The latest US job figures [2] may have been better than expected but they're still far from encouraging. 117,000 new jobs were created in July but the unemployment rate was little changed at 9.1 per cent (down from 9.2 per cent in June) and even this fall can be largely attributed to the fact that more people have simply dropped out of the labour market. Perhaps the most sobering statistic, as the New York Times [3] notes, is that only 58.1 per cent of the US adult population is in work, the lowest level since 1983. Little wonder that the "relief" provided by the figures was decidedly temporary [4].

All of which doesn't bode well for Obama's re-election chances. No US president since Roosevelt in the 1930s has won re-election with the unemployment rate above 7.2 per cent, and it is estimated that by November 2012 it will be 7.8 per cent. So, how worried should Obama be? It all depends on context. FDR was able to win a second term because unemployment was falling, not rising. When he ran for re-election in 1936, unemployment stood at 17 per cent but this was still down from 22 per cent in 1934 and 25 per cent in 1932. The public were satisfied because the figures were moving in the right direction. Similarly, as the NYT has previously noted: "Ronald Reagan won, despite 7.2 percent unemployment in November 1984, because the rate was falling and voters decided he was fixing the problem."

Thus, Obama's challenge is to reduce unemployment to a level that voters, given the global economic context, are willing to tolerate. The Roosevelt precedent suggests that this could be significantly higher than 7.2 per cent and, in the absence of a credible Republican candidate, the smart money is still on Obama celebrating his 54th birthday in the White House.