KEEPING THE HOUSE, PART 2: What can Democrats do to overcome the GOP's built-in advantage and keep control of the House of Representatives?

By Dan Drasin

In Part 1 of this post, I made the argument that the Democratic hold on the House is much more precarious than it might otherwise seem. The crux of the discussion is that by looking at the Partisan Voter Index (or PVI, see below) of the districts, we can see that Republican voters are much more efficiently distributed, and so in an evenly divided election, they have a 40-seat starting advantage. As promised, I want to suggest now some things that the Dems can do to minimize this advantage.

PAUSE FOR A QUICK RECAP ON PVI

The PVI is a measure of the partisanship of each House district relative to the rest of the country, based on the last two presidential elections. A PVI of +2R means that a Republican candidate in that district is likely to get 2% more of the popular vote than the Republican share of the vote nationwide. So if in a particular election year the Republicans get 49% of the vote, then this candidate would (all other things being equal) get 51%.

1. GET MORE VOTES AND VOTERS

The most obvious tactic is to keep getting a much larger share of the vote.

(1a) Keep the "wave" going

There is considerable evidence from exit polls that the 2006 election wasn't just about "throwing the bums out" but also represented a fundamental realignment of Democratic vs. Republican voting patterns. In order to maintain control of the House, the Dems will obviously need to maintain their edge in the popular vote. In fact, there still are a number of takeover targets that almost fell in this election and will be likely to fall in the next one assuming a strongly pro-Dem environment: CT-4, DE-1, NY-25, PA-6, PA-15, WA-8, NM-1, to mention just a few.

So it is vitally important for practical (as well as ethical and moral) reasons that the Democrats demonstrate that their governance is effective and valuable for America as well as contrast it with the disaster of the last 12 years of GOP congressional leadership. That means not only passing popular programs, but also highlighting the failure of previous congresses and the Bush administration.

This last part is crucial, because dissatisfaction with Republicans was a key component to the 2006 election outcome, and unless reminded, people's memories tend to get fuzzy. In two, four or six years, absent a concerted effort by Democrats, many people will remember GOP control as "the good old days." (For those of you who think it's impossible, consider how Newt Gingrich, after slinking away from the House speakership following his party's poor performance in the 1998 midterm election, is now rehabilitated enough to be considered a 2008 presidential contender.)

(1b) Get out the vote--for challengers

A less-known effect of high voter turnout is that it tends to be bad for incumbents. It's usually thought to be the opposite, but in fact low voter turnout tends to be good for the incumbents.

The Democrats should focus GOTV efforts especially in districts where they are working to unseat a Republican. Especially good targets are those districts that already have a Democratic PVI--i.e., FL-10, NV-3, NY-13, NM-1, NY-3, PA-6, PA-15, WA-8, NJ-3, NY-25, NJ-2, CT-4 and DE-1. And given that 2008 is a presidential-election year, in some of these races a strong GOTV effort can yield a nice two-for-one return.


2. RETAIN PRESENT DEMOCRATIC SEATS

The next important tactic is to focus on retaining seats that are Democratic in the 110th Congress. Even allowing for all the factors that I argued in Part 1 favor Republicans, incumbency is definitely an advantage. Studies show that, on average, it translates to an 8% electoral advantage.

(2a) Limit retirements of incumbents

To retain the incumbency advantage, Democrats need to limit the greatest threat to incumbency reelection: retirement.

With the Democrats just taking the reigns of power, a slew of retirements probably isn't likely. After all, who wants to throw in the towel just when they've finally got some pull? However, in the 46 Republican-PVI districts, there are six (TX-17, MS-4, MO-4, AL-5, MN-7 and WV-1) where the incumbent has been in office for more than 15 years--a magic number where retirements start happening. (In statistical terms, it is +1 standard deviation beyond the mean number of terms served.) These districts would be hard for Democrats to hold if the seat were open.

Of course, the principle of limiting retirement maximizes the Democratic advantage for ALL races, but it’s particularly important in the races where the PVI is +5Dem or less--and there are 75 of those seats.

(2b) Protect the freshmen of the 109th and 110th Congresses

Studies show that a congressman is by far most vulnerable in the first two election cycles after he/she is first elected. (Interestingly enough there is also a statistically significant increase in vulnerability between the 8th and 14th terms.)

There are about 35 Democratic House members from the Classes of 2004 and 2006. And of those, a troubling number are in Republican PVI territory: 23 seats ranging from TX-22's 15R down to a cluster of 1Rs that will be doubly vulnerable in 2008, more than enough to tip the House back to the GOP. So the party needs to work hard to bolster these House members--with money, choice committee assignments, pork, independent expenditures, etc. Support needs to come early and often.


3. WIN MORE SEATS VIA THE 50-STATE STRATEGY

This is probably the most important point because the ones outlined above are considered standard strategy.

This election demonstrated that in the right environment, Democrats can win and retain seats anywhere in the country, even in districts that have a long history of voting Republican. In a previous post I focused on the criticality of the "50-state strategy"--fielding candidates in virtually every race--in terms of the Democrats taking advantage of the strongly anti-GOP leanings of the electorate in 2006. The main point there was that 18, 12 and even 6 months out, it's hard to predict which races will be the most favorable, and so it's important to hedge your bets by challenging, at least to some degree, everywhere.

But the other important stipulation of the 50-state strategy is not only to challenge with quality candidates but to work hard at the levels of state and local races to build up a "bench" of quality candidates. Studies show that experienced challengers pick up an average of 2 points, so having a quality candidate helps eat into the incumbent advantage (no surprise).

In one race, 2% may not seem like much, but when applied across the board, as the 50-state strategy envisions, this brings 11 more races into Democratic-leaning territory. In fact, one study even suggests that one of the key components to incumbency advantage is the ability of incumbents to limit the quality of the challengers they face. This suggests that the 50-state strategy can provide tools to expand the Democratic House majority and overcome the 40-seat pro-GOP districting bias.

With that said, here's the most critical aspect of the 50-state strategy, one that makes it so much more powerful for the Democrats than for the Republicans. If we look again at the chart that shows the distribution of House district PVI margins:
we can see an opportunity. In Part 1, I pointed to all the "wasted PVIs" for Dems, above +15D. But we can also see here the large number of Republicans at 0-5R. When Republicans win these seats, they represent a much more efficient use of their votes, as it means that on average the same popular-vote percentage will get them more seats than the Dems. However, this also means that the GOP has much less margin for error.

Every one of the Republicans' 77 seats in the 0-5R range (almost twice the number of similar Dem seats) should be constantly targeted by the Dems. As this year demonstrates, if you're challenging seats, you may catch a good wind, or a scandal, or an unexpected retirement (or indictment) and pick up seats.

The 50-state strategy amounts to a way to turn the challenges I described in Part 1 into opportunities.


4. MAXIMIZE INPUT INTO REDISTRICTING

Finally, it would be blind to ignore the impact that Democratic-controlled redistricting could have in a number of states. Leaving aside the question of when and under what circumstances redistricting should occur, analysis of PVI shows that there are a number of states where the Democratic majority in popular vote is not represented in the House membership because of the way that Democratic votes are concentrated. Whether this concentration is purposeful or accidental, it has the effect of diluting the power of Democratic voters, and so a reasonable argument could be made in the redistricting process that his imbalance deserves to be rectified.

Here is a short list of states that fall in this category (remembering, of course, that control of redistricting depends on control of state legislatures and governorships):

• PENNSYLVANIA: 10 Democrat-leaning districts, 9 Republican-leaning districts, with a per-district PVI average of 3D. That's a lot of extra Democratic voting percentages for a near-evenly split delegation. Arizona, Tennessee, Virginia and Missouri, with similar partisan splits (in those cases leaning GOP), have delegations leaning more than two-thirds GOP. By that standard, a 13-6 Dem-GOP split of the Pennsylvania delegation seems reasonable.

• OHIO: 6 Democrat, 12 Republican, with PVI average showing an even split. And the state seems to have shifted significantly more Dem since the last two presidential elections (which you'll recall is what the PVI is based on), so it's probably more pro-Dem now. A reasonable split would be 9 Dem, 8 GOP.

• FLORIDA: 8 Democrat, 17 Republican, with PVI average of 2D across the state. The split should be at least 13 Dem, 12 GOP. This will be tough given continuing GOP control of the governorship and legislature.

• MICHIGAN: 5 Democrat, 10 Republican, with PVI average of 2D across the state. The split should be at least 9 Dem, 6 GOP.

• ILLINOIS: 8 Democrat, 10 Republican, with PVI average of 3D across the state. The split should be at least 12 Dem, 6 GOP.

It seems unlikely that much can be done about Florida (yeah, things can change, I suppose), but the others are possibilities. Together that adds up to 14 seats alone that would be easier for Dems to either pick up or defend.


SO . . . THERE'S LOTS THAT CAN BE DONE--

and must be done, to overcome the GOP districting advantage.

Some of these strategies fall under the heading of House Campaign 101, and are well-known to all election strategists. But even with these, it helps to see the underlying numbers, in order to understand why these strategies work and also make it more likely that they will be taken seriously and properly implemented and not just paid lip service.

Other strategies are less-known and/or accepted, notably the potential rewards of the 50-state strategy. They deserve to be embraced and emphasized by the Democratic establishment as a whole.