The jobs wars are back. A few years ago, the monthly jobs statistics produced by the Bureau of Labor Statistics suddenly became hot. As job growth failed to materialize when the economy began to expand in the fall of 2001, inquiring minds began to wonder whether Bush would be the first president since Hoover to see payroll jobs decline over a full term. Bush partisans, frustrated with the slow pace of job growth as determined by BLS's "establishment" survey, began to argue that another BLS report, its "household"survey, was a more accurate job gauge.
After the 2004 election, partisan fevers subsided, payroll jobs showed steady (if not impressive) growth, and the BLS data sank back into obscurity—until last week, when the jobs fight re-emerged in a different form. Now, the issue is: Who measures job growth better, a vast public bureaucracy or a joint venture between two successful private-sector organizations?
Earlier this year, ADP, the giant payroll processing firm, and Macroeconomic Advisers, the highly respected St. Louis-based forecasting outfit, rolled out their own monthly jobs report. Under the arrangement, MA's whizzes crunch numbers from ADP's huge client base—500,000 companies representing more than 20 percent of private-sector payroll employment. As this overview notes, they believe their methodology produces estimates better than the consensus forecasts compiled by news services and other organizations, and as good as the BLS figures over the long term. By releasing the data a few days before the BLS publishes its results monthly report, ADP/Macroeconomic Advisers could thus provide the public with an early look at the strength of the job market.
The report, they crow, "is likely to become one of the most—if not the most—closely followed economic statistics produced regularly in the private sector." And it has become just that. Last Wednesday, when ADP reported that the economy created a blowout 368,000 private-sector jobs in June, it sent the bond market into a tizzy—and right-wing econo-pundits like Larry Kudlow into ecstasy.
Perhaps based in part on this optimism, President Bush scheduled a splashy news conference for Friday morning in Chicago to coincide with the BLS jobs report. But as it has been so many times in the last several years, the government-issued payroll data were disappointing—just 121,000 total jobs (89,000 private-sector jobs) were created in June, far below ADP's projection. The tepid BLS numbers didn't stop Bush from boasting about them: "This morning we got some good news—the nation added 121,000 new jobs for the month of June. That's over 5.4 million jobs since August of 2003; that's 34 months of job increases. … People are better off, things are working."
Or not. The jobs figures are classic examples of misleading aggregates—large, economy-wide numbers that, shorn of context, are either confusing or empty. Who would say that the creation of 121,000 jobs, roughly equivalent to the entire work force of Best Buy, in a single month, isn't good news? But this isn't a static economy. The labor force grows by more than 150,000 people per month, as people graduate or leave high school and college, or immigrate, or return to the work force after absences. Bush talks about the creation of 5.4 million jobs in 34 months—or about 158,000 per month—as if it is a Herculean achievement. But since August 2003, job creation has basically kept pace with growth in the labor force. And for many months before that, it didn't. Keep in mind, too, that these jobs were only created with the massive stimulus of super-low interest rates (thanks, Mr. Greenspan), gigantic increases in government spending, and several tax cuts on individuals and businesses. As the Center for American Progress shows in this chart, job creation in this economic expansion has lagged job creation in other expansions.
ADP and Macroeconomic Advisers are, understandably, somewhat defensive about the large difference between their figures and those put out by BLS. On its Web site, Macroeconomic Advisers notes that it stands by its report. (It is assuredly not a conspiracy to make the Bush administration look good. Macroeconomic Advisers was co-founded by former Federal Reserve Governor Laurence Meyer, a Clinton appointee; its excellent staff has appointees from both parties. One of ADP's co-founders was Democratic Sen. Frank Lautenberg.) The new job predictors are confident they'll be proved right over time. A chart in the report shows that when they back-test the methodology, it tracks the BLS data closely over time. ADP/Macroeconomic Advisers notes that when BLS makes its annual adjustments to its job estimates, it pushes them in the direction of the ADP/Macroeconomic numbers. But the model has clearly misfired over the past six months. ADP/Macroeconomic Advisers say that since January 2006, some 1.3 million private-sector payroll jobs were created. According to BLS, however, that figure is only 581,000. While it may not be of much solace, the economics Ph.D.s at Macroeconomic Advisers aren't the only practitioners of the dismal science whose models have been upset by recent events. The current economic expansion, now in its fifth year, has shown that when it comes to jobs and wages, what held true in the past does not hold true in the present. A lengthy spell of robust GDP growth is not pushing up median wages or creating lots of jobs, as it has in the past. It's true that because of structural changes in the economy, the BLS model of surveying businesses to calculate payroll trouble may not capture all of the nation's job growth. But the ADP/Macroeconomic model, which also starts by examining payroll data, may not do much better.
More accurate or not, the new employment report does provide a political benefit to the president and his allies, just at the moment when campaigns are heating up. If the ADP report continues to issue optimistic assessments of job growth, expect calls from the American Enterprise Institute for the Bush administration to shut down BLS and outsource the job-creation figures to ADP.