Incentives are Important: Why our DUI prevention incentives don’t work but our mishap prevention incentives do

On the surface, incentives seem really simple. You want people to behave a certain way so you establish a reward for that behavior. But incentives can be tricky, especially if poorly implemented. A well-known example of misaligned incentives is real-estate sales commissions. A home seller incentivizes a real-estate broker by giving them 3% of the home sale price. The broker is therefore incentivized to maximize the sale price. If you’re trying to sell your house for $210K, and the broker gets a buyer willing to pay $200K, the broker is already getting $6K in commission, and will only going to get another $300 if they get up to $210K. That extra 10K might be really important to you the seller, but that $300 doesn’t mean jack to the broker. The real incentive for the broker isn’t to maximize sale price, it’s to sell as many homes as quickly as possible.

The Navy has limited positive incentives like cash bonuses for high performance. What the Navy does have is a lot of negative incentives, like punishment. This idea is that people want to avoid punishment, so they don’t do the behavior that gets them that. A note on negative incentives: using too many of them creates a culture of risk-aversion (not in scope for this article, but it’s definitely a problem).

For both positive and negative incentives to work, they must be carefully constructed so that you actually get the behavior you want. I’ll use two examples – one we do right and one we do wrong.

1)     Airplane Safety: We all know that we have a very safe platform because when things go wrong, everyone does a deep dive, learns from what happened, and then those lessons are disseminated and incorporated into our training curriculum to make us safer. This requires people to divulge mistakes, safety incidents, etc. So our incentive is to reward squadrons for the number of HAZREPs they produce. Further, each squadron and wing publicly rewards individuals for doing safety related acts of kindness. Great alignment to get the behavior you want.

2)     DUI and the friend: We all know that if you’re out with a buddy, and he offers to drive but gets pulled over and arrested for a DUI, not only does the DUI offender get in heaps of trouble, but everyone that “allowed” him to drive drunk gets punished too. The idea is that you want people to take care of each other, so you will punish when they do not. There are two choices If I have to decide what to do in this situation: 1) take care of a sloppy drunk (really hard) or 2) steer clear of them (really easy). The result (avoiding punishment) is the same if I do either one successfully. Because the incentive is the same I’d much rather do the easier of the two options and just steer clear. Further, even if I work really hard to take care of people, there’s no guarantee I’d be successful. This is a perverse disincentive combined with a risky situation where it’s more likely to cause people to do the opposite of the behavior you want. How many people have seen the senior most person just disappear when things were getting fun? How many of you have been the senior person that recused themselves from going out in the first place?

Incentives should also not be universally applied. Consider the Skipper that punishes the JO pilots when they make a mistake. Didn’t like how they handled that emergency? Pull their papers. This negative incentive is directly opposed to the airplane safety incentive listed above. You run the risk of causing people to simply not report safety incidents for fear of punishment. It makes you less safe, not more. This happens often enough that I’m convinced we don’t understand how to effectively use incentives.

Group incentives are also a useful tool – there’s a social component to it that we respond to. Remember when the drill instructor made everyone else do pushups except for the ass that showed up late? Everyone pays a heavy price except for one person, who shoulders all the blame. That person probably never shows up late again. This works great for negative incentives, but is pretty weak with positive incentives. You remember the 365 days without a DUI and everyone gets a day off? The incentive rests on the behavior of 300+ other people… am I really going to respond to this? Not really.

The interesting thing is that preventing DUIs and increasing aviation safety should be in the same category of incentives. Use negative incentives only for the things you really want to punish – safety incident cover ups and actual DUIs should be punished severely and publicly (add the social element). Punish the right people, not the bystanders. Punishments should also be public enough that everyone understands exactly the reason for it. If you’re severely punishing someone for covering up a minor safety incident, it could be misconstrued as a big overreaction. But also incorporate rewards – reward the individual who stops a DUI with a day off (and don’t punish anyone for reporting it).

We tend to rely heavily on intrinsic (or innate) motivations rather than incentives. Here are some common rebuttals to incentive schemes that I’d expect to hear from typical VP. “Stopping DUIs is the mission of every good leader. Work hard because you’re patriotic and doing it for your country. A prestigious but really hard job is your reward for high performance.” Innate motivation is great, and we in the military have it in spades compared to the corporate world. It’s a cultural attribute unique to the military and should be celebrated. But intrinsic rewards can’t be universally applied nor are they universally effective at motivating everyone. Not everyone has Captain America level patriotism. Not everyone cares about prestigious roles. Not everyone is a leader that cares deeply about the people around them. Adding a robust incentive system applied in the right way could build on that innate motivation. It would not only fix a lot of problems but would also maximize the productivity of our best asset: our people.

A couple of key notes about incentives:

1)     Incentives need to be aligned to the behavior you want to achieve. Be honest about what you’re really incentivizing (harsh punishments for safety incidents is more likely to cause cover-ups than improve safety). “On the folly of rewarding A while hoping for B” is a classic management article that talks about this in several industries, including the military.

2)     Group incentives add a social component which is powerful, but the benefit is essentially split between all the participants.

3)     Incentives have a time component to it: the longer it takes to get the reward, the less sensitive people are to it (people just don’t internalize rewards that are too far out).

4)     Too much reliance on negative incentives creates a culture of risk-aversion.

5)     Any meaningful change to VP culture should include changing our incentives for everyone from the Skipper down.

6)    There is no reason not to experiment with different incentives and share the results amongst community leaders. I’d insert this into the DH and CO/XO curriculum at VP-30.

  1. #1 by anymouse1 on 2March2017 - 1:28 pm

    Any innovative ideas to prevent DUIs? I like the programs that offer safe rides home, but how to get the word out that these options truly are “free” without reprisal? Requiring DoD ID numbers leads to mis-trust of the system. Having an Uber style (technology backed) option to get both you and your party AND your vehicle home without cost would help, but also would be costly. I’ve heard of a grass roots effort to offer such a service, with volunteer hours credited to those willing to volunteer their time to get fellow sailors home safely, but haven’t seen much action at the Big Navy level on such an innovative idea. Feels like we pay lip service to these kinds of programs.

    • #2 by Voltaire on 13March2017 - 7:13 am

      Why is there fear of reprisal for needing to use a DUI-prevention service? Our negative incentive culture has been so overused that we can’t even implement things like this because people are so risk averse they’d rather take the chance on driving hoping that they won’t get caught. Simple fix is for the Admirals on down to the PO1s need to clearly define what gets punished and what doesn’t.

      Eliminating DUIs is the wrong goal. Minimizing them should be the objective. Changing the drinking culture is probably the only way to eliminate DUIs (my guess is that DUI rates are non-existent in devout Mormon communities). I think one reasonable thing you can do is adjust the incentives as discussed in the article (add rewards and eliminate punishment for bystanders).

      Another would be to change the negative incentives for leadership. I think there’s been a DUI every year in every squadron I’ve been in. That means there are a lot of people drinking and driving, because most don’t get caught. It’s wrong to punish a Skipper for having 3 DUIs in one year because maybe that’s just how the dice rolled in that squadron. Because the Skippers are hypersensitive, they’re implementing all these heavy handed policies that, at best, aren’t helpful, and at worst, cause more problems than they solve. This is so that when the inevitable happens, they can clearly point to this policy and say they’re not culpable (again, back to the perverse behavior that negative incentives can cause.

      Side note – Why have an Uber-style app when there’s actual Uber? The Navy should simply partner with Uber for a product that active duty can use when they’re in a bind and need a ride. It’d be way cheaper than the Navy inventing it’s own thing, utilize an already existing network of drivers, and help Uber manage some of it’s own PR image problems.

  2. #3 by Terry Rothchild on 2March2017 - 9:36 pm

    My last Skipper knew that morale was in the shutter, so he came up with a game plan to skyrocket morale into the Kuiper Belt.

    His ingenious plan? A t-shirt design contest.

    He figured people would be chomping at the bit to design our new shirt, and with its 100% cotton blend it was to be ten times more comfortable than the last shirt.

    Well, guess what? Nobody fucking cared. It ends up designing logos is something people with high morale and free time end up doing (both of which had been running on E for quite some time). Seeing the writing on the wall, I did what any good Sailor would do: Stall.

    I put that contest hard on the back burner, sucked and weaved the Skipper’s inquiries, and dumped that shit on my relief (who deserves a NAM for how long they also put that off). Anyway, the morale:

    Incentives don’t have to be super amazing or expensive, but they need to be something people actually want. If I loved P3s and my Squadron, I would have been ecstatic to get my design on every shirt or command coin in the hangar. If my crew did something amazing, I’d love to be Crew of the Quarter. If we could sit down and work out what’s wrong with our crews, what’s wrong with our Squadron, what’s wrong with our community, we could be the finest community in the Navy.

    But we can’t. And we won’t. And I fucking hate P3s.

  1. A Retrospective on the P3Reform Project | p3reform
  2. Final Thoughts | p3reform

Leave a Reply to Voltaire Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: