What was the impact of outside spending and especially “dark money” in the 2014 election? A Yale University political scientist sent me two graphs that he made from early Senate returns and these data on outside spending. He writes:
The basic takeaway is this. After the GOP victory, there is even more attention to the Super PACs and concern of their influence. This attention seems misplaced. Check out the Senate graph below:
Graph by Eitan Hersh
There is no relationship between which side spent more money and vote share. In the close races, both sides spent a lot of money. Republicans won in places where they were outspent by the other side (NC, CO, KS, SD) and they lost some places where they were outspent by the other side (NH, OR). The correlation between spending advantage and vote share advantage is 0.02. There’s nothing there.
The House graph is similar:
Graph by Eitan Hersh
Most of the money is concentrated in very close races. There is a stronger correlation here than in the Senate, but there are plenty of races in the upper left and lower right quadrants.
What kind of picture would we be concerned to see? I think if we saw a lot of races in which one side way outspent the other side were in the upper-right and lower-left quadrants, that would make us wonder about the whether the money affected the outcomes. But we see all four quadrants here populated with data points.
Nothing about this bivariate relationship gets at the real effect of the money, but the picture doesn’t seem to me to be consistent with a story of hugely influential money.
Of course, the concluding sentence is an important caveat (for more, see this), but it seems worthy to factor even this preliminary analysis into the (sometimes overheated) conversation about outside spending in elections.