job-loss
rates, and the total number of hours employees work.
The fact that the labor market will stabilize as the recession
eases does not mean that the Administration can then
justifiably claim the stimulus bill improved the labor market. In
order to credit the stimulus bill with improving the economy, these
labor market indicators should show real improvement in the
The rule of thumb that forms the basis for the President's
claims says that increasing government spending by one percentage
point of gross domestic product (GDP) creates one million new
jobs. The government spent $56 billion of the
stimulus by the end of June 2009, with billions more already
obligated for expenditure. The President's estimates that
justified the stimulus imply that this spending should have created
400,000 new jobs. If these estimates are correct, these jobs
should be reflected in the data as higher job-creation rates, lower
job-loss rates, and longer hours worked.
Job Creation Continues to Drop
President Obama claimed that the stimulus would create jobs and
two BLS surveys shed light on job-creation rates. The Current
Population Survey (CPS) shows the proportion of unemployed
workers who find a job each month--the job-finding rate for
the unemployed. The Job Opportunities and Labor
Turnover Survey (JOLTS) shows the proportion of workers starting at
a new job each month. It reveals how many total new jobs
employers create and includes the unemployed who find jobs,
workers who switch jobs, and workers who re-enter the labor force.
Chart 1 shows rates since the recession began in December 2007.
Chart 1 explains why unemployment has risen so rapidly: Job
creation has fallen sharply since the recession began. At the start
of the recession, over one-quarter of unemployed workers found a
new job each month. By June 2009, that figure had fallen to below
Employers and businesses have cut back sharply on investment.
Equipment and software investment has fallen by one-fifth since the
recession began. In the current economic climate, many
businesses and investors have retrenched, holding off on
non-essential projects until they are more confident they will
recoup their investment. With less investment, employers are
creating fewer jobs, making it more difficult for the unemployed to
find new work. Employment will not increase to previous levels
until businesses and investors resume risk-taking and
The stimulus bill will not promote such job-creating
investment. Spending hundreds of billions of dollars on federal
projects does little to increase private-sector
entrepreneurship. The prospect of steep tax hikes to fund this
excessive spending also discourages wealth creation.
Since President Obama signed the stimulus bill into law in
February, the job creation rate has decreased. In February 18.9
percent of the unemployed workers found jobs and 3.2 percent
of all workers started at a new job. By May 2009, the new-hire rate
had fallen to 3.0 percent--the lowest level recorded since the BLS
began the JOLTS survey in 2000. In June 2009, the job-finding
rate for the unemployed stood at 18.4 percent. The stimulus
bill simply has not promoted job creation.
Layoffs Still High
President Obama also claims that the stimulus will "save jobs,"
while both the CPS and JOLTSsurveys report job-loss rates. CPS
data show the proportion of employed workers that lose their
job and become unemployed each month. JOLTS reports the
proportion of workers laid off or discharged each month. Chart 2
reports both job-loss measures.
Both measures show that job-loss rates increased sharply since
the start of the recession and have remained high since President
Obama signed the stimulus bill. JOLTS provides some evidence that
layoffs have decreased, showing that in February, 1.9 percent of
workers involuntarily left their jobs. By May 2009, that figure
fell slightly to 1.7 percent. The CPS data show no improvement,
reporting that 1.9 percent of workers employed in January 2009 lost
their jobs in February--a figure unchanged in June.
Both data sources show that, despite the President's pledge to
"save" jobs, job losses have--at best--slowed only slightly since
the stimulus became law.
Employers who are cutting back on labor costs may lay workers
off. They may also reduce employees' working hours. Many
employers have adopted the second approach. While layoff rates have
slightly improved, the latest data show that employers are
cutting back hours. Chart 3 shows the length of the average work
week and an index of the aggregate hours worked. Both have
fallen sharply since the recession beganat the end of 2007 and that
the decline has continued unabated since the stimulus was enacted.
This past June, the employment survey reported that the
average number of weekly hours worked had fallen to its lowest
record level in the survey's 45 years of existence.
Pro-Growth Policies Needed
The job-creation rate among private-sector employers will have
to increase before employment begins rising again. If the
job-creation rate had remained steady since the recession began,
employers would have created 900,000 more jobs a month this
year than they did. Instead of payrolls falling by 500,000
jobs, they would have grown by 400,000 jobs a month and the
economy would now be gaining--not losing--jobs.
Redirecting resources toward special-interest priorities
through massive government spending does not stimulate
private-sector job creation. Instead, it crowds out private-sector
While some want to changethe current stimulus bill by trying to
spend more quickly or even pass a new stimulus package, the 2009
stimulus bill shows that this would be a waste of money. Again.
Instead, Congress should pass policies that encourage investment,
risk-taking, and entrepreneurial activity. Congress should
reduce regulatory barriers, make the 2001 and 2003 tax relief
permanent, and cut unnecessary government spending.
Americans depend on it.
Sherk is Bradley Fellow in Labor Policy in the Center for
Data Analysis, and Rea S. Hederman, Jr., is Assistant Director of
and Senior Policy Analyst in the Center for Data Analysis, at The
Romer and Bernstein, "The Job Impact of the
American Recovery and Reinvestment Plan," p. 3.
government had paid out $56 billion of the stimulus bill by June
26, 2009--0.40 percent of the $14.09 trillion economy.
example, this rate is the number of workers with jobs who were
unemployed in the previous month divided by the total number of
unemployed workers in the previous month. Note that CPS data
contain a serious limitation when used to measure labor flows: The
CPS does not follow individuals when they move. So the CPS does not
record unemployed individuals moving to take a new job. These
individuals simply fall out of the sample and are ignored.
Consequently, the JOLTS data provide more reliable estimates of
changes in hiring rates, although the CPS also provides useful
Department of Commerce, Bureau of Economic
Analysis, National Income and Product Accounts, "Table 5.3.6. Real
Private Fixed Investment by Type, Chained Dollars," July 17,
2009 is the most recent month of JOLTS data at the time of the
writing of this paper.
That is the number of currently unemployed
workers who had a job in the previous month divided by the total
number of employed workers in the previous month.
The surveys report different results for two
reasons: (1) They are separate surveys using different
methodologies, and (2) not all workers who involuntarily lose their
jobs become unemployed and look for new work. Some leave the labor
force, such as retirees or parents who stay home with their