Paying Rent on a Seat

When the Linkery got a little bit of national attention for our service-charge/no-tipping business model, it started a chain of events that led, among other things, to my becoming friends with Bruce McAdams and Mike von Massow of the University of Guelph.  Over lunch with them on a beautiful Toronto day, Mike, in passing, said something like (to the best of my recollection):

I’ve talked at length about this with consulting clients when I work with them on pricing. We’re not charging for the food and drink. We’re charging rent on the seat. Everything else we do, the pricing, the upselling, it’s all just a way to extract rent for the time a guest is occupying a chair in our restaurant.

The genius of this observation, in my opinion, is hard to overstate. It cuts to the quick of restaurant pricing, restaurant value, and the weird reaction restaurant operators have on hearing the words, “it’s expensive for what you get.”

To really understand the math, we need to lay out a typical restaurant operating budget. A simple model that is often used in textbooks is approximately this:

Cost of food & drink (COGS): 35% (of sales)
Labor: 35%
Rent: 5%
Operating Costs (utilities, insurance, bank fees, etc): 15%
Profit/Return on Capital: 10%

A more typical model, that we independent restaurant operators would discuss over bars in San Diego, is:

COGS 27%
Labor & Management: 33%
Rent: 10%
Operating Costs: 30%
Profit/Return on Capital: 0%

Two asides:

1) At a certain type of larger restaurant, they can get their COGS below 20%, which is a good way to make profits although not good for food quality.

2) In any circumstance, when you do a really high volume of sales — perhaps through a lot of to-go orders, or just being full from open to close — the percentage that goes to labor costs and variable costs starts to come down quite a bit and profits quickly improve. “Volume,” a friend used to say, “cures all ills.”

Back to pricing. It turns out that, once a restaurant pays for its cost of goods, labor and rent are by far its largest remaining costs. Furthermore, its other operating costs are largely fixed — specifically, they are a function of maintaining a brick-and-mortar location.

So, on the cost side, after paying for the raw product, costs are driven principally by the size of the establishment. Every hour you’re open, you need a minimum amount of people working the facility, determined by the number of seats available. Because, if you suddenly get a wave of guests, you have to take care of them.

All this means that, whether occupied or not, every seat costs money every minute.

On the revenue side, for table-service restaurants, the limiting factor is the number of seats in the restaurant. Or specifically, the amount of sales that can be generated by any seat. Sometimes guests complain that certain American restaurants are too focused on getting them out of their seat after dinner; I suppose the truth is the restaurant doesn’t care whether they stay or leave as long as they order a certain amount of food and drink per hour.

The fundamental strength of the business model of a table-service restaurant is measured by the equation:

number of seats * (seat revenue after COGS – seat cost)

The last part of that equation, seat revenue minus seat cost, has to be positive for the restaurant to work financially. The name of the game, then, is to generate as much revenue per hour per seat as you can while making it seem like your prices actually reflect the value of the goods that you’re selling. This a weird situation, right? Economically, the seller is selling the opportunity to sit in a restaurant and enjoy some food and drink; and the buyer is buying the food and drink itself!

No wonder this leads to a lot of situations where the guest feels as though she’s gotten bad value; that guest looks at what’s on her plate, or compares the retail cost of the wine or beer she ordered, or judges how satisfied she is with the culinary preparations — and meanwhile, the cost of providing those things is largely dictated by unobvious, orthoganal forces like the commercial rent market, loan rates, labor costs in the community, investor expectations, and so on.

As I pondered what Mike had said for a few weeks, it became obvious to me that there is a solution. A pricing model for restaurants that actually reflects their cost structure — and it’s totally do-able.

Food would be sold more or less at cost, perhaps with a 30% markup — as opposed to the usual ~300% markup you’d currently find. In other words, a steak that now sells for $32 in a restaurant would now sell for $11 on the menu. However, there would be a seat surcharge — a cover charge, if you will — for every second you were in the restaurant. Because, eating or not, your presence in that seat generated labor and operating costs, and you also have to pay the base rent for that square footage. When you walk in and sit at the table, the clock would start running, and when you went to pay, the seat surcharge would be added to your check.

Of course, the seat surcharge would cover labor, operating costs, and have a little added for profit.

The end result would be that you would get a hell of a deal on the food, and pay for all the other costs of the restaurant in a way that reflected your use of them.

Am I going to institute this pricing model at my next restaurant? Nope. But I do think it’s interesting to keep it mind as a possibile alternate universe, and, when I go to a restaurant, I often mentally break down my bill as though it had been itemized this way.

Because, I now understand, I wasn’t paying much for the food. I was mostly paying rent on a seat.