Econ 101: Economies of scale
Why do hotels want to get bigger?
When you hear someone say “Econ 101”, it is usually not a reference to an academic class so much as it is a bludgeon by which to assert that something is obviously prudent: “it’s Econ 101.”
Uber’s surge pricing? Econ 101. Free trade? Econ 101. Abolishing the minimum wage? Come on, it’s Econ 101!
It’s an argument from obviousness, like “even a child understands that…”, or “if you think about it for just a second you’ll realize that…”, or “it’s just common sense.”
The problem is that the lessons you learn in an introductory class are often un-done at the more advanced levels. Gravity is not a force. Electrons are not necessarily small. DNA is not the sole carrier of hereditary traits. Options cannot be valued like a continuous heat diffusion processes. A computer is not a Turing machine. And consumers may prefer time-rationing to price-rationing of goods like taxi rides.
That is, learning more does not always push your beliefs in the same direction. Sometimes it pushes back toward where you came from. And there’s no guarantee that the flip-flopping will ever “settle down”. Beliefs do not converge with increasing knowledge, and certainly not monotonically.
We have lots of memes and sayings that attest to this. There’s the joke, “everyone knows what money is, except economists.” There’s the Good Will Hunting scene where Matt Damon is like four levels beyond the Harvard jerk and says he’ll believe such-and-such about the antebellum Southern economy until he gets to the next book. There’s the bell curve meme where the dumb guy and the smart guy both see the truth that the “mid-wit” in-between misses.
This also gives rise to the genre of books telling you that “everything you thought you knew about” whatever “is wrong”. I personally feel like those books have gone too far, p-hacked to the point that time-tested conventional wisdom has been supplanted by a novel slate of phenomena that don’t replicate.1 As G.K. Chesterton wrote in 1905, “The word ‘orthodoxy’ not only no longer means being right; it practically means being wrong.” So I guess this isn’t new.
But I’m not surprised that people are receptive to those ideas when so much common sense has been proven wrong. The ground feels solid, and yet it moves — e pur si muove.
No job of mine to date has borne any resemblance to the principles I learned in Economics 101. My dad used to say things like: you should work to charge customers more than it costs you to provide some goods or services; the difference between your revenues and your expenses is your profit, which you should try to increase. But that has never mattered.
My first job was working for the magazine Bloomberg Businessweek, which is heavily subsidized by Bloomberg’s enormously profitable financial services business. Our job was, in effect, to make a splash — to win awards, to increase the cachet of the brand, and to entertain and inform the sole owner. My second and third jobs, Kensho and Observable, have been venture-backed technology startups, which feels kind of like farming for miracles. People spend millions to attract a nexus of weirdos to commune with the inscrutable ethereal muses until one in a hundred lands on an idea that scales almost for free to naturally monopolize an as-yet-undiscovered niche of the human condition.
Of course, you can’t think that way while doing the job. You think about pursuing excellence in your craft, out of faith that success springs, not from selfishness, but from (as Musil wrote) “a deep love of what might be called the generally and suprapersonally useful, in other words, from a sincere veneration for what advances one’s own interests—and this not for the sake of advancing them, but in harmony with that advancement and simultaneously with it, and also on general grounds.” And of course a loss-leader magazine or a venture investment can be seen from a distance through the orthodox lens of maximizing the net present expected value of a stream of future earnings. But, on the ground, day to day, in practice, it feels like capitalism incubates within itself a more alien kind of enterprise.2
The Apple Tree Inn, in contrast, is the closest thing I have ever experienced to a classically recognizable business. It’s also the most quaint and the least corporate, and is very much a product of Claire’s personal passion. But I think the smallness of “small business” makes it a petri dish in which to more clearly observe economic phenomena. You really have to care about supply and demand, profit margins, opportunity costs, principal-agent problems, the time value of money, and certainly menu costs.
In this series, I will reflect on those concepts through life at the hotel, starting with…
Economies of scale
I wrote almost a year ago about how our first-ever guests left just a couple hours after they checked in because they were the only room booked that night and the hotel felt “creepy”.
In retrospect, they must have partly been reflecting our own nervous energy. There have been plenty of times since then when someone has been the only guest, and it has never again been such a problem. But one complaint was that, when they got back from dinner, there was no fire burning in the lobby’s fireplace.
For the next week, I was almost manic about keeping that thing going. I had it all ready to go and as soon as a guest drove up, I tried to light it before they got far enough in to see. I remember the fear that gripped me whenever a group beat me, when they turned that corner and it had gone out or hadn’t been lit yet. Seeing me scramble for the hearth probably created its own nervous energy, but I didn’t want to repeat that first mistake!
Tending to the fire is an absolute pleasure, my favorite part of living at the hotel. Instead of that Apple Watch feature that reminds people to stand up, flame turning to embers reminds me to go flip a log over and put another on.
But it quickly became clear that the math of making it part of our service standard is brutal. Having one couple at a hotel is not like having one couple staying at your house. On average, they are on their way to somewhere; they check in, scope out the lobby, either experience the charm of a fire or don’t, and then barely appear again until check-out. If you want the fire going when they arrive, and don’t know when they’ll arrive, you have to keep it going for like an hour for every minute of guest enjoyment. The possible upside is a nice comment; the possible downside is a painful cancelation.
In those quiet early days it felt hard to justify going through the wood, let alone the monitoring. I can’t live in the lobby, and Claire can’t see the fireplace from the front desk.
The front desk has a similar payoff profile: sparse utilization, low upside from getting it right, high downside from getting it wrong. You have to staff the front desk for like an hour for every minute of use it gets, because if someone’s not there when a guest needs something, it could be a disaster. Lots of negative reviews involve standing waiting alone at an empty front desk.3
That means you can’t afford to have a lot of slack in the labor. For the basic functions of check-ins and answering questions, you need, on average, much less than one person — but you can’t have zero. So that one person is simultaneously underused and unable to leave. In the idle time, you can only do things that you can do at the desk and drop at a moment’s notice.
At that scale, where the “table stakes” are expensive relative to your labor, it’s very hard to justify adding job responsibilities that add delight. There is some disconnect here between the ideas people have for us and the realities of how a very limited staff needs to spend its time. Picking flowers for breakfast, shuttling people to Tanglewood, putting out cookies in the afternoon, lighting candles for dinner — all fun ideas for delightful and memorable touches. But your one person can’t leave the desk to do them, and you can’t afford a second person. You can do them as one-offs, but guests’ expectation of consistency and fairness imposes a categorical imperative: in the long run, you can only do that which you can always do for everyone. So it’s hard for any task to live in the zone where it’s in-between zero and one employee’s worth of work.
Think of ten tasks with diverse job requirements. Then think of the ideal distribution of labor among them. Like 15% on task A, 5% on task B. If you are small, you cannot hire 5% of a person. But if you are bigger, you can assign that task to 1 out of your 20 people. “One” is a very special number of people that produces a floor for what’s worth doing, and scale gets you over that floor for more kinds of things, which lets you more closely match your desired distribution of labor. A software startup of five generalists already clears that floor — but a hotel is a “full-stack” enterprise spanning especially diverse specializations, each of which must clear that floor.
Something similar happens with the distribution of space. We have thirty-four rooms (across both buildings), one front desk, one lobby, and two dining rooms. I wouldn’t have been able to guess this before Claire took over, but that turns out to be way too much common space per guest room. I’d guess that for every ten rooms booked you get like one person using the common space. Every room of the main house can be full, and there are still times when people ask “Is anyone else here?”
And in the back office, many operational costs of the hotel are basically fixed, including branding, marketing, photography, social media, web development, accounting, and “revenue management” (the competitive research by which you set how room rates vary over time). We could probably have ten times as many rooms and spend basically the same on those.
This is why Claire’s old employer Starwood would never even think about hotels of fewer than like a hundred “keys” (rooms).
We can fill the space and taste that bigger scale in three ways: Tanglewood concerts; private events; and the food & beverage operation, especially with live music.
The fireplace math changes — nonlinearly! — when it gets busy. Saturday night was the second-busiest night we’ve ever had at the bar, and it felt great to pile the logs in the fire high, fast and often, knowing that each lick of flame is charming fifty people per second instead of, like, a fraction of a person. And if it gets low, the guests stoke it! It becomes totally self-sustaining. And then you get people reading by it, huddling around, raving about it, playing chess, and so on. That’s when you really feel the primeval charm, progenitor of all other fixations on flickering light. (My favorite moment at the hotel was when Garrett, Peter, and Matt befriended all the Bates girls around the outdoor fire pit last October.)
The same fire can be seen by one person, or by fifty people. The fire is broadcast. That’s an economy of scale.
When I hang out at the hotel, I can serve as “the slack in the rope”. Everyone else is busting their asses bussing tables and cleaning dishes and fixing leaks and otherwise satisfying the expensive table stakes — but I have zero stated responsibilities, so I can float around and have extended guest conversations, make annotated placards for books of historical significance, film interviews with the bands, and drive carless people to their dinner in town. (There’s no Uber here!) I think of these tasks as a sort of preview of what the hotel could do systematically if it had a little more scale.
And sometimes, when I befriend guests, they remind me of the diseconomies of scale. Many of them have sought us out because we’re small, quiet, and independent. Big service becomes impersonal, routinized, mechanical. Sometimes Claire tells people they’ll be the only ones and they’re like, “amazing!” Scale is evil! Why do these corporate types want to make everything bigger!
But when there’s literally one room booked, and it’s cold and icy, and Claire has just learned that she needs to file something called a certification of entity tax status with the Massachusetts Department of Revenue, and she can taste the tantalizing opportunity to make everything much better for everyone, it gets hard for her to wake up early and set out the continental breakfast. It doesn’t feel worth it. It’s an emotional drag and distraction from investing in more scalable longer-term work.
There’s scale in space, but also scale in time — the scale of a stream of guest-facing pleasantries stretching out day after day into the future. For the first few weeks you feel like you’re welcoming someone into your home or throwing a dinner party for friends. It can’t feel like that forever — but you kind of have to tell guests it does. You have to wake up and exult in the chance to do it again. So you have to make it sustainable. And scale in space can alleviate the monotony of scale in time.
The front desk rotation is far more mature today than it was even six months ago. When Abby’s there, she’s not idle in the times between guests interrupting her — she’s working on private events. When Christian’s there, he’s overhauling the phone systems and booking engine. When I’m over at the hotel, I still get that pang sometimes where I’m scared the fire has gone out — but, increasingly, I find that someone must be keeping an eye on it. I don’t even know who. The bartenders, the bussers… someone. One way or another I more often find it already roaring.
We always knew that it’d be nice to have real candles at dinner. But it was only when things got going well enough that Claire could hire Sean as bar manager that it started happening consistently. He’s not the chef, he’s not the bartender, he’s not the dishwasher — he’s not tied up in table stakes. He’s able to own a holistic category of the experience and bring his imagination to unnecessary tasks like lighting candles every night for a better ambiance. He thinks up special timely cocktails each weekend and prints little insert blurbs to clip atop the main the menu. He throws a New Year’s Eve party. It’s amazing, and has improved the F&B operation greatly. But that takes hiring a whole person.
Recently asked questions
“Is this original?” — Guests fond of historic houses love asking this about the floors, tiles, chandeliers. I have no idea.
“How many people does the Ostrich Room fit?” — Someone’s writing a story about our music program and wanted to know. It’s about fifty.
“Can I get water with a little cranberry juice in it?” — Sure! A woman at the bar on Saturday asked Claire.
“What weekends in April is room 2 available?” — Not the first two, yes the second two. Some prospective guest only wanted to come a time they could stay in room 2.
“Could we do the continental buffet breakfast, but with alcohol?” — For private events, we offer either basic continental breakfast package, or the hot foods and cocktails menu. They wanted something in-between. I don’t actually know the answer.
Case in point: or do they? Ugh… I’m tired of updating my beliefs!
I don’t think “Bullshit Jobs” is quite the right framing, partly because I’ve been lucky enough to love my jobs, but Graeber does get at some profound opacity and dislocation between the theoretical microfoundations of business and the embodied microfoundations of working. (I haven’t read the book yet.)
This is why the previous owners chose not to staff the front desk at all. Its curtains were clamped shut; calling the hotel got you a call center in the Philippines, and you checked in by a mobile app. That is not Claire’s vision for the space, but it is an understandable economic response to this problem of scale.