The Times’ Caucus blog has now returned to the subject of spending by independent organizations and the effect on the midterm election results. The post however leaves a great deal to be desired.
After all the talk about the impact of independent groups on this year’s midterm elections, the debate over at least one important question can now begin in earnest: Which groups were the most effective?
It is surely a question that donors who contributed giant sums will want to know, although an organization’s impact is not measured solely in wins and losses.
Nevertheless, a New York Times analysis of Federal Election Commission records found the American Future Fund, an Iowa-based nonprofit that reported spending just over $8 million on 25 House and Senate races, won in 76 percent of them, giving it the highest winning percentage among the biggest-spending Republican-leaning groups.
American Action Network, another nonprofit group that has been active this year, ended up with Republican victories in about 56 percent of the contests it invested in.
American Crossroads and Crossroads GPS, the two groups tied to Karl Rove, won just over 58 percent of their races combined; the U.S. Chamber of Commerce’s winning percentage when backing Republican candidates was just a few percentage points higher.
Another active Republican-leaning group, Americans for Job Security, saw Republicans it backed win in about two-thirds of its races.
Meanwhile, major Democratic-leaning groups, predictably, fared worse. America’s Families First Action Fund, a group financed mostly by wealthy liberal donors that was active in House races, won only about 18 percent of its contests; the American Federation of State, County and Municipal Employees did only slightly better, winning about 20 percent of its races.
Faring the best among the top-spending Democratic groups were the National Education Association, the teacher’s union, which won about 25 percent of its races, and the Service Employees International Union, which won about 29 percent of its contests.
The problem is that this assumes spending has an effect. Comparing what organizations won what percentage of the races they invested in might be good fun (I mean some people might find it is fun just to know that Karl Rove can’t always get what he wants!), but it does not actually answer the question of whether third-party spending is effective or not. Even the fact that there was variation in the percentage of races won by different organizations does not tell us anything useful for answering that puzzle.
Some might be quick to assume that the fact that there was variation between organizations means third-party spending does not matter. Yet, even that ignores critical questions like, would things have gone even worse in those races where organizations unsuccessfully spent for their preferred candidate? Or did the level of investment in those races not meet some essential threshold? Or were the organizations with less success outspent by competing organizations in the same race? Most importantly, would all these races have had the exact same result (or a close approximation) without a single dollar being spent? If not, how big is the difference?
Answering the question of whether third party spending effects (positively or negatively) election outcomes — whether it is possible to buy or heavily influence election outcomes — requires a much finer grade of analysis and more robust methods and measurement. And frankly it is probably not possible to come to a definitive answer.
So those that invested might like the outcomes of the races they spent money on. But that is fundamentally different than bang for one’s buck. The later implies that spending actually has an effect. This Caucus post just leaves us in the dark.