Infrastructure Is What We Notice Only After It Fails

Some of the most important things government owns are also the things most people rarely think about.
A water main beneath the street.
A bridge carrying thousands of people every day.
A network switch connecting employees to critical applications.
A heating system working through January.
A roof keeping water outside.
A retaining wall holding up someone’s property.
A staircase connecting two streets on a steep hill.
A catch basin that allows stormwater to leave the roadway.
When these things work, almost nobody celebrates them.
When they stop working, they can become the most important issue in the organization.
I learned that lesson early in my career.
Around 2005 and 2006, I worked as a capital budget analyst at the New York City Department of Transportation. It was one of my earliest opportunities to see infrastructure not as an abstract capital-budget category, but as an enormous collection of physical assets somebody ultimately had to own, inspect, repair and pay for.
After more than two decades working around government budgets, technology, procurement and capital planning—and years of owning and maintaining real estate—I have developed a simple view of infrastructure:
Maintenance is usually cheaper than replacement, and the best time to address deterioration is before it becomes a capital project.
Sometimes the Hardest Question Is: Whose Wall Is It?
One of the most interesting infrastructure questions I encountered during my DOT years involved retaining walls.
In May 2005, a massive retaining wall beneath the Castle Village cooperative complex in Upper Manhattan collapsed onto the Henry Hudson Parkway near West 181st Street. The City had to respond immediately to stabilize the site and restore one of Manhattan’s major transportation arteries.
But incidents like that raise a deceptively difficult question:
Who is responsible for the infrastructure?
If a retaining wall supports private property but sits alongside a public highway, whose obligation is it to maintain?
The property owner?
The City?
Some combination?
And even if the legal answer says the responsibility belongs to a private entity, government doesn’t necessarily have the luxury of ignoring the asset if its failure can close a highway or endanger the public.
That distinction between legal responsibility and practical responsibility has stayed with me.
I remember similar questions involving Rockefeller University and the structures associated with its campus over the FDR Drive.
Rockefeller University today describes part of its campus as being suspended over the FDR Drive.
One can reasonably say: it’s university infrastructure, so the university should maintain it.
Fine.
But what does New York City do if infrastructure above one of its major highways isn’t adequately maintained?
Government still owns the risk beneath it.
That’s what makes infrastructure management so complicated.
Sometimes the government doesn’t own the asset but cannot afford the consequences of its failure.
And Then There Are Step Streets
New York City’s geography creates infrastructure most people elsewhere probably never think about.
Step streets are a perfect example.
In steep parts of the city, particularly in Upper Manhattan and the Bronx, enormous public staircases essentially function as streets for pedestrians, connecting roadways at different elevations.
They’re fascinating pieces of urban infrastructure.
They also have to be maintained.
They have to be repaired.
And in winter, somebody has to remove the snow.
You can’t exactly send a conventional snowplow up a staircase.
NYC DOT still specifically recruits emergency snow shovelers to help clear locations including step streets, and the City’s snow-response system relies on supplemental manual labor for pedestrian overpasses and other locations that ordinary plows can’t effectively service.
That illustrates a larger reality:
New York City owns more infrastructure than its permanent workforce can possibly touch every day.
That isn’t necessarily an indictment of the workforce.
It’s a question of scale.
Thousands of streets. More than 12,000 miles of sidewalks. Bridges. Step streets. Overpasses. Retaining walls. Signs. Signals. Buildings. Drainage systems. Technology. Parks. Public spaces.
There is finite maintenance bandwidth.
That’s Where Contracting Can Make Sense
I work in procurement, so perhaps unsurprisingly I see contracting as one legitimate part of the solution.
Government doesn’t necessarily need a permanent employee available for every specialized or seasonal maintenance requirement.
Contractors can provide surge capacity.
They can supply specialized equipment.
They can perform work requiring skills government needs intermittently rather than every day.
New York City already uses outside partners for functions including snow and ice removal, graffiti removal and maintenance of certain public spaces.
The question shouldn’t be ideological—public employee or contractor?
It should be operational.
What arrangement gives taxpayers reliable service at a reasonable lifecycle cost?
Sometimes that’s City employees.
Sometimes it’s a contractor.
Sometimes it’s both.
And occasionally, it’s the person standing nearby who decides to help.
My Wife Couldn’t Understand Why I Was Cleaning a Catch Basin During a Nor’easter
During a recent Nor’easter, I noticed a catch basin near me becoming obstructed by leaves and sticks.
So I cleared the debris.
My wife couldn’t quite understand why I was outside doing this.
It wasn’t my job.
Nobody had asked me to do it.
I wasn’t getting paid.
But I looked at it differently.
I could see the problem.
I could safely do something simple that helped prevent water from accumulating.
So I did it.
That’s part of how I think about community service.
Not everything needs a work order.
Not every useful act requires compensation.
And not every responsibility for the condition of the place where we live belongs exclusively to government.
There are obvious limits to this. Residents shouldn’t climb bridges, enter roadways, approach electrical hazards or attempt infrastructure work requiring professionals.
But there is still room for civic responsibility.
President John F. Kennedy captured that idea much more eloquently in his inaugural address:
“Ask not what your country can do for you—ask what you can do for your country.”
Sometimes what you can do for your country is historic.
Sometimes it’s clearing leaves away from a catch basin.
The scale is different.
The civic instinct is the same.
Not Every Infrastructure Problem Needs a Capital Project
Government loves capital projects.
A new building gets a ribbon cutting.
A reconstructed bridge gets an announcement.
A new technology system gets a launch.
Nobody holds a press conference because somebody painted a railing before it rusted.
But perhaps we should pay more attention to exactly that kind of work.
Paint the railing.
Paint the bridge.
And I don’t mean only the George Washington Bridge or Brooklyn Bridge. Think about the countless smaller overpasses and structures residents pass every day.
Remove the graffiti.
Patch the wall.
Repair the roof.
Seal the water intrusion.
Clean the catch basin.
Do it early.
The broken-windows concept is usually discussed in the context of public safety and has a complicated history. But there is a broader asset-management principle that doesn’t require embracing any particular policing philosophy:
Visible deterioration shouldn’t simply be allowed to accumulate.
Graffiti, broken fixtures, peeling paint and neglected public spaces communicate that nobody is paying attention.
Government shouldn’t wait for routine maintenance to become a capital project.
NYC’s Own Accounting Rules Recognize the Difference
Working in New York City government also makes the distinction between maintenance and capital improvement very real.
The Comptroller’s Directive 10 governs what may appropriately be charged to the Capital Projects Fund, while Directive 30 addresses the City’s accounting and control of capital assets.
Those rules matter.
But they also demonstrate something important:
Something doesn’t have to qualify as a capital project to be important.
Sometimes expense-funded maintenance is precisely what protects the capital asset taxpayers already purchased.
Paint the railing today so we don’t replace it tomorrow.
Maintain the roof so we can defer replacing it.
Seal the building envelope before water damages what’s behind it.
Maintain technology so replacement decisions can be made strategically instead of during an outage.
“Can we capitalize this?” and “What is the most economical way to preserve this asset?” are different questions.
Government needs to ask both.
My Own Roofs Have Taught Me the Same Lesson
I intentionally defer roof replacement on my own home and rental properties when repairs can reasonably extend a roof’s useful life.
If a localized repair can solve the problem, I’d rather make the repair than automatically spend tens of thousands of dollars replacing an otherwise functional roof.
That preserves working capital.
But eventually repairs stop making economic sense.
When I replace a roof, I increasingly favor metal where appropriate. It may cost more upfront, but I’m thinking about decades rather than this year’s invoice.
Doing a major job twice over 40 years can be more expensive than spending more to do it once.
But even that isn’t an absolute rule.
Sometimes preserving cash today is more valuable than buying the longest-lasting option.
That’s capital allocation.
Government faces the same tradeoff.
The Hardest Maintenance Job May Be the One Nobody Can Reach
I have another small example at one of my properties.
Some vinyl siding beneath a third-floor window needs attention.
If it were six feet above the ground, it would probably already be fixed.
Instead, it’s three stories up, near electrical wires, and may require a lift to reach safely.
Suddenly an inexpensive piece of siding becomes an expensive maintenance project.
I know what may happen.
Unless I find a practical way to address it, the work may eventually wait until the problem becomes significant enough to justify the mobilization expense.
Government encounters that problem everywhere.
The paint may be inexpensive.
Closing traffic lanes isn’t.
The replacement part may be cheap.
Getting workers safely to it isn’t.
That is why maintainability should be considered when infrastructure is originally designed.
Don’t Build What You Can’t Maintain
This brings me to what may be the most important infrastructure lesson.
Government has finite maintenance bandwidth.
Every new building, vehicle, computer, park feature, camera, bridge component or technology system creates another obligation.
Someone has to inspect it.
Clean it.
Paint it.
Patch it.
Secure it.
Upgrade it.
Repair it.
Eventually replace it.
We tend to celebrate acquisition.
Maintenance comes later.
Before government builds something new, we should therefore ask not only:
Can we afford to build it?
We should ask:
Can we afford to own it?
That includes money, but also people, equipment, contracts, institutional knowledge and time.
If government continually expands infrastructure without expanding its ability to maintain it, we shouldn’t be surprised when the maintenance backlog grows.
Sometimes the responsible infrastructure decision may be to rehabilitate what we already own rather than add something else.
Infrastructure Is Ultimately a Shared Responsibility
Government has the primary responsibility for public infrastructure.
I don’t want residents replacing guardrails or repairing bridges.
But maintaining a city as complicated as New York ultimately requires more than capital appropriations.
It requires government employees.
Contractors.
Property owners.
Universities.
Co-ops.
Community organizations.
And, occasionally, residents willing to clear the leaves away from a catch basin because they’re already standing there and can help.
Good infrastructure management starts with knowing what we own and who is responsible for it.
Then maintaining it while maintenance is still economical.
Repairing it while repair remains sensible.
Replacing it when lifecycle economics—not neglect—make replacement the better decision.
And resisting the temptation to build more than we have the institutional capacity to maintain.
Capital projects will always be necessary.
But a capital project shouldn’t be our maintenance strategy.
Sometimes fiscal responsibility is a new bridge.
Sometimes it’s a 40-year roof.
Sometimes it’s black paint on a railing before rust takes over.
And sometimes it’s somebody noticing that a drain is clogged during a storm and deciding that, while it may not technically be his responsibility, he can do something about it.
Infrastructure works best when somebody takes responsibility before failure forces somebody to.






Join the discussion
Members can read and add comments on every article.