Why the 2014 Senate Races Matter So Much
Yesterday I laid out some scenarios for the 2014 Senate races. At President Obama's current job approval rating, Democrats would likely suffer substantial losses in the upper chamber. On the other hand, if the president were to improve his standing, and if Republican candidates implode in some races, the GOP could break even or even lose a seat or two.
But aside from bragging rights, 2014 is a sideshow for the following two years. Obama is a lame duck, and the GOP already has blocking power with the House. A Republican majority in the Senate could provide added leverage by forcing Obama to use his veto, and it could force vulnerable Democrats to cast tough votes. But where it really matters -- where it could make a huge difference -- is in the next round of Senate elections.
Given the playing field that year, Republican losses are at the very least probable in 2016. In fact, they are likely. So if the GOP blows it this year, and if a Democrat wins the presidency in 2016, he or she could have a sizable -- perhaps even filibuster-proof -- majority. Moreover, if the GOP doesn’t win a decent-size majority this year, it stands a good chance of losing it in 2016 if Democrats have a good year.
Because 2010 was such a great GOP year, with Republicans winning seats in a deep-blue state like Illinois and in bluish-purple states like New Hampshire and Wisconsin, they will have an unusually high number of vulnerable seats in 2016. Similarly, Democratic vulnerabilities are limited to Nevada (where popular Gov. Brian Sandoval could challenge Harry Reid) and in Colorado. But Republicans failed to win those seats in 2010, and there’s hardly any guarantee that the playing field will be as favorable as it was that year, and candidates will be facing a presidential electorate to boot (which tends to be more favorable to the Democrats).
With the caveat that 2016 is a long way off, and that a lot could change, I identify the following seats where the GOP will have varying degrees of vulnerability: Florida, Illinois, New Hampshire, North Carolina, Ohio, Pennsylvania and Wisconsin. In addition, John McCain and Charles Grassley will be 80 and 83, respectively, on Election Day. If they were to retire, their seats could be quite vulnerable, depending on the national environment.
To try to get a better sense of this, and without getting into too much detail (e-mail me if you want more particulars), I’ve run 10,000 simulated elections in each of those states. The simulation randomly assigned outcomes for the presidential race (mean Republican vote share is 49 percent, including third parties; standard deviation is 2 percent), and calculated what vote share the Republicans would get in each state, using the 2012 elections as a baseline.
I then looked at the competitive Senate races for 2000, 2004, 2008 and 2012, and noticed that Republican candidates tended to run about two points behind the GOP presidential candidate (in only 14 of 41 races have they run ahead). Now, there’s quite a lot of variance here, but I went ahead and ran another set of simulations for each race, based on the presidential outcomes (mean = presidential outcome minus 2 percent; standard deviation = about 6 percent). I also built in possibilities for a recruiting failure/collapse (keyed off of how strong the presidential candidate is running) and whether McCain and/or Grassley retires.
After running these 10,000 simulations, I added up the total number of seats Republicans won and lost in each instance. The following chart gives the percentage of the time that Republicans lost or gained seats.
Truth be told, I think this estimate is a bit hard on the party -- for whatever reason, it gives Republicans a worse shot in New Hampshire, Ohio, and Nevada than I think is reasonable. But regardless, as you can see, the most likely outcome for the GOP is a loss of four or five seats -- about a one-in-three chance of that occurring. This isn’t as easy to see, but there’s about a 40 percent chance that the GOP will lose fewer seats, or even gain seats, and a little better than 25 percent chance that the GOP will lose more than five seats.
The following chart gets at this from a slightly different angle. It shows the cumulative probability that the GOP will lose at least “n” seats:
You read this chart from left to right. The basic idea is that there is a 1 percent chance that the GOP will lose nine seats in 2016. There is a 4 percent chance that it will lose exactly eight seats. Adding this to the 1 percent chance that it loses exactly nine seats gives us the chance that it will lose at least eight seats (5 percent). There’s a 9 percent chance that it will lose exactly seven seats, which means a 14 percent chance it loses at least seven seats. And so forth.
To pull this all together, the following chart looks at the readily conceivable outcomes for 2014 (assuming neither an astronomical increase in Obama’s job approval nor a catastrophic collapse), and then looks at the odds that the Democrats would gain control of the Senate (with both a Democratic and Republican vice president as a potential tiebreaker) and of getting a filibuster-proof majority:
As you can see, if the GOP wins a bare majority in 2014, the odds are very, very good that the Senate will revert back to Democratic hands in 2016. In fact, if GOP gains are confined to the “traditional seven” Democratic races (the three open seats and the four incumbents in states Mitt Romney carried), they’re still favored to lose the chamber two years later. On the other hand, if Republicans get to 54 seats, their chances of retaining control are very good, and given the horrific playing field for Democrats in 2018, they would be extremely unlikely to lose it that year.
Perhaps of more interest, if Republicans gain only a seat or two, a filibuster-proof Democratic majority in 2016 is at least plausible. If Republicans break even or lose seats -- and remember, no one thought that Republican losses were plausible at this point in 2012 -- a filibuster-proof Democratic majority might even be likely in 2016. A year good enough to net Democrats six or more Senate seats would probably given them control of the House as well, giving them an unlikely trifecta for the second time in eight years.