The national unemployment rate is not good for the president’s reelection campaign. And if Team Obama digs a bit deeper, it will find something even worse, exactly where it matters – the swing states.
Jobless rates and their pace of improvement in seven battleground states have predicted the winner of the last three presidential contests involving an incumbent, a National Journal analysis reveals. While other factors are always at play in presidential elections, the correlation is worth considering. It shows that unemployment rates in these states may need to not only improve, but also to keep pace with the national average for an incumbent president to take that state’s electoral college votes.
That spells trouble for Obama in 2012, if the historical trend and current economic projections hold.
The unemployment rates in seven critical states – Colorado, Florida, Michigan, Nevada, Ohio, Pennsylvania, and Virginia – include some of the worst in the country. They are unlikely to show significant, or even statistically meaningful, improvement over the next year and most will fall demonstrably short of the improvement in the United States overall.
The single exception might be Florida, which should see its unemployment rate decline slightly more than the national average – or down 0.8 percent from the fourth quarter of 2011 to the fourth quarter of 2012 compared with the country’s anticipated 0.1 percent drop, according to December data provided by Moody’s Economy.com.
That’s a conservative forecast. The White House Office of Management and Budget predicted in its September Mid-Session Review that the national rate will drop to 8.2 percent in the fourth quarter of 2012. If that comes to pass, states will have an even tougher time keeping pace.
The rate of change in the other six states indices does not come close to Florida’s expected improvement, according to Moody's predictions. Even without major shocks to the economy, unemployment rates in all of them should remain statistically unchanged through 2012.
While Obama’s reelection team might take some comfort from Florida’s numbers, it should be tempered. The expected improvement in unemployment there also runs just shy of the percentage change considered statistically significant for that state. And Florida’s 29 electoral votes would not be enough to secure a second term if the president loses in the other states National Journal sees as battlegrounds.
Look at the math. National Journal’s Reid Wilson reports that Obama is likely to start his reelection campaign with at least 196 electoral votes from blue states; the GOP nominee can count on 180 from red states. Eleven true swing states remain, including the seven analyzed here, accounting for 162 electoral votes. Obama needs 74 of those to win.
Historically speaking, unchanged state unemployment rates correlate with no electoral votes for the incumbent. Just winning Florida’s votes—and not Colorado (9), Michigan (16), Nevada (6), Ohio (18), Pennsylvania (20), and Virginia (13)—is not enough for Obama.
Some pollsters argue only the national unemployment numbers matter. “National unemployment rate is the whole ballgame,” said Quinnipiac University Polling Institute Assistant Director Peter Brown. “There’s no one number more important than the national unemployment rate.”
But historical data does not bear that out. National Journal examined the last three presidential races in which an incumbent was reelected, focusing on the change in the unemployment rate in the year leading up to the incumbents’ victories. Voters appear to be especially impacted by the economic situation in their state in the year prior to the election, said Charles Prysby, a political science professor at the University of North Carolina-Greensboro.
In 1984, the unemployment rate fell by more than 1 percent in each of the seven battleground states, and President Reagan took them all.
President Clinton in 1996 carried five of the seven states National Journal examined. Florida, Michigan, Nevada, Ohio, and Pennsylvania all saw their unemployment rates fall more than the nation’s 1.5 percent decline, and all chose Clinton. The rate in Virginia fell 0.9 percent—well under the national rate’s change—and Colorado’s unemployment rate increased by 0.2 percent. Both went to Republican challenger Bob Dole.
In 2004, all of the states President Bush carried saw unemployment rate declines that met or exceeded the 0.5 percent national drop. The battleground states Bush did not carry -- Michigan and Pennsylvania -- fell short of the national mark with unemployment declines of -0.1 and -0.3 percent, respectively.
But the historical trend did not hold in two battlegrounds in 2004. Bush won Ohio and Virginia despite both states posting unemployment rates that improved less than the national average.
Still, in 16 of the 21 cases considered by National Journal, the incumbent saw victory in states whose unemployment rate declined at a rate that matched or exceeded the national one, and defeat when the rate was lower than that of the country. In the five states that went to the incumbent with a decline in unemployment that was lower than the national average, two were just 0.1 percent shy of it—and none saw gains in unemployment.
Prysby explained the correlation in the data: “How you see the national situation depends upon what’s happening in your state to a large extent.”