If this is your first time reading, please check out the overview for Poor Ash’s Almanack, a free, vertically-integrated resource including a latticework of mental models, reviews/notes/analysis on books, guided learning journeys, and more.
Executive Summary: Incentives Mental Model
If you only have three minutes, this introductory section will get you up to speed on the incentives mental model.
The concept in one quote:It is difficult to get a man to understand something when his salary depends upon his not understanding it. - Upton Sinclair Click To Tweet
Key takeaways/applications: it is extremely difficult to discourage a behavior if it is incentivized, and equally difficult to encourage a behavior if it’s disincentivized.
Three brief examples of incentives:
Greedy agent… or dumb principal? A classical challenge in the business literature is “principal-agent problems” – the fact that the incentives of management (maximize their own pay) and that of shareholders (maximize their own earnings) are not always aligned. In “ Misbehaving” ( M review + notes), Richard Thaler offers a unique behavioral twist on this – the “dumb principal” challenge – that’s discussed more in the local vs. global optimization model.
More frequent bad things… are a good thing? The legal system is inherently based on disincentives: do illegal things and you’ll be fined or thrown in jail. Unfortunately, the addictive pull of drugs and alcohol can provide a powerful counterincentive to keep breaking the law. In “ The Up Side of Down” ( UpD review + notes), Megan McArdle explores, among other topics, a unique Hawaiian parole system that combines strong incentives and consistent feedback to yield better results for everyone involved.
If this sounds interesting/applicable in your life, keep reading for unexpected applications and a deeper understanding of how this interacts with other mental models in the latticework.
However, if this doesn’t sound like something you need to learn right now, no worries! There’s plenty of other content on Poor Ash’s Almanack that might suit your needs. Instead, consider checking out our discussion of the selective perception, product vs. packaging, and nonlinearity mental models, or our reviews of great books like “ The Autobiography of Benjamin Franklin” ( ABF review + notes), “ Polio: An American Story” ( PAAS review + notes), or “ The Pleasure of Finding Things Out” ( PFTO review + notes).
A Deeper Look At The Incentives Mental Model:
– Charlie Munger
I’ve had similar experiences. Here’s an amusing anecdote: in the summer of 2017, as I was working on creating PAA, I asked readers of my blog for book recommendations, offering a $25 Amazon gift certificate and public recognition to whoever’s recommendation ended up being the best.
Even if people assumed only a modest number of responses, the expected value of sending a recommendation hardly amounted to anything – maybe a few bucks – certainly not enough to make a difference for the financial professionals who at that time comprised the majority of my readership, right?
Well, I ended up with so many recommendations that I still (as of summer 2018) haven’t gotten through all the books… and as you know, I read a ton. Lesson learned: incentives make a difference.
The model at its most basic is intuitive and pops up in tons of books ranging from “ The 7 Habits of Highly Effective People” ( 7H review + notes) to “ How Doctors Think“ ( HDT review + notes). One of my favorite examples comes from Meredith Wadman’s “ The Vaccine Race” ( TVR review + notes).
It is now known that any strain of human cells that can divide indefinitely is cancerous, but for much of the early to mid 1900s, the prevailing wisdom was the opposite: that any healthy human cells would divide indefinitely.
In the 1930s, one smart lab technician picked up on some of the problems in the experimental methodology and pointed it out. But it was the Great Depression, and she was told, as Wadman relays:
“to forget what she was seeing or risk losing her job.”
So, naturally, the myth persisted for decades, slowing scientific progress in some key areas.
As usual, to provide fresh/new insights, let’s skip right to the interactions. (Do note that there are more scattered around the site: there’s a nice one in the humans vs. econs mental model… and I’ll let you find the rest!)
Incentives x Marginal Utility x Social Connection
Bob Porter: Space out?
Peter Gibbons: Yeah, I just stare at my desk. But it looks like I’m working […] I’d say in a given week, I probably only do about fifteen minutes of real, actual work […] the thing is, Bob, it’s not that I’m lazy, it’s that I just don’t care.
Bob Porter: Don’t… don’t care?
Peter Gibbons: It’s a problem of motivation, alright? Now if I work my ass off and Initech ships a few extra units, I don’t see another dime, so where’s the motivation? […] my only real motivation is not to be hassled. That, and the fear of losing my job. But you know, Bob, that will only make someone work just hard enough not to get fired.
Office Space does a better job with incentives than pretty much any book around. The exchange above epitomizes the challenge faced by the employers of many of my friends who went into fields other than finance or law: absent opportunities for rapid career advancement or raises, many corporate jobs with a high fixed salary and a low bonus disincentivize extra effort because the marginal utility to doing more than the bare minimum is exceedingly low.
Companies that get incentives right tend to get nonlinearly more production from their employees. My favorite example is The Container Store, which recognizes this explicitly: one of their bedrock “Foundation Principles” is that “1 Great Person = 3 Good People.” They pay up accordingly: wages for front-line employees are 50% higher than the industry average, and raises are directly linked to performance, providing a strong marginal incentive for marginal effort.
Of course, those of you who’ve read the humans vs. econs model may recall that I noted there that one consequence of humans not being unfeeling robot automatons is that we don’t respond only to financial incentives. In fact, there are certain situations where financial incentives can, perversely, decrease the likelihood of people doing something.
What are those non-financial incentives? Status, recognition, and so on. Social proof serves as a powerful incentive – a phenomenon explored by Sunstein/Thaler in “ Nudge” ( NDGE review + notes) and Cialdini in “Influence” – because social connection matters to real humans.
In an environment where everyone’s doing the bare minimum (like Office Space’s “Initech” – emblematic of many corporate environments), there’s a strong pressure to go along with it.
Conversely, it’s totally possible to intentionally construct a work environment where reciprocity and social proof serves as a positive rather than negative incentive for giving it your all: Kip Tindell’s “ Uncontainable” ( UCT review + notes) explores The Container Store’s unique approach to business, including creating an “Air of Excitement” throughout their operations:
“anyone walking through our doors can’t help but be affected by the passion, energy, creativity, and, yes, love that our employees demonstrate every day on the job.”
This not only incentivizes employees to do their best, but actually incentivizes customers, too: Tindell notes that customers frequently explain they shop at The Container Store because they like being around the happy employees.
As Tindell discusses extensively, the Container Store recognizes that motivation is multicausal and tries to address all the relevant categories, not only by paying top dollar and creating a great work environment, but also with frequent positive recognition and agency (allowing employees to use their judgment rather than being micromanaged). The Container Store didn’t invent this, of course – Dale Carnegie discusses similar ideas in “ How To Win Friends and Influence People” ( HWFIP review + notes) – but they’ve certainly applied it to the max.
What’s the result of this Lollapalooza of correct incentives? The Container Store is consistently ranked as a top place to work by rigorous external reviews, and that’s translated into business success: TCS managed to thrive in a category so difficult that well-capitalized, well-managed competitors couldn’t make it.
Recent results have been pressured by the generally tough retail environment and a debt overhang from a vastly overpriced private equity buyout more than a decade ago, but that doesn’t take away from the company’s long-term accomplishments.
Application/impact: an approach to incentives that considers marginal utility, as well as non-financial incentives, will perform more effectively than one that does not. This works equally well for companies or individuals; the whole business model of “Weight Watchers” is providing individuals with social connection and social proof as a motivating factor to diet and exercise!
Incentives x N-Order Impacts
There’s a famous, maybe-apocryphal story in which some college had a rodent problem. The administration came up with the bright idea of paying students a small amount for each carcass they turned in.
That’s an example of n-order impacts, or, less formally, “what-then” problems. Part of the challenge in structuring incentive systems is that in any complex adaptive system, you have to take into account not just the immediate impacts, but the longer-term impacts as the new incentives filter through the system.
There are a lot of “standard” examples for this: healthcare policies that make it easier for people to access healthcare usually dramatically underestimate the accompanying surge in volume. Bankers with annual bonuses and no clawbacks may not care about the long-term risk of bad loans, just like a mayor with a reelection coming up might not care if the city can pay back a bond in 30 years, when everyone will like the new schools and playgrounds today!
The first involves bankruptcy law: as McArdle explores in depth (pages 248 – 253), American bankruptcy law is extremely lenient compared to European bankruptcy law, to the extent that new, stricter bankruptcy laws that were passed (in the U.S.) were perceived by Europeans as being ridiculously lax.
Many readers’ first impression – like mine – might be: hey, lax bankruptcy laws are bad, we don’t want to incentivize people to spend beyond their means! But McArdle points out an often-overlooked n-order impact: entrepreneurship is a much bigger risk in Europe than in the U.S.
Stateside, the worst that can happen to you if you start a business is that you lose everything and go find a day job. In Europe, it turns out that you can lose more than everything, because if your business venture goes south – regardless of whether it’s luck orskill – you’re still on the hook for any debts you may have, pretty much forever.
I don’t know enough about the underlying data here to come to any firm conclusions on what appropriate bankruptcy laws should look like – an example of low- ideology, humble probabilistic, scientific thinking, by the way – but it’s at the very least an interesting, thought-provoking starting point for thinking about unintended consequences.
McArdle, similarly, presents some data suggesting that simply putting homeless people in apartments – which, initially, seems like an expensive solution that incentivizes homelessness – ended up being cheaper in some situations, because it reduced the n-order impact of excessive emergency room visits.
Application/impact: incentives can have unintended consequences as people respond to them. Be cautious and try to think through the second and third-order impacts of changing an incentive!
Incentives x Salience/ Vividness x Contrast Bias x Loss Aversion
You’ve got places to be, so we’ll keep this last bit short (and funny)…
One of the fundamental challenges with incentives is, you know, they cost something. We’ve explored (above) less costly methods of providing incentives – an enthusiastic and supportive corporate culture, for example – but at the end of the day, sometimes it boils down to cold hard dinero. I also know, despite my best efforts to keep the audience here broad, that some of y’all are value investors who eat breath and sleep “bang for your buck” – so how can we stretch our incentive dollar a little further?
Richard Thaler has the answers (as usual) in “ Misbehaving” ( M review + notes) as well as “ Nudge’”( NDGE review + notes), the latter with co-author Cass Sunstein. Thaler provides a number of examples addressing this exact question; I’ll highlight a couple and let you find the rest on your own in the book.
Here’s one from Nudge (page 46). Thaler once had a graduate student who – wait for it, wait for it, wait for it – procrastinated on his thesis. (What a plot twist!) Naturally, being an economist – albeit a behavioral one – our future Nobel Laureate decided to try incentives.
With a twist: using what he knew from his behavioral / psychology research, Thaler realized (and I’m heavily paraphrasing here, to be clear) that sometimes it’s not the size of your incentive that matters, but rather how noticeably you display it.
Thaler frequently references the concept of salience / vividness bias and contrast bias inMisbehaving, and it perfectly explains how writing a series of small checks (much smaller than the graduate student’s stipend) helped the graduate student cure his procrastination. The thesis was broken down into smaller milestones and Thaler was allowed to cash one of the checks for each month that the student didn’t meet a milestone. If I recall the story correctly, Thaler (mischievous by nature) would use the money to throw a party to which the student was not invited.
The student, naturally, met the milestones; Thaler cites other similar stories regarding weight loss pledges between friends with monetary enforcement penalties: never was a meaningful amount of money involved, but the in-your-faceness of having someone withdraw money from your bank account and throw a party without you – when you had the agency to prevent that from happening – is a very real incentive.
Other examples are provided in the interlude of Misbehaving where Thaler helps out a ski resort, which doesn’t have budget to spare. Using disaggregation to separate utility into its “transactional” and “actual” components – product vs. packaging, in other words – Thaler realized that providing people with discounted “ten-packs” of tickets (that most people would never fully utilize) made the value proposition more attractive.
Finally, the concepts of marginal utility and contrast bias discussed throughout Misbehaving lend credence to some of Achor’s discussion of loyalty cards in “ Before Happiness” ( BH review + notes) – it turns out that making the finish line look closer, even if it’s not actually closer, can make you accelerate toward it.
Businesses, similarly, can incentivize their customers to spend more by giving them a “head start” on loyalty cards that allows them to feel closer to a reward, even if the absolute spend required to reach a reward is exactly the same – i.e., giving customers a free two points out of eight rather than having them start at zero out of six…
Application/Impact: A thorough understanding of psychology and human behavior can allow us to create incentives – for ourselves or others – that are costless, or at least more cost-effective.