Donors' real problem: Shiny object syndrome

And, how to make good on past mistakes in Wisconsin in 2023

In today’s New York Times is a story on how partisan polls gave a false impression about the November election. Today I focus on the subheadline, “Politicians and Donors Spooked by Surveys” - and how to escape this trap in a race happening soon, in early 2023.

First off, a correction of the Gray Lady. Partisan polls may have been off, but the aggregates were fine. The Princeton Election Consortium probabilistic calculation showed 47-51 Democratic Senate seats, and the eventual outcome was 51 seats, within range. PEC showed a toss-up for control, a “meta-margin” of R+0.4%, basically no advantage; on election day the tipping-point state for Democrats was Nevada, D+0.8%, a difference of 1.2 points - minimal. And in the House, the generic Congressional average suggested a national Republican win of 2 points in contested races - which is exactly what happened.

However, I have a bigger concern: bad polls weren’t the biggest driver of suboptimal donor decisions.

Donors and shiny objects

Even if we suppose that partisan polls caused some donors to shift their giving, there are far larger factors that influenced donor behavior. These factors add up to hundreds of millions of dollars, and an enormous misdirection of resources. The fact is that donors followed quite different cues.

Of the five most expensive Senate races in 2022, three were not that close: Arizona (5 points), Pennsylvania (5 points), and Florida (16 points). Five points is possibly excusable, though Pennyslvania’s and Florida’s large populations make them additionally ineffective on a per-voter (and therefore per-dollar) basis. That money would have been better spent in Wisconsin or Nevada.

House races were worse: the top-funded five races were decided by 39 points on average, and included Kevin McCarthy, Steve Scalise, Nancy Pelosi, and Adam Schiff. Now, some of those are party leaders. But other top fundraisers include Majorie Taylor Greene, not a leader - and her opponent, who is not even in Congress.

Overall, a simple explanation is that donors are not moved by data, but by news coverage, star power - and their emotional reactions to those stories and stars. Call it the charismatic megafauna hypothesis of donor behavior.

Data+feelings > feelings

Now that I have shown you the dangers of paying too much attention to charismatic megafauna, here is how to find the lowly races that make a bigger difference. The secret is…data!

With data, it was possible in 2022 to do much better. On our optimized-giving ActBlue and WinRed pages, the top six races by donation, accounting for half of the giving, were decided by a median of 2.2 points. (This calculation was done using ActBlue data. WinRed donations were not itemized by recipient.) Across 18 races, the overall median margin was 4.2 points, with 10 wins for Democrats and 8 wins for Republicans. Evidently, it is possible to target resources in an optimized and unbiased manner.

A short-term focus does come with one burden: you have to take a side on individual races. Let me explain.

Suppose that your goal is to make democracy more responsive to voters. This is the goal here at Fixing Bugs In Democracy. “More responsive” means helping build conditions where outcomes are sensitive to voter sentiment. There are multiple ways to get there: nonpartisan redistricting, ranked-choice voting, and voting-rights lawsuits. In the short term, these paths sometimes favor Democrats: for example, redistricting reform in Michigan. They can also favor Republicans: for example, redistricting lawsuits in Maryland and New York. But in the long term, such approaches are conceptually neutral. They help voters, not parties or politicians.

In the short term, candidates for state office can also affect the responsiveness of elections. But here, an asymmetry can arise. As I’ve written, secretary of state and judicial races are especially important. Key secretary-of-state races went in a pro-democracy direction in Arizona (Fontes-D), Nevada (Aguilar-D), and Georgia (Raffensperger-R), making the 2024 Presidential election likely to be conducted accurately in those swing states. However, key judicial races went in an anti-voting-rights direction in Ohio and North Carolina.

In 2023, a key judicial race in Wisconsin

The next chance to optimize in the short term comes in Wisconsin. Key elections in February and April 2023 deserve attention.

The Wisconsin Supreme Court is a classic case of the shiny-object problem: a decade’s worth of representation and other rights could go in either direction, yet the 2023 election will get a fraction of the attention that went to last month’s Senate or governor’s race. Indeed, a similar close state Supreme Court race in 2019 flew under the radar and elected Justice Brian Hagedorn, the deciding vote in this year’s big redistricting cases.

Wisconsin’s legislative map is among the nation’s most partisan. It allows Republicans to control the General Assembly with less than 45% of the statewide vote. Alternative maps exist that do not do this, but instead give both parties a chance to gain control. In a lawsuit, the Wisconsin Supreme Court split 3-3 along liberal and conservative wings. The seventh swing justice, Hagedorn, allowed the map to stand. That same court also hears other voting-rights issues such as the use of drop boxes, as well as future abortion-rights cases.

Now one of the three conservative justices, Patience Roggensack, is retiring. To replace her, a two-stage election will be held: a primary in February, and a runoff in April. Depending on who wins, the pivot point of the court may change.

The primary has four candidates, two liberals and two conservatives. The liberals are Everett Mitchell and Janet Protasiewicz. One conservative candidate, Daniel Kelly, helped defend the 2010 partisan gerrymander. You can read about all four candidates here.

The February all-candidate primary will advance the top two vote-getters to the April general election Such a top-two system carries the risk of advancing two candidates from the same side. Until some happy future in which Wisconsin adopts a ranked-choice system, turnout in February will be crucial. And, of course, the April election will matter enormously.

Microfauna donations in 2023

Democrats, to support Wisconsin candidates Mitchell & Protasiewicz, go to this link:

Republicans may contribute to a candidate to fill a Pennsylvania state supreme court vacancy, Carolyn Carluccio. That sitting judges are split 4-2 favoring Democrats, so a win by Carluccio would introduce closer balance. Her position statements are here, and once it goes live, her website will be here.

Until next year…

With that, I thank you for your readership in 2022. It’s been quite a ride at the Electoral Innovation Lab, with lots of successes. I’ll give a recap in the new year.

Have a happy and safe New Year’s celebration, and see you in 2023!