To view a chart of job growth vs. electoral success, click here.
There is widespread discussion of what the September employment numbers mean for the future of the economy and the sustainability of the recovery. In September, job creation fell far short of expectations and of most reasonable standards when only 96,000 new jobs emerged. But there is also a good deal of discussion about what the numbers mean for the November presidential election. Only time will truly answer that question, but it could be useful to examine historical precedents.
Monthly employment data are available starting in 1939. During the current presidential term of office (Jan. 20 of the year following the election until Jan. 20 four years later) there are either 44 or 45 employment reports between the date the president is sworn in and the date of the election. This varies based on whether the date of the November election is late enough for the October employment report to be released. If the election is held on or after Nov. 6, there will be 45 reports.
Examination of the employment data indicates that prior to this administration, there were two occasions on which total employment in the United States grew at a rate of less than 1 percent a year based on the monthly employment reports leading up to Election Day. Those occasions were during the second Eisenhower administration, during which employment grew by 0.6 percent a year, and the first Bush administration, during which it grew by only 0.5 per year (as of Election Day). In both instances, the party that held the White House lost the election.
Prior to 1939, employment data is available only on a yearly basis. The data is an average for the year and should roughly reflect the employment situation in June or July of that year. If one were to average the employment level of the year a president assumed office with the prior year, one could approximate the level at the beginning of the year or close to the beginning of the term. Similarly, if the employment level of the year of the election were averaged with the employment level of the following year, one could approximate the end of year level for the election year. By doing this, we find that there were two other administrations during the 20th century in which employment grew by less than 1 percent annually. Those were the second Wilson administration, during which employment decreased by 1.1 percent a year, and the Hoover administration, during which it decreased by 4.7 percent a year. In both of those instances, the party that held the White House was defeated at the polls.
The September Employment Report from the Bureau of Labor Statistics at the U.S. Department of Labor indicates that 96,000 new jobs were created in the month of September, bringing total national employment to 131.6 million, or 821,000 fewer jobs than when the current administration took office. That would amount to an annual rate of decrease in national employment of 0.2 percent and would place this administration in the mid-range of the administrations that experienced less than 1 percent a year job growth.
The average growth rate for the 25 administrations prior to the current one totaled 2.1 percent. Six experienced average annual employment growth of between 1 percent and 2 percent. These included the Taft administration, the first Wilson administration, the second Roosevelt (FDR) administration, the first Eisenhower administration, the Ford administration and the first Reagan administration. Of these, only two saw their party lose control of the White House, the Taft administration and the Ford administration. Both clearly faced other problems more daunting that the relatively slower rate of job growth during their term of office. Taft faced a massive split in his party, with his Republican predecessor seeking to replace him as a third party candidate. Ford faced the backlash of the Watergate scandal and the ill feeling directed toward him as a result of his pardoning his predecessor, Richard Nixon.
The 15 administrations that enjoyed employment growth of more than 2 percent annually during their tenure experienced only a slightly greater probability of seeing their party retain office. Of these administrations, four were succeeded by the opposition party. They included the second Truman administration, which presided over a 2.8 percent a year rate of employment growth, the Johnson administration, which presided over a huge 4.2 percent rate of annual employment growth, the Carter administration, which presided over a 3.3 percent rate, and the second Clinton administration, which presided over a 2.5 percent rate.
Overall, administrations with annual employment growth rates of more than 2 percent had a 73 percent probability of being succeeded by an administration of the same party. Administrations with an annual employment growth rate of 1 percent to 2 percent had a 67 percent probability of their party retaining power. Administrations with less than a 1 percent annual growth in employment had a 0 percent probability.
These numbers provide statistical evidence of what many political observers would contend. Strong employment growth does not guarantee electoral success, but weak employment growth presents a very significant obstacle to electoral success in a presidential election.
The difference between somewhat less than 1 percent a year employment growth and growth that is only marginally above that may appear statistically small to account for such a strong dividing line between success and failure. One possible explanation is that at least in the period since World War II, it marks approximately the average rate at which the population is growing. Employment growth at less than that rate marks declining employment opportunities, family incomes and living standards. Employment growth above that level provides voters with a better chance of getting a job or of moving up to a better job and the prospect of some overall improvement in national living standards.
Scott Lilly is a senior fellow at the American Progress Action Fund.