When policies fail to reach their stated goal, just move the goal posts. That is the obvious lesson of the new report from the White House’s Council of Economic Advisors (CEA) claiming that last year’s stimulus bill created or saved somewhere between 1.5 and 2 million jobs.
Defining Success Downward
A year ago, White House economists predicted that the stimulus bill would create (not merely save) 3.3 million net jobs by 2010. Since then, 3.5 million more net jobs have been lost, pushing the unemployment rate above 10 percent.
More specifically, in January 2009, non-farm payroll employment totaled 134.3 million jobs. The White House estimated that without stimulus, the number of jobs would fall to 133.9 million, while the stimulated economy would total 137.6 million jobs by fall 2010. The stimulus was enacted, yet the number of jobs has since fallen all the way to 130.8 million. This represents:
- A net loss of 3.5 million jobs overall;
- A net loss of 6.8 million jobs relative to the White House’s projection with the stimulus; and
- A net loss of 3.1 million jobs relative to the White House’s projection even without the stimulus.
The White House is spinning this failure by throwing out its earlier economic projections. Now it claims that, without the stimulus, the economy would have lost approximately 5.5 million jobs overall–or 2 million more than the actual 3.5 million job loss. With some tweaks to their internal economic models, the White House has defined success so far downward as to make 3.5 million job losses a success.
Analyzing the New Estimates
When analyzing the new estimates, it becomes clear that the “it would have been worse” argument is completely non-falsifiable and therefore ultimately unprovable. In addition, the White House report does not actually share its specific economic model, leaving the reader to take their word that the baseline economy would have performed so much worse without the stimulus spending. Of course, the White House’s admission that last year’s economic models were so far off should not inspire confidence in this year’s economic models.
Furthermore, a recent paper by economists John F. Cogan, Tobias Cwik, John B. Taylor, and Volker Wieland shows that the White House’s economic models are primitive and outdated. Even if one accepts the controversial Keynesian theories that have been used to defend the stimulus, the authors point out that the White House economic models used in January 2009 were based on “old Keynesian” models that do not adequately capture how individuals and firms respond to changes in economic policy and also often rely on unrealistic interest rate assumptions. The authors estimate that the more modern “new Keynesian” analysis shows economic and jobs impacts of one-sixth the size of those estimated by the White House last year. Any White House estimates based on old Keynesian economic models should not be taken seriously.
The problems do not stop there. The CEA study claims that $263 billion in stimulus spending has created 2 million jobs and added approximately 2.5 percent to the economy. Yet virtually all Keynesian models are based on simple multipliers assuming that every dollar of new deficit spending grows the economy by pre-determined figure. Thus, last year’s $1 trillion increase (from $455 billion to $1.417 trillion) in the budget deficit is the correct measure of Keynesian stimulus. The fact that only $263 billion of it came from legislation commonly called “stimulus” is not economically relevant.
Extrapolating out the CEA’s figures suggests that the full $1 trillion in new deficit spending over the past year has created nearly 8 million jobs and added nearly 10 percent to GDP. Since the economy actually lost 3.5 million jobs and contracted by 2.3 percent, these Keynesian theories suggest that the economy would have lost more than 11 millions jobs and contracted by more than 12 percent without this “infusion” of added deficit spending. That is simply not plausible, and few if any economists have claimed otherwise. One can therefore conclude that the White House is basing its estimates on multipliers that are way too large.
Why Government Spending Does Not Stimulate
The larger problem is that these Keynesian theories have long since been disproven. The repeated failure of government spending stimulus programs shows that governments cannot spend their way out of a recession. Every dollar Congress injects into the economy must first be taxed or borrowed out of the economy. No new purchasing power is created; it is merely transferred from one part of the economy to another.
For instance, the government’s Recovery.gov Web site claimed that $200 billion spent from the stimulus had financed 640,000 jobs. Even if that is true, the private sector now had $200 billion less to spend, which–by the same logic–must lose the same number of jobs, leaving a net jobs impact of zero. But this single-entry bookkeeping simply ignored that side of the equation.
Some respond that people are saving instead of spending, and government spending is therefore needed to unlock these savings. But redistributing money from savers to spenders does not add new demand, because savings do not fall out of the economy. They are invested or deposited in banks–which then lend them out to others to spend.
Even when recession-weary banks hesitate to loan money, they invest it in Treasury bills instead. They do not hoard customer deposits in massive basement vaults. One person’s savings quickly finances another person’s spending. Consequently, all income is applied somewhere in the economy, regardless of whether it is initially consumed or saved.
Removing water from one end of a swimming pool and pouring it in the other end will not raise the overall water level–no matter how large the bucket. Similarly, borrowing money from out part of the economy and redistributing to another part of the economy will not create new growth–no matter how big the stimulus bill. Thus, it is no surprise that the stimulus has failed to create net jobs.
If This Is Success, What Would Failure Be?
A recent directive from Office of Management and Budget Director Peter Orszag instructed staff tracking stimulus spending to replace the term “jobs saved or created” with “jobs funded.” This allows for the reality that stimulus spending does not create additional net jobs but rather funds the redistribution of jobs from one part of the economy to another.
Unfortunately, the CEA continues to argue that stimulus spending is creating millions of jobs–even as the economy loses millions of jobs. They do so by continually defining jobs success downward. One wonders just how many jobs must disappear before the White House would concede that the stimulus has failed.
Brian Riedl is Grover M. Hermann Fellow in Federal Budgetary Affairs in the Thomas A. Roe Institute for Economic Policy Studies at The Heritage Foundation.
White House Council of Economic Advisors, “The Economic Impact of the American Recovery and Reinvestment Act of 2009 Second Quarterly Report,” January 13, 2010, at http://www.whitehouse.gov/sites/default/files/microsites/
100113-economic-impact-arra-second-quarterly-report.pdf (January 13, 2010).
Christina Romer and Jared Bernstein, “The Job Impact of the American Recovery and Reinvestment Plan,” January 9, 2009, p. 4, at http://otrans
.3cdn.net/45593e8ecbd339d074_l3m6bt1te.pdf (January 13, 2010).
John F. Cogan, Tobias Cwik, John B. Taylor, Volker Wieland, “New Keynesian versus Old Keynesian Government Spending Multipliers,” February 2009, athttp://www.volkerwieland.com/docs/CCTW%20Mar%202.pdf (January 13, 2010).
For a longer discussion of why government spending fails to stimulate economic growth, see Brian Riedl, “Why Government Spending Does Not Stimulate Economic Growth: Answering the Critics,” Heritage Foundation Backgrounder No. 2354, January 5, 2009, at http://www.heritage.org/
Peter Orszag, “Updated Guidance on the American Recovery and Reinvestment Act–Data Quality, Non-Reporting Recipients, and Reporting of Job Estimates,” Office of Management and Budget, December 18, 2009, p. 11, athttp://www.whitehouse.gov/omb/assets/memoranda_2010/m10-08.pdf (January 13, 2010).