By the time you read this, the “yes” side will have likely raised more than $50 million to persuade Virginians to vote in favor of a proposed constitutional amendment to allow a rare mid-decade redrawing of congressional lines.
The “no” side lags far behind, at about half that. We’ll know more once a new round of campaign finance reports become available today.
To put this in context, that $50 million or so is more than all but one of the statewide candidates spent in last fall’s election; only Abigail Spanberger raised more (just a little north of $70 million). It’s more than three times as much as Jay Jones raised in his successful campaign for attorney general and 5.5 times more than what Ghazala Hashmi raised in her successful campaign for lieutenant governor.
Another difference is that we know exactly where all that campaign money last fall came from. Much of these funds come in the form of “dark money,” where it’s hard to trace the source to know who’s digging into their pockets to influence our votes. For more on that, see today’s story by Cardinal’s Richmond-based reporter, Elizabeth Beyer.
Instead, I’ll tackle a different angle: What if, instead of trying to change Virginia’s congressional map, the “yes” side simply spent its money on campaigns in the existing districts? This skips over a lot of technicalities involving federal campaign finance laws, so it’s mostly a thought experiment. However, here’s one likely answer: New polling suggests that Democrats could win three seats and possibly four without having to resort to redistricting. This is significant because the proposed map is intended to knock out four of the state’s five Republican House members. These latest figures suggest Democrats might be able to achieve the same result the old-fashioned way: by winning elections in the current districts.
This math comes from G. Elliot Morris, one of the nation’s premier crunchers of political numbers. He’s the author of “Strength in Numbers” and now runs the Strength in Numbers newsletter, which is fascinating. If you really want to know how he arrived at these figures, you can see the full explanation on his site, but be warned, here’s a taste: “We use Multilevel Regression and Post-stratification (MRP) to estimate Trump’s approval rating for small geographic areas and demographic subgroups. MRP is a statistical technique that combines survey responses with Census data to produce reliable estimates even for small populations. The model is a Bayesian multilevel logistic regression programmed in Stan, where approval is coded as 1 (approve) or 0 (disapprove). ‘Don’t Know’ respondents are excluded from the analysis.”
For those of us who still count using our fingers (and sometimes toes), here’s what to know: Morris used this math to estimate President Donald Trump’s approval rating in every congressional district in the country. The keyword there is “estimate.” He started with a national poll, and then sorted the numbers based on the demographics of each district, on the theory that if people of a certain demographic profile nationally feel a certain way, they probably feel the same way at the congressional district level, too. Example: If older rural white voters like Trump, then he probably has high approval ratings in Virginia’s 9th District, which has a lot of older rural white voters. Indeed, this math projects that in the 9th District in Southwest Virginia (now represented by Republican Morgan Griffith), Trump’s approval rating is about 54.0%.
Polling nationally shows that Trump’s approval ratings are down, and 54% approval in the 9th seems soft for Trump (or any Republican president), but it’s still on the plus side of the equation.
Where things get more interesting is with the other four congressional districts now represented by Republicans. Morris’ math shows that Trump is under 50% in 134 Republican-held House seats across the country. This includes all four districts in Virginia where Democrats are trying to squeeze out the Republican incumbents through redistricting. Moreover, in all four, Trump’s disapproval rate is higher than his approval rate. That’s a warning sign for Republican incumbents because if voters in a Republican-voting district are unhappy with Trump, they may transfer that unhappiness to the nearest Republican on the ballot this fall. You can bet that all the Democratic candidates will spend a lot of time running against Trump, while the Republican incumbents will be focusing on something else, anything else.
If voters approve redistricting on April 21, none of this really matters since the purpose of the new map is to produce a 10-1 Democratic outcome rather than the current 6-5 Democratic majority. However, for our calculations today, let’s assume the “no” side prevails, and we stick with the current congressional lines. If that happens, what’s the political lay of the land?
Morris’ math shows Trump in bad shape in all six Democratic districts (no surprise there). In the 8th District, represented by Don Beyer of Alexandria, Trump’s approval rating is projected at just 20.8%, which means his disapproval rate is 79.1%. (Morris’ math doesn’t allow for undecideds; it’s an either/or proposition). There are very few districts nationally where Trump fares worse, although Trump scoring poorly in Northern Virginia shouldn’t surprise anyone. The most competitive district in Virginia now held by a Democrat is the 7th District, represented by Eugene Vindman of Prince William County. This math shows Trump’s approval rating there at 36.2%.
Now let’s look at the Republican districts. We’ve already seen that Trump is likely doing well in Southwest Virginia. In the other districts, not so much:
The 2nd District on the south side of Hampton Roads has long been a swing district. Republican incumbent Jen Kiggans faces a tough reelection battle no matter who the Democratic nominee is, although it’s likely to be former Rep. Elaine Luria. Here’s the figure that should worry Kiggans: Trump’s approval rating in the 2nd is projected at 37.3%. To win, she’ll have to separate herself from Trump to make the case that she’s done a good job, even if voters think Trump hasn’t. The Democratic nominee will no doubt tie Kiggans to Trump at every opportunity. That’s how politics works.
The 1st District in eastern Virginia is now held by Republican Rob Wittman, but Democrats have officially targeted it as potentially winnable, likely with Henrico County Commonwealth’s Attorney Shannon Taylor. Here’s why: Trump’s approval rating in the 1st is 39.5%, according to Morris’ math.
The 5th District in Southside Virginia (along with Charlottesville and Albemarle) is a tougher district for Democrats. I’ve gone through the math in the past that makes it hard for me to believe a Democrat could dislodge Republican incumbent John McGuire. However, Morris’ math shows that Trump’s approval rating in the 5th is just 41.8%, with a disapproval rate at 58.2%, which certainly creates an opening for a strong Democratic candidate — potentially former U.S. Rep. Tom Perriello, who has experience in running and winning (and also losing) in rural parts of Southside. Even before Morris’ numbers, the Democratic Congressional Campaign Committee had listed the 5th as a target district; this math suggests they’re on target.
Then there’s the 6th District. As currently constructed, it’s the second-most Republican district in the state (behind only the 9th) and is represented by Ben Cline. Morris’ math shows Trump at 48.3% approval, 51.7% disapproval. That’s close enough to fall within the margin of error, and also close enough that even if the numbers are accurate, they could easily change, and also close enough that a popular incumbent could overcome a president being “under water,” as the polls say. Still, this raises the prospect of whether Democrats could pull an upset in the 6th if they nominated a strong and well-funded candidate — of which they have one in the form of Roanoke author Beth Macy. The difficulty for Macy is that if redistricting passes, she’d be paired in a district with Perriello. It would be a district more favorable to Democrats, but the only public polling so far shows Perriello with a commanding lead, which makes sense given his name ID. The irony is that if Roanoke Democrats vote “yes” on redistricting, they’re essentially voting to doom their hometown candidate to an early exit. If they vote “no,” she’d almost certainly win the nomination in a district that would be a long shot but which this data shows might be within reach in the event of a blue wave of historic proportions.
Got questions about redistricting?
I’ll have a column later this week where I attempt to answer reader questions. Submit your questions here.
A case can be made — using these numbers — that Democrats don’t need redistricting because they might be able to win four seats anyway. Or, at worst, they could win three, which means redistricting isn’t about picking up four seats, it’s really just about picking up one. Of course, from a Democratic point of view, there are still reasons to back redistricting: It would send a signal nationally and make Democratic wins more certain, and more permanent. These numbers, even if solid now, might not be so solid come November, and voters’ moods may swing back to a more favorable Republican view in the years ahead. In other words, Democrats may not really win all these seats, and some they do might flip back Republican in 2028. That’s why a “yes” vote gives Democrats more certainty that they can lock out Republicans until the 2032 elections, when we’d get a new map after the next census.
Let’s go back to the original question: What if Democrats took the $50 million they’re spending on redistricting and used it for these congressional races? Let’s skip the 9th District — Democrats statewide concede that one, even if the local Democratic candidates don’t. That means for our figuring purposes, we could divide $50 million four ways — that’s $12.5 million for each candidate.
Two years ago, the Virginia House candidate with the most money was Vindman, who spent $17.9 million to win the 7th District. He was by far the outlier; the second most-funded candidate was Kiggans, with $6.4 million. Most candidates, though, were in the $2 million-or-less range.
In the 1st District, Wittman spent just over $2 million while his Democratic opponent came in at $722,445.
In the 5th, McGuire spent $1.8 million while his Democratic opponent had $353,529.
In the 6th, Cline spent less than $1 million — $969,485 — while his Democratic opponent spent $188,758.
We’ll get updated campaign finance numbers later this week, but Macy has already said she’s raised $1.1 million, which would be more than Cline spent in his whole campaign two years ago. Perriello is at $1.4 million, which is almost at McGuire’s campaign-long total last time.
Give each of these candidates $12.5 million, and it would dwarf what any of their predecessors had to spend. Money doesn’t always produce desired election outcomes, but it sure helps. Democrats could have funded these candidates in eye-popping ways. Instead, they’re choosing to spend that money on redrawing the maps. The Morris math raises the question of whether they’re spending all that money unnecessarily when they might be able to win these seats anyway.
Want more politics and analysis?
I talk about redistricting in our latest Cardinal News podcast; find it here or wherever you get your podcasts.
I’ll give a final update on the early voting trends (and whatever other political news happens) in West of the Capital, our weekly political newsletter that goes out on Friday afternoons. Sign up here:
