The Setup
Let’s pretend for a moment that you own a car wash. For $5, customers can drive in for a hand wash and wax. One day, in the midst of the stream of Hondas and Subarus, a customer pulls up in a brand-new Ferrari (probably a janitor at Facebook, you think to yourself) and asks for a wash.
“That’ll be $10,” you say.
“Hey, what gives?” the driver shoots back. “Your sign says $5. Why are you trying to gouge me? Do I look like I’m made of money??”
You then proceed to explain to the driver that, no, you’re charging her double because it costs you more money to wash her car. After a little more explanation, Miss Ferrari nods and hands over $10, and your crew gets to work sudsing up that prancing horse hood ornament.
The Punch Line
You’re probably wondering what, exactly you could have said to your customer to convince her that it does, in fact, cost you more to wash her car. After all, it doesn’t take particularly more soap, water or time to wash a Ferrari than it does to wash a Ford. The added cost doesn’t come from increases in materials or payroll, it comes from an increase in risk. Specifically, it’s the risk that someone will put a big scratch in that rosso corsa paint job, and you’ll be out $10,000 spent on a museum-quality fender restoration. For a client in an economy car, that same damage would cost you perhaps $500 to repair.
We can calcuate our risk with the simple equation:
risk = probability x cost
Let’s assume that the probability of damage remains constant, and that you end up paying out a claim once for every 2000 customers. Plugging in the numbers, we get a risk of $0.25 for your everyday clients, versus $5 for the Ferrari. Assuming that your goal is to make the same profit on each car through the line, your $5 exotic car surcharge is looking just about right.
You might be thinking that a good insurance policy would help here, and it would, but not in the way that you might think. Although we’ve modeled the risk as a continuous probability, the real outcome for this one driver is either that her car gets through unscathed, in which case you make $10, or it gets damaged, in which case you lose $9,990. There is no middle ground. With insurance, you can trade off the risk of losing big-time for the certainty of earning the average return, minus a small profit for the insurer.
Real-World Ferraris
I cooked up this analogy a few years ago after encountering a similar problem on the job. We were working with a vendor who would take a relatively expensive electronic component from us on consigment, do a few processing steps in a factory in China, and then return the finished component to us. When they provided us with a quotation, they included a line item for “breakage”. The vendor had estimated that they would damage a certain percentage of the incoming parts that we sent to them, and so they added this risk to their quote to offset the claims that we would inevitably file for lost materials. Just for the sake of argument, let’s say that the component cost was $10, and that the vendor was getting paid $0.10 for each part; for each part that gets ruined, they’d have to process 100 more without incident just to make back their money.
More generally:
“If the risk of loss on a transaction is a large fraction of your potential profit, then you’re washing someone’s Ferrari.”
In the time since that first supplier meeting, I’ve learned to spot Ferrari-washing in other people’s businesses. For an easy example, take any product, tack on the phrases “medical”, “aerospace” or “defense”, and watch the price increase by 10x. Sure, there’s some profiteering going on here, but the penalties for a single critical failure (plane crash, pacemaker failure, inadvertent missile launch) can easily wipe out several years of profits, or worse.
A Lesson For Entrepreneurs
My takeaway is this: in the early stages of planning a new business, don’t let “disruption” be another word for “turning a blind eye to risk.” If your business model is telling you that you can offer a product or service for much less than the incumbent competition, take a careful look at how much it’s costing your competitors to cover their risks. Especially when your business is small, you might get lucky, but as you scale, even risks that seem improbable will become likely.
risk=probability x cost is OK theoretically but with just 10 bucks you are not going to compensate for any scratch on the ferrari’s paint. And you do not get those many ferrari’s in your car wash to make up for that single event.
This reasoning (risk=a single number from a probability completely unknown from an unreal model) is at the core of today’s crisis.
Agreed. One of the better books dealing with risk to come out over that last 5 years was the “The Black Swan” by Nicolas Taleb. If you aren’t familiar with the book. It basically states that your estimate P in R= P*C is probably wrong. Also, C is probably wrong. The whole thing is probably just wrong.
In your case, what your model didn’t account for is that you don’t perform background checks on your employees. Turns out one of your employees did 3 years for auto-theft and larceny. While you were worried about the scratch he just stole the car and has crossed state lines. Have you never seen Ferris Bueller?
We live in a world of regime shifts.
No, corporations (banks) being forced by the government (US) to take on unreasonable risks (sub prime loans) is a big portion of the cause for today’s economic mess. We could get into the long chain of events which led to the current economic woes, but let it suffice that it began well over a decade ago.
risk doesn’t equal probability x cost, it equals probability x severity.
Severity might = equal big bill for scratching a Ferrari but no car wash would be open to that liability (‘clients coming here can get their car totally fucked up and we’ll just laugh’ kind of clause)
But I like the blog, excellent.
Wouldn’t cost and severity be the same, for insurance purposes?
Obviously it’s a little complicated if there’s a question over contesting that the damage was caused by your business, and the legal costs of fighting it, but I’d have thought that, in a basic, simplified version, severity and cost would be the same?
Just another note on that, more directly on how corporations handle risk. I handle bids where clients say (eg) ‘if you run over the end date we’ll charge you $100,000 per day up to a max of $500,000′.
Guess what we put in our bid to cover the risk? $500,000. (and so do all our peers so we’re not losing a job through doing that). And if the risk doesn’t materialise, we keep the money.
I don’t know how many times I’ve lectured client groups on this – YOU CANNOT SELL YOUR RISK TO SOMEONE ELSE.
But do they listen??
Nice equation for risk-analysis and a great message.
Regardless of the correct answer, you have everyone thinking about entrepreneurship and problem-solving and this is what has made America great over the years. Now if we all just apply this type of thinking to the election being held on November 6th, surely we will come up with the right answer to the question, “For whom should I vote?”
Good one, i like the explanation…..