Bill Clinton’s campaign advisers are still right: For Barack Obama, it really is the economy, stupid. One respectable forecasting model puts the US president with a hair over half the two-party vote on November 6, 2012. A Breakingviews calculator shows how even slight changes in the economic growth rate before then could tip the result dramatically.
Among the best economic indicators for American elections are measures of personal income. The unemployment rate, which fell sharply in November but at 8.6 per cent is still worryingly high for a sitting president, is a favorite of pundits – but it’s not well correlated to election votes. The rate of change of unemployment may be better, but that’s probably already reflected in income growth.
Ray Fair of Yale University has developed a model that predicts the vote share reasonably well when tested on the 24 presidential elections since 1916. Its main inputs are the real growth rate of per capita GDP in an election year and inflation throughout the ending presidential term. Fair has a third input to incorporate the “feelgood factor” of short periods during the term when growth is particularly robust.
The Breakingviews calculator is based on Fair’s model, but simplifies it and turns the result into a probability of winning. Assuming 2.5 per cent annual GDP growth from the last quarter of 2011 until the election and inflation running just under 2 per cent, Obama’s share of the two-party vote comes out at 50.3 per cent. (This ignores the potential distortions of an independent candidate, and of the electoral college that actually chooses the president.) That slim-looking margin converts into a probability of winning of about 54 per cent.
For more evidence that things are on a knife-edge 10 months out, consider figures from Intrade. On Wednesday, the prediction market gave Obama a 52 per cent chance of re-election. His approval rating – around 45 per cent as of January 2, according to Gallup – seems to bode less well. It is early to read too much into that, and Obama’s approval has lately edged upward, but he will be in trouble if it stays too far below 50 per cent.
Breakingviews has built a “swing factor” into the calculator to incorporate users’ own adjustments. But the economy is the key. For instance, if growth hits 3 per cent, the Democrat Obama’s probability of winning veers to 70 per cent or so. Conversely, weaker growth than hoped could easily put a Republican opponent – let’s call him Mitt Romney given his Iowa win – in the White House.