# Restaurant Lifespans Are Measured in Dog Years: The Cost Structure Behind 29 Years in Chicago

Source: Realpha Blog (blog.getrealpha.com)
Original article and charts: https://blog.getrealpha.com/en/blog/oddlots-2026-10-24-how-a-chicago-restaurant-has-stayed-open-for-nearl/

> Notes after listening to Bloomberg Odd Lots, 24 Oct 2026: North Pond owner Richard Mott on why independent fine dining has negative returns as an industry, why wine held at 33–34% of sales for 28 years and then fell to 24%, and how a shared kitchen spreads capital and licensing across many small food businesses. Educational notes on industry structure, not investment advice — no tickers, no price targets.

Published: 2026-10-24
Locale: en
Tags: Odd Lots, restaurants, moats, shared kitchens, inflation, entrepreneurship

TL;DR: Independent fine dining has negative returns as an industry; North Pond lasted 29 years on a site nobody else bid for and a farmer network it built itself — while wine falling from 33% to 24% of sales removes its most profitable line.


![A prairie-style timber pavilion restaurant at the edge of a city-park pond at dusk, warm interior light glowing, a thin distant skyline on the horizon](/covers/oddlots-2026-10-24-how-a-chicago-restaurant-has-stayed-open-for-nearl-cover.png)

> In governing people and serving heaven, nothing matches frugality. Only through frugality does one yield early; yielding early is called accumulating virtue; with accumulated virtue nothing cannot be overcome… This is called deep roots and a firm stem, the way of long life and lasting sight.
>
> — Laozi, *Tao Te Ching*, ch. 59 (Spring and Autumn period; translation mine)

On the 24 October 2026 episode of Bloomberg Odd Lots, "How a Chicago Restaurant Has Stayed Open For Nearly Three Decades," hosts Tracy Alloway and Joe Weisenthal talk with Richard Mott, owner of North Pond and of Spark Shared Kitchen. North Pond opened in 1997, which makes it 29 years old, and Mott says independent fine dining as an industry earns a negative return. What kept his restaurant alive sits on the cost side: nobody else bid on the park building, and his supply base is close to a hundred small local farms. He also brings one piece of bad news from the past year and a half — wine ran at 33–34% of sales for 28 years and has now fallen to 24%.

## 1981: A coffee counter in the business-school basement

Mott studied finance at Chicago Booth and spent a summer interning in the finance department at Progressive Insurance. Three months in a corporate office convinced him he never wanted to work for anyone else. He bid on the coffee counter in the school basement, added bagels and sandwiches, turned it into a small lunchroom and quadrupled revenue — on rent negotiated for a coffee counter. By graduation he had four locations on campus. Back then Booth had no entrepreneurship courses and no seminars on small-business finance; classmates asked him, "What's the matter, Rich, didn't you get a real job offer?" Today the largest single concentration at Booth is entrepreneurial studies, bigger than marketing or finance.

Coffee was a different trade then. A seven-ounce small, a ten-ounce large, beans bought in five-pound bags and scooped out, and whatever was left when school closed in June went into a container until September. His wife came back from a new Starbucks saying the place sold little but coffee, yet the coffee and the service were great. His reaction: how long can a store that only sells coffee last? He went on to open twenty-odd coffee shops himself.

## Late 1980s: Squeezed out by the comprehensive contract

At the peak he ran 23 shops across five universities, small operations inside academic buildings, staying away from the main board plans. Then universities started signing comprehensive contracts — laundry, cafeterias, buses, stadium concessions all handed to Aramark or Compass — and nobody put a single coffee shop out to bid any more. A tiny niche company gets swallowed by that. Campus coffee runs September to May, so he took on Chicago Park District concessions to fill the summer, and the summer fix turned into his next business.

![On the left, five services each go out for their own bid; on the right, one large frame swallows four of them, while the fifth, the coffee shop, is left outside the frame with no bid to enter.](/figures/bundled-contract-squeezes-out-niche-en.svg)

Looking back, he says he opened that first business with no idea what he was doing and made everything up as he went; his advice now is to go work for someone else first and let them pay for your mistakes. Hiring was the hardest part — it took him years to learn to hire for the job rather than hiring people because he liked them. One more reversal: fifteen years ago he would have said success is 80% the idea; now he says the idea is 20% and the rest is the team, the environment and the support network. When Pat from Home Chef came to him, he thought buying food online and having it shipped was a terrible idea, but he loved the young man — and that company had a great run.

## 1997: The building nobody else bid on

Next to the Nature Museum in Lincoln Park stood an old warming house. Mott says that corner is one of the prettiest spots in the park now; back then it was abandoned ground — a homeless hangout, prostitution in the area, a pond that looked like a suburban drainage ditch. Friends later asked whether he had someone at City Hall. He didn't: "We got the contract basically because no one else bid on it."

The Park District's frame of reference was concessionaires running bad hot dog carts that stole from them, so a fine dining restaurant in the park with a liquor license went over their heads. Permission alone took two and a half years, the build-out another year. It was the first liquor license in the park, and he was not allowed a bar and had to include a takeout window. It opened as North Pond Cafe, a little more casual. A few years later a new superintendent and park board president let him remodel, add the front room and add a bar, and the place started being treated as a fine dining restaurant.

## After 1997: Before anyone said "farm to table"

The second chef was a Wharton finance major who switched careers; he stayed more than twenty years and left in December 2019. They decided to build the restaurant around local and seasonal food before the phrase for it existed. The sign read "North Pond Seasonal American Dining," and people called to ask which season they were open. The menu began listing farmers, and guests asked who Mickey Klug was and what a Mickey Klug pear might be. That chef went on to co-found the Green City Market.

The supply network now runs to nearly a hundred small farmers, fishers and cheesemongers. These are not the third-generation Iowa families farming 1,200 acres of corn — one of them got thrown out of the sociology PhD program at Northwestern and now grows organic beets in northern Wisconsin. Twenty-nine years on, that network works as a cost shield: the man growing organic artichokes in Wisconsin has a cost basis that tariffs and the Strait of Hormuz don't touch, and diesel prices move it less.

## The arithmetic of the trade: no moat

Tracy asks about margins directly; Mott jokes that the weather looks good this week, then answers. Independent fine dining is one of the worst businesses to be in: zero barriers to entry, anyone can open, and nothing you make is proprietary — invent the croissant-donut mashup and the whole world has it the next day.

Two things get less airtime. First, if you intend to obey the law — labor law, immigration law, overtime — you operate at a 5–10% cost deficit to your competitors. Wage theft and overtime abuse were rampant in this industry, and the people absorbing it were not going to complain to the labor board. Second, you never want to be in a business for economic reasons while competing with people who are in it for non-economic ones. Nobody pays $12 billion for the Dodgers because of the cash flow. A doctor who made millions would not buy an auto towing business, because he knows he knows nothing about towing — yet everyone who eats out thinks they can run a restaurant, and everyone whose mother has a good chocolate chip cookie recipe thinks they should start a cookie company. What the business actually looks like, Mott says, is six o'clock on a Sunday morning, unclogging the drain.

He cites SBA and Wharton statistics for how ugly the failure rate is: behind every name that made real money in this trade there are forty you've never heard of whose restaurant closed and who went back to being a dentist. Restaurant lifespans are measured in dog years; a place that runs twenty years is the *Cats* of Broadway, and rare.

![One large circle on the left stands for the one success whose name you can recall, while forty small grey circles packed on the right stand for all the restaurants that closed.](/figures/one-name-forty-closed-en.svg)

## The last 18 months: The missing percentage points

North Pond's prix fixe is $130 for five courses, with nine or ten people in the kitchen and pastry department on a Saturday night, plus waiters, back waiters and busboys on the floor. The same table orders a bottle of wine, average price $110; the waiter walks to the cellar, brings it back, opens it, sets it down — two minutes — at roughly the same margin as the food and with almost none of the associated cost.

![Ten figures on the left stand for the people it takes to put out one five-course menu, a single figure on the right stands for pouring one bottle of wine, and two gross-margin bars of equal height below show the profit is about the same.](/figures/wine-versus-tasting-menu-effort-en.svg)

So here is the number that makes every restaurateur nervous: for the first 28 years, wine ran at 33–34% of sales, like clockwork. Over the past 18 months it fell to 24%. The slice that disappeared was the most profitable business in the house. Younger guests drink less, some for health, some pulled away by cannabis and CBD — Lollapalooza sold THC margaritas for the first time this year. North Pond has put in a non-alcoholic cocktail menu and has a proposal with the Park District to add THC drinks, not yet approved. On GLP-1 drugs, Mott says he can't separate the effect from other megatrends and can't recall the adoption figure; his instinct is that someone on one of those drugs doesn't want five courses, the desserts and a glass of port.

![A line that held between 33% and 34% for twenty-eight years drops steeply to 24% in its final short stretch, a fall of about nine to ten percentage points.](/figures/wine-share-breaks-after-28-years-en.svg)

## Now: Costs arrive the same week

Starting cooks cost about 30% more than three or four years ago, and food is one of the highest-inflation categories in the economy. The business has no lag: what goes on the plate was bought that week, so inflation shows up in costs immediately. They have raised prices four times in two years, roughly every six months, where the old rhythm was one annual review.

![Cost is a fine staircase climbing one small step every week, price is a coarse staircase that jumps only once every six months, and a gap stays open between them where price has not caught up.](/figures/cost-weekly-price-semiannual-en.svg)

The cost lands on the guest's side: dinner for two at North Pond with the wine pairing runs about $400, and to justify that you have to raise your game on food and service alike. People come for the experience. The shift he sees is a rising share of guests arriving for the 25th anniversary, the 50th birthday, the college graduation — and fewer of the "it's Thursday, let's treat ourselves" kind.

## Spark Shared Kitchen: Spreading capital and licensing

The other line is a 60,000-square-foot warehouse in Brighton Park. Shared kitchens and ghost kitchens are both third-party facilities used by multiple vendors; the difference is purpose and location. Ghost kitchens host takeout-only brand outlets and have to sit in dense residential areas to reach customers in minutes; his building is in an industrial zone two-thirds of the way to Midway Airport, worth maybe a third as much per square foot as the real estate a ghost kitchen needs.

![A densely built city block on the left carries a tall column standing for expensive ground, while a sparse industrial district on the right carries a column only a third as tall.](/figures/ghost-kitchen-versus-shared-kitchen-rent-en.svg)

His tenants are mostly small caterers, and there are two ways to rent. Fully equipped kitchens go by the hour; or you take a kitchen of 400 to 1,600 square feet with your own key and 24/7 access, bring your own equipment, and he supplies the hood, the sinks, the flooring and the floor drains.

Opening your own cookie business means finding a licensed facility, a landlord who wants a three-to-five-year lease plus a deposit, and the whole licensing process: six months minimum. Through a shared kitchen, where he holds the operator license, you pick up a user license at City Hall on Monday and you are in business on Tuesday, with ingredients and the license fee as your only upfront spend.

![A long progress bar on top is split into finding a space, lease and deposit, and the permit process, marked six months at best, while a very short bar below holds only two cells, getting the license and opening the door.](/figures/six-months-versus-two-days-to-open-en.svg)

Three people came to him wanting to start ice cream businesses, so he built out an 800-square-foot space without a hood and bought a $26,000 five-gallon industrial ice cream machine. One of them makes Hispanic-themed flavors, another Asian flavors; a tenant can come in for a day, make 200 gallons, put it in the deep freeze and spend about $1,000 including ingredients. On your own, $5,000 buys a one-gallon machine that makes ten gallons a day. The equipment is expensive and any one person needs it two to eight hours a week — sharing it is where the gap comes from.

![Twenty squares on the right stand for the shared machine's two hundred gallons a day, a single square on the left stands for an owned machine's ten gallons a day, and only a short segment is used on a one-hundred-sixty-eight-hour axis below.](/figures/shared-ice-cream-machine-capacity-en.svg)

Short of hospital rooms, he says, restaurant kitchens are about the most expensive thing you can build per square foot, but tenants are easy to find. On licensing he gives a grade: the eighth floor at City Hall used to be the department of no, things started changing while Rahm was mayor, and it has gone from a D-minus to a B-minus.

## The question on your mind: Good news arrives, so why doesn't the company hold up?

The mistake I make most often is seeing a "structural growth story" for an industry and assuming the companies inside it benefit. This episode makes the counterexample plain: an industry everyone can see, and everyone thinks they could run, earns a negative return as a whole. The difference is whether anyone can lock that demand down. With no barrier to entry, no proprietary recipe, and a croissant-donut copied by tomorrow, the gains from growing demand get split among new entrants.

So I ask one more question: who collects the rent here? North Pond's answer isn't on the menu. It's the building nobody bid on, the first liquor license in the park, and a farmer list assembled over two decades — three things a competitor would need years to rebuild. Spark is blunter still: it earns on capital and licensing, the two chokepoints a tenant's cooking skill can't route around. Next time I hear "this industry is taking off," I'll look for which layer is the chokepoint and whether the people in it can raise prices, before I look at the loudest name in the sector.

![Three horizontal bars of different lengths show how many years it would take someone else to rebuild each of these three things, the longest being the roster of small farms at 29 years.](/figures/how-many-years-to-rebuild-en.svg)

## The question on your mind: Is a falling number noise or structure?

Wine going from 33% to 24% of sales deserves a second look, because it has the shape of every "bad quarter" number while the structure underneath is different. I check three things: the series sat at 33–34% for 28 years, so the move is already outside its historical range; the part that fell is the highest-margin line, which amplifies the profit effect; and the explanations offered — young people drinking less, cannabis drinks on menus, weight-loss drugs spreading — all run for years.

Mott also names the soft spot himself: he can't isolate the GLP-1 effect. So the hard part of this reasoning is "the wine share fell nine or ten percentage points," and the soft part is "how much each cause contributes." The way I file it: a long-stable ratio breaking ranks above any single revenue miss, while attributing the break to one fashionable cause waits for more periods of data.

There's also a way to check his conviction against his behavior. He has already added the non-alcoholic cocktail list and filed the THC proposal — those are his bets on his own read. I'll treat those two menus as the observation point: in six months, has the non-alcoholic column gotten longer?

## Sources worth your time

- Bloomberg Odd Lots, 24 October 2026, "How a Chicago Restaurant Has Stayed Open For Nearly Three Decades," hosts Tracy Alloway and Joe Weisenthal, guest Richard Mott
- North Pond (Chicago)
- Spark Shared Kitchen (Chicago)
- Green City Market
- University of Chicago Booth School of Business and the Polsky Center
- Restaurant failure-rate statistics: the guest attributes them to the SBA and Wharton; no specific figures are given in the episode

## One thing to take with you

Businesses that last often win at the spot nobody else could be bothered with. North Pond got its site because no one else bid; Mott spent two and a half years on permission and another year on the build-out, turned an abandoned corner into the best location in the park, and has stood on that spot for twenty-nine years.

Something I've tried: write down one thing you're willing to do that the people around you find tedious or beneath them — taking an older relative to a hospital appointment, proofreading a friend's draft, turning your neighborhood group chat's running tab into an actual spreadsheet — then actually do it once this week, and note who comes to you afterward because of it. That list is your own version of the site nobody bid on.