Playing both sides: the downside of dual candidate endorsements

If you are an organization that attempts to influence politics in your favor, you’re almost certainly making candidate endorsements.

(And if you aren’t, you probably should be.)

Occasionally, I have noticed organizations that "dual endorse" candidates. Endorsing multiple candidates is acceptable, as long as there are an equal number of vacancies available. For instance, it is perfectly acceptable and sensible to endorse two candidates for two open seats.

However, I disagree with the idea of an organization advising voters or donors to support either of two candidates for a specific seat by saying "both of them are good for us." In my opinion, it is not typically a sound decision.

The first reason is more of a math reason.

Let’s pretend your organization represents the workers of the Willy Wonka factory. And a large segment of voters really sympathize with the workers of the Willy Wonka factory, and will generally look to your organization to tell them which candidates are the most pro-Willy Wonka factory workers.

Let’s also pretend there are three candidates running for one seat on the local government board that sets policies that can positively or negatively impact Willy Wonka factory workers.

Two of those candidates are varying degrees of good for your organization. One is terrible (they hate candy).

Some organizations may look at this scenario and say “well, let’s just endorse both the good candidates, since they are both good for us and we should likely win either way!”

Under this scenario, let’s assume 60% of voters love Willy Wonka factory workers, and 40% are generally apathetic or even unsympathetic to them.

So the organization endorses both of the “good candidates”, and tells voters, “Hey, voting for either of these guys is totally fine with us!”

And so half of the pro-Willy Wonka factory worker voting bloc supports Good Candidate A, and the other half supports Good Candidate B.

Final Results

Good Candidate A: 30%
Good Candidate B: 30%
Bad Candidate: 40%

Oops.

(This doesn’t just apply to votes, either. It can apply to donors – in this case, if Willy Wonka factory workers tend to donate based on the instruction of the organization, their money, or the money of other sympathetic donors, could be split between the two Good Candidates.)

Of course, the math rarely works out this perfectly. But you can see how some organizations – namely those who have a subset of voters who will vote on their issues – can dilute their own coalition by dual endorsing. They’ve basically done the work of “dividing and conquering” on behalf of their opposition (the Bad Candidate).

Even if the figures aren’t as clean and pronounced as they are in this example, a dual endorsement can still have a similar impact around the margins – and in a close race, the margins could be what make the difference.

Math isn’t the only argument though. There’s a less tangible reason, and it has to do with the message you are sending about your organization.

If you are an organization whose endorsement is in demand, you should use that to your advantage, and force the cream of candidates to rise to the top.

You should be endorsing the candidate who has the best combination of being a) good on your issues and b) having a reasonable chance of victory.

There will be plenty of candidates who are good on your issues but have no chance of winning, and there will be plenty of candidates who have a great chance of winning who are terrible on your issues.

You want to incentivize candidates to be good on your issues. And you want to incentivize candidates who put themselves in a good position to win (through hard work, acumen, and fundraising).

And so you want to endorse the candidate who maximizes both categories. And by doing so, you are sending a message to future candidates (and incumbent elected officials who want to be re-elected) that they need to be good on your issues, and that they need to do the work that puts them in the best position to win.

Will there be times where you have two candidates who are great on your issues and each have a decent shot of winning? Absolutely. And sometimes when this happens, the organization will want to bail itself out from making a tough choice by supporting both. It feels like a win-win! They get to make a choice without having to actually make a choice, and they don’t have to deliver bad news to either candidate.

But a dual endorsement doesn’t always mean you are delivering “good news” to both candidates. Remember, people running for political office are in a particularly competitive state of mind, especially when trying to wrangle a handful of scarce resources (votes, money, endorsements). Knowing that the organization supported their opponent often feels worse than the satisfaction of knowing that they supported the candidate. There’s the chance – depending on the candidate – that the dual endorsement has generated more resentment than goodwill toward your organization.

But sometimes the organization’s leadership can’t come to consensus around one candidate, for any number of reasons. And that’s OK. Not endorsing either candidate is a viable option. Is it ideal? No. But sometimes it’s unavoidable.

I would rank the hierarchy of ideal outcomes in the following manner:

1. Endorsing one candidate (High risk, high reward)
2. Not endorsing any candidate (Medium risk, medium/low reward)
3. Dual endorsing (High risk, low reward)

#1 is ideal, but not always possible. I would add that when the alternative to the two Good Candidates is a Very Bad Candidate, the most calculating move would be to endorse one candidate and try to ensure they receive as much support from voters and donors sympathetic or aligned with your organization, in order to deprive the other Good Candidate of political oxygen and preventing them from splitting the vote (and letting the Very Bad Candidate slip through). This will of course make the other Good Candidate sad and possibly angry. But if the risk of the Very Bad Candidate is significant (they want to outlaw candy), you’ll have to make some tough decisions.

#2 is almost always better than #3.

Are there instances where #3 might make the most sense? Almost certainly. But those instances are the rare exception, at least in my experience.