Is it a Capital Problem or an Operating Problem?
When to build, and when to adjust operations
The Story
In the movie National Lampoon’s Christmas Vacation, Clark Griswold (Chevy Chase), in an effort to ensure a “fun old-fashioned family Christmas” for his wife and kids, his visiting parents and in-laws, and his uninvited brother’s family, sets about installing 25,000 twinkle lights on his house. At first, they don’t work, but once Clark realizes that he wired them through the garage light, he throws the switch, turning on the lights, and causing a city-wide power shortage, not to mention inconveniencing his snooty neighbors. He also seeks out other forms of holiday perfection, like taking the family out to buy the biggest Christmas tree ever, yet, despite his best efforts to create a storybook experience, he cannot control the behavior of the other characters, who don’t get along very well and aren’t getting in line with Clark’s idea of the Christmas spirit. Clark’s story is about overreach—and doing more than is required in the situation.
Let me illustrate the idea with a pair of more pedestrian stories, one about overdesign and turning an operating problem into a capital project, and the other about trying to solve a capital problem with operational band-aids. I live in a condominium building that has 60 units. The building is a converted warehouse, and a big feature is a three-story, 45-foot-high atrium. One side of this atrium is a large, blank concrete block wall, so about twenty years ago, we decided to hang art on it and create an art gallery. A wonderful artist in our building runs the gallery, which mounts three shows per year. We have a big opening for each show, with wine, cheese, and music, but there is just one small problem: The developer of the building did not think to include a toilet for visitors in the public part of the building. For the first couple of years, the artist who manages the gallery would leave her unit open so visitors could use her bathroom, and pretty soon the condo association began to hire a security guard to stay in her unit so she could be at the opening just for that one evening, three times a year. And then, after a few years, an idea surfaced: Should we consider spending money to build a dedicated bathroom in the public area of the building for visitors to use during these events? So the condo association’s board of directors had a study done and determined that it would cost over $20,000 to build the restroom. But cost wasn’t the only issue: There was no available space to put the bathroom, and on top of that, the construction would be very disruptive and have a permanent impact on the aesthetics of the space, because it would require chopping out a trench diagonally through the finished, industrial concrete floor in the atrium. In the end, the board decided—I think wisely—to stick with the rent-a-cop for three nights a year. This is an example of how we can almost talk ourselves into a capital solution—a design and construction project—when an operating solution is easier, less expensive, and more flexible.
Here’s the second story. Our building was completed in 2002, and 23 years later, some systems had reached the end of their useful lives, so we have had to complete several capital improvement projects. For example, ten years ago, we repointed the exterior brick facades, and a couple of years ago, we replaced the entire roof after years of patching to control leaks. More recently, some of the several hundred recessed can light fixtures in the common areas have stopped working, one at a time. Originally fitted with incandescent bulbs that were later replaced with compact fluorescent bulbs and, most recently, with LEDs, the fixtures and housings themselves are now failing. Simply changing the bulbs is no longer a solution, as some of the fixtures/housings break while being handled. What to do? There were two choices: Replace all the fixtures as a single project, or just replace them one at a time, as they fail, in what is known derogatorily in the operations world as a “break-fix” maintenance model. There are a couple of problems with this approach. The first is that, on a per-fixture basis, it is a lot more expensive in the long run to mobilize an electrician or two—including travel time, tools, materials, and ladders—to replace an individual fixture than it is to do them all at once. A second problem is that you are constantly experiencing outages, one at a time. A third problem is that the color temperature of LED bulbs can vary depending on the manufacturer and age, leading to differences in color temperature among bulbs in the same space. And finally, the design and manufacture of the fixture housings can change over time, making it more difficult to source the same parts. All of which argues for replacing them all at once. Fresh off the recent assessment for the new roof, however, and fearful of the residents’ reaction to yet another assessment, our condo board decided that it would be better to just continue replacing the fixtures one at a time as they fail.
The first story is about mistaking an operating problem for a capital problem. The second story is about mistaking a capital problem for an operating problem. Here are three city-scale examples that illustrate the same kinds of mistakes (thanks to ChatGPT):
The $2.8 billion expansion of the Katy Freeway in Houston, completed in 2008, widened the highway to 26 lanes in an effort to reduce severe traffic congestion. Instead, it induced more demand, and within a few years, travel times had increased. Operating solutions that were considered but rejected included congestion pricing, expanded HOV enforcement, investments in transit, and incentives for telecommuting.
The New York City Subway system had been experiencing chronic delays and service problems for years, so the MTA implemented a series of management reforms, schedule modifications, and other operational adjustments. But the real problem was an obsolete, century-old signal system. The obvious solution was the complete modernization of the signal system, which was the primary cause of the problem, and which the MTA ultimately did.
Downtown San Francisco was experiencing growing traffic congestion and unreliable bus service, so MUNI added minor bus lane markings, adjusted signal timing, and rolled out enforcement campaigns, all of which had little effect. Finally, MUNI built dedicated bus rapid transit (BRT) facilities on two major downtown streets—Van Ness and Geary—reducing travel time and improving reliability. I like this example because it also highlights the unique capital-operating balance that BRT strikes by providing dedicated routes while avoiding more costly rail infrastructure.
The Theory
I’ve talked enough about capital versus operating in this and several previous posts, so now, I’d like to touch on a tangential but related subject that appeared above in my condo’s lighting problem, as well as the New York Subway example: The idea of “spending down your capital.” Think of it like a homeowner: One rule of thumb is that you should spend between 1% and 4% of your home’s value every year on maintenance and repairs. If you do, your home will hold its value, and you shouldn’t experience many unexpected expenses because you have taken care of it over the years. On the other hand, if you invest minimally or not at all, the home will deteriorate and sooner or later you will experience big capital shocks, like the unexpected need for a new roof or water heater. And even if you don’t, when you go to sell, you will receive lower offers that reflect both the poor condition of the house and, conversely, the money you saved by deferring maintenance along the way. When you avoid spending money on upkeep, you are instead using up the capital value of the asset itself, which is why it is called “spending down your capital.” It may be because you can’t afford to invest, or you’re just cheap, or you know exactly what you are doing and can afford to take the risk. There is nothing wrong with spending down your capital, as long as you recognize it as a conscious decision. I remember the first time I heard the term “spending down your capital,” from an old professor and former Philadelphia city official who once said to me:
“I was up on the 7th floor of City Hall the other day and saw that the linoleum floor we put down when I was Managing Director in the 1970s is still there. Geez, that floor doesn’t owe us anything.”
The Lesson
Understand the differences between capital and operating. Recognize when a discussion is driving towards a capital solution, when an operating solution may be more appropriate and cost-effective, or vice versa. As in the Katy Freeway example, be on guard when a group or individual promotes a major capital project over viable operating solutions. Capital projects can seem sexier, but just because someone wants to be photographed cutting a ribbon doesn’t mean that the project is worth it, or that it will solve the problem. On the other hand, if the infrastructure is the underlying problem, no amount of operational tweaking can correct for it, and kicking the can down the road will surely only make things worse—and make the future solution even more expensive, when you are finally forced to reinvest. Last but not least, don’t spend down your capital unless you are doing it on purpose with a plan—and some emergency funds in your bank account.
“The Katy Expressway is a shitshow”
- Comprehensive-Let232, on Reddit
“Overbuilt? It goes from autobahn speeds to parking lot.”
- Adam Mantz, on TikTok (on the Katy Freeway)
Aunt Bethany: “Is your house on fire, Clark?”
Clark: “No, Aunt Bethany, those are the Christmas lights.”
- Clark Griswold, in National Lampoon’s Christmas Vacation