Co-op & Condo Boards

The stake: Building-wide obligations under the NYC Housing Maintenance Code and Local Law 55.

What is actually at risk for a co-op or condo board?

A duty the board cannot delegate to the person who noticed the problem, and a record that will be read by someone else later.

That is the honest summary, and it is worth sitting with, because it is different from what most boards think they are buying when they call a pest contractor. A board is not purchasing the absence of insects. It is discharging an obligation that attaches to ownership of a multiple dwelling, and it is creating a file. If a resident escalates — to the managing agent, to counsel, to the city — the question asked will not be whether the building sprayed. It will be what the building knew, when it knew it, what it decided, and whether the decision was reasonable given what was in front of it.

Three things flow from that framing, and they are the reason this page exists.

The first is that scope is a compliance question, not a budget question. If the obligation runs to the building and the pathways run through the building, then work confined to the apartments that complained is, on its face, a partial response. It may be a defensible one in a small, isolated incident. It is not defensible as a standing approach in a building generating reports on multiple lines.

The second is that the record is a deliverable in its own right. Field work that happened but was never written down in a legible way is, from the point of view of a dispute, work that did not happen. Boards routinely have the substance and lack the paper.

The third is that the board’s real adversary is time. Pest matters in shared buildings do not stay the same size. A single reported apartment becomes a line, a line becomes a stack, and an unresolved complaint chain turns into correspondence with an attorney’s letterhead on it. The cost of acting building-wide in year one is almost always lower than the cost of the same work in year three plus everything that accrued in between.

Graduate Pest Control has worked on residential buildings across Manhattan, Brooklyn, Queens, Nassau and Suffolk since 1983, and the pattern is consistent enough to be predictive. Boards that treat pest pressure as a building-repair item resolve it. Boards that treat it as a monthly service line manage it indefinitely.

Who actually carries the obligation — the corporation or the shareholder?

Ownership carries it. In a cooperative, that means the cooperative corporation, not the shareholder whose apartment happens to be where the cockroaches were seen.

This is the single most misunderstood point in board pest management, and it produces more wasted argument than any other. The NYC Housing Maintenance Code, at Title 27, Chapter 2, Article 4 of the Administrative Code, places extermination and rodent eradication duties on the owner of a multiple dwelling. A shareholder in a co-op holds shares and a proprietary lease; they are, in the structure of the thing, closer to a tenant than to an owner of real property. The corporation owns the building. The duty attaches there.

A condominium is not identical and the distinction is worth understanding rather than glossing. In a condo, each unit is separately owned real property and the board of managers controls the common elements rather than owning the building outright. Who the “owner” is for the purposes of a given obligation can therefore be a more layered question in a condo than in a co-op, and it may differ depending on where the condition sits. That is exactly the sort of question a board should put to its counsel rather than to its exterminator, and we say so on the first visit. Read the code text itself, and take advice on how it applies to your form of ownership.

What we can be useful about is the practical consequence, which does not change much between the two forms. Telling a resident to hire their own exterminator is not a strategy. Even where the governing documents permit the building to charge the cost back to a unit — and many do, in defined circumstances — the responsibility to see that the condition is actually addressed does not move with the invoice. Cost allocation and legal duty are two separate questions, and boards conflate them constantly.

There is a second layer sitting on top of the code. Local Law 55 of 2018, the Asthma Free Housing Act, requires owners of multiple dwellings to inspect dwelling units for indoor allergen hazards — a category that expressly includes pests alongside mold — and to address what those inspections find. It carries an annual inspection obligation, which is a meaningful shift in character. The older model was reactive: a complaint arrives, a response follows. This one makes the looking itself the duty, on a cycle, whether or not anybody has complained. Again, read the law rather than a summary of it, and take counsel on how the inspection and remediation obligations apply to your building.

For a board, the operational translation of those two instruments is short. You are expected to look, on a schedule. You are expected to act on what you find. You are expected to address the condition rather than the sighting. And you should be able to show all three.

Where does building responsibility end and unit responsibility begin?

At a line drawn in the proprietary lease or the bylaws — and almost every serious dispute a board has about pests is an argument about where that line falls.

The typical allocation gives the corporation the structure and the building systems and gives the shareholder the interior of the apartment: finishes, fixtures, appliances, the contents. Read literally, that seems to settle it. The wall is the building’s; the kitchen cabinets are the shareholder’s. In practice it settles nothing, because the population is almost never on one side of the line. A German cockroach harborage behind a base cabinet sits in the shareholder’s kitchen and is fed by a leaking supply line inside the building’s wall, reached through a penetration in the building’s chase. Whose problem is that?

The argument then runs along entirely predictable grooves. The shareholder says the insects come from the building and from the neighbors. The board says the apartment’s housekeeping invites them. Both positions contain some truth, which is why the dispute persists rather than resolving, and why it usually escalates into a matter of tone rather than fact.

Evidence is what ends it. The diagnostic question is not who is at fault but where the population actually lives, and that is answerable. A population resident in a chase shows up along a vertical line — the same relative location in apartments stacked above and below one another — and it shows up in units with very different housekeeping standards. A population resident in one apartment shows up in that apartment, concentrated around a specific harborage, with the neighboring units clean on inspection and clean on monitoring. Those two pictures look nothing alike once someone competent has looked at more than the complaining unit.

So the sequence we recommend to boards is: inspect before adjudicating. Assess the reported apartment, the units immediately above and below it on the same stack, and the common space behind the wall. Put monitors in and read them. Then decide the cost question, with the managing agent and with counsel, on a finding rather than on an assertion. A board that does this once discovers something clarifying: the answer is frequently that both are true — there is a chase population and there is a contributing condition inside the unit — and a program that addresses only one half of that will fail regardless of who was right.

One further point deserves stating plainly, because it saves boards a great deal of grief. Bed bugs do not respect this line at all, and treating them as a housekeeping matter is both factually wrong and a reliable way to make residents conceal reports. A resident who fears blame, cost or stigma delays reporting, and delay is the only variable that genuinely determines how expensive a bed bug problem becomes in a shared building.

Why does apartment-by-apartment treatment fail in a shared building?

Because the apartment is not the boundary that the animals recognize. The building is.

It helps to look at what a residential building actually is, structurally, from the point of view of something small that wants to move around inside it.

Plumbing chases. Kitchens and bathrooms are stacked vertically so that a single set of supply and waste lines can serve every floor. The void those lines occupy runs from the cellar to the roof. Unless it was firestopped and sealed at each floor — and in pre-war stock it very often was not, and in post-war stock it was frequently done badly or opened later by a trade — it is a continuous shaft joining every kitchen on the line. A cockroach population living in that shaft is not resident in any apartment. It is resident in the building’s plumbing and it forages into whichever unit is currently the most rewarding.

Steam and heating risers. Older buildings run risers through floor slabs, and the penetration is usually oversized to allow for expansion. Some are sleeved. Very few are sealed at the annulus, and the gap around a riser passing through a slab is a purpose-built vertical route. Radiator boxes and the void beneath them are among the most productive harborages in pre-war apartments and among the least frequently inspected.

Electrical conduit and low-voltage runs. Every generation of cable pulled through the building since it was built has cut its own path. Cockroaches move inside conduit and behind switch and outlet boxes, which is why a treatment confined to a kitchen and a bathroom leaves a population intact in the wall two rooms away.

Party walls and joist pockets. In attached buildings — brownstones, row houses, and the countless converted structures across Brooklyn and Manhattan — the floor joists of two adjoining buildings often bear into the same shared wall from opposite sides, leaving open pockets that connect one structure to its neighbor. A rodent or bed bug problem in one house arriving next door within weeks is not a coincidence and is not a failure of the treatment. It is the joist pockets.

Compactor chutes and refuse rooms. This is the one boards underestimate most. A refuse chute is a vertical duct with a food residue film on its interior surface, opening onto every floor through a door whose gasket wears out, terminating in a room with a floor drain and a staging area. It supplies food, moisture, warmth and vertical access in a single assembly. A building can be immaculate on every landing and still support a substantial population in that one shaft.

Dumbwaiter shafts. Many older buildings have them, decommissioned decades ago and closed off with whatever was to hand. Closed off is not the same as sealed, and an abandoned shaft is an undisturbed vertical corridor running the height of the building.

The cellar and everything below it. Pipe sleeves entering from the street, sidewalk vaults and hatches, elevator pits, and the drainage connection. Norway rats characteristically arrive from below through a failed house trap or a cracked cellar line rather than walking in the front door, and a building that responds to rat sightings with exterior baiting alone can run that cycle for years without touching the source.

Now put treatment into that picture. Service one apartment on a line in isolation and the count inside that apartment falls while the chase population is untouched and a share of it redistributes. Within weeks the units directly above and below it report. The building experiences a complaint pattern that appears to spread despite continuous service, and the file — which someone may eventually read critically — records exactly that.

The alternative is to make the network the client. Assess a block of units around every report rather than the reporting unit alone. Seal the pipe penetrations floor by floor through the whole stack. Firestop and close the chases. Close the joist pockets at party walls. Repair the chute doors and rebuild the refuse process. Resolve the drainage below. The physical detail of this work is set out under structural exclusion, and the species-specific approaches sit under cockroach control, rodent control and bed bug treatment.

How is a building-wide program actually phased?

In four phases across roughly two budget cycles, and the phases are sequential for reasons that are physical rather than administrative.

Phase one — survey. A full building assessment before any recurring schedule is agreed. Every common area, the cellar and mechanical spaces, the refuse system end to end from chute door to curb, the roof and bulkhead, the exterior at grade, and a meaningful sample of apartments selected by stack rather than by who complained. What comes out of it is a baseline: what is present, where, how heavily, by what route, and which conditions are sustaining it. Setting a service frequency before this exists is guesswork dressed as a contract, and it is how buildings end up paying for monthly visits that were never sized to anything.

For a board, this phase has a second function. It produces the exhibit. A written baseline is what turns “we think we have a roach problem” into a document the board can act on, minute, and hand to the next board.

Phase two — reduction. Bringing the existing population down, with methods and placements constrained by the fact that people live here. This phase is intensive and it is finite. It is not the program. It is the thing that has to happen before the structural work can proceed, because sealing a building that has an active population inside it removes the animals’ route out rather than the animals — and in a rodent job that means the next thing the board hears about is an odor inside a wall.

Phase three — correction. The structural and sanitation work the survey identified: sealing penetrations through the stacks, closing chases and joist pockets, chute door gaskets and chute cleaning, door sweeps on cellar and service doors, drainage repairs, storage practice in the cellar, and the refuse handling interval. This is the phase that decides whether the result is durable, and it is the phase most likely to stall, because it lands in a different budget line and involves trades that are not the pest contractor. Our job is to specify it precisely enough that a carpenter, a plumber or a mason can price it, and then to verify it once it has been done.

Phase four — monitoring and verification. A recurring cycle sized to the building’s actual risk rather than to a standard contract: monitors read and logged, the corrective register worked through, conditions re-checked, and reporting produced on a rhythm the board’s own meeting schedule can absorb. If phase three was done properly, the frequency should be able to come down after the first year. If it never comes down, that is the diagnostic — the correction did not happen, or it did not hold.

Phasing across stacks rather than across the whole building at once is usually the practical route for an occupied co-op. It limits disruption to one line at a time, it lets the board fund the work in tranches, and it produces an early, visible result on the worst stack that makes the rest of the program very much easier to approve.

How does a board get building-wide work approved?

By presenting it as a repair to the building rather than as a bigger pest contract, and by giving the undecided directors something they can defend to a shareholder in the elevator.

The politics here are real and they are not cynical. A board is being asked to spend other people’s money on a problem most of those people have not personally experienced. The shareholder on the fourth floor of a quiet line has had no cockroaches, has no complaint, and is being asked to fund work in a stack she has never entered. She will ask why. A director who cannot answer that question in two sentences will vote no, or will vote to defer, which amounts to the same thing.

Several things make the case land.

Lead with the survey, not the proposal. A board approves work it understands. The finding — this line has a chase population, here is where it is living, here is why the last three years of treatment did not touch it — does the persuading. The scope of work follows from it. A proposal presented without a baseline reads as a contractor upselling, and directors are right to be skeptical of that.

Name the alternative honestly, including its end date. The alternative to correction is not zero. It is the current spend, continuing, with no terminal point, plus the complaint volume, plus the staff time, plus whatever the escalation eventually costs. Unit-by-unit response is a permanent operating expense. Correction is a finite one. Put both on the page and let the numbers be the building’s own, not ours.

Address the fairness objection directly. The reason the quiet line should fund work on the affected line is that the chases connect them, that pest pressure migrates along stacks, and that the building’s obligation is a building obligation regardless of which apartment currently has the problem. That is also the answer to the shareholder in the elevator, and it has the advantage of being true.

Bring the managing agent in early. The agent will be operating this — notices, access, coordination with the super, invoicing — and an agent who first sees the scope at the board meeting will find reasons it cannot be done. An agent who helped shape it will carry it.

Give the board a decision it can minute. Directors are exposed when they act on a vague recommendation. They are much better placed when the minutes record a written finding, a specified scope, a phased plan and an approval. That record is also, later, the evidence that the board acted reasonably on professional advice.

Sequence it against the rest of the capital plan. Pest correction competes with façade work, elevators, roofs and boilers, and it will lose every time if it is presented as an equal claimant. It usually does not have to be. Much of the work is small-trade labor that can be phased, and some of it — sealing chase penetrations, closing joist pockets — is dramatically cheaper if it is scheduled alongside work that is opening those walls anyway.

Should the building assess for it or absorb it?

That depends on what the work is, and the useful distinction is between repair and service rather than between large and small.

Structural correction — sealing penetrations through a stack, rebuilding a refuse room, replacing chute doors, repairing a cellar drain line — is work on the common elements. It has a defined scope, an end point and a durable result, and it tends to be treated in the same way as other building repairs. Recurring monitoring, inspection and response is an operating cost with no end point, and it belongs in the operating budget alongside the other things the building pays for every year.

Which of those gets funded from reserves, from a line item, from a special assessment or from an increase in maintenance is a question for the board with its managing agent and its accountant, informed by the governing documents. It is genuinely not a question for a pest contractor, and any contractor who answers it confidently is overreaching. What we can do is scope the work in a way that makes the distinction clean, so that the board is not trying to split a single undifferentiated figure into categories after the fact.

Two practical notes are worth passing on. First, a scope that itemises openings and locations rather than presenting one number for “building-wide exclusion” is far easier to phase, fund and defend, and it lets the board decline part of it knowingly rather than deferring all of it by default. Second, the charge-back question — whether a cost can be allocated to a particular unit — should be settled as policy before it arises rather than negotiated case by case under pressure. Boards that improvise this create precedent they later regret.

What records does the board actually need?

Six things, and buildings usually have the first two and are missing the rest.

The baseline survey. The written assessment of the building at the point the program started, with findings by location. Everything afterwards is measured against it, and without it there is no way to demonstrate improvement to anyone.

The service log. Dates, who attended, which areas were covered, what was found and what was done. Specific enough to be worth reading later: a note recording two captures in a named monitor on a named floor is evidence, and a note recording that the building was serviced is not.

The unit-level inspection and access record. Which apartments were notified, on what dates, which were inspected, which were not, and why. Refusals recorded individually with the notice date, the scheduled window and the re-attempt. This is the record that boards most often lack and the one that matters most when a matter is disputed, because the difference between a building that could not gain entry and a building that never tried is entirely a documentary difference.

The corrective-action register. Every structural or sanitation finding with a named owner, a target date and a closure record. The crucial detail is that most of these items are not the pest contractor’s to close. Sealing a chase is carpentry. Repointing a cellar wall is masonry. A cracked drain line is a plumber. A chute door gasket is building maintenance. Findings without an assignment and a date do not become work; they become an accumulating record of a known and unaddressed condition, which is the worst possible file to be holding.

The trend. Counts and complaint volume over time, by area or by stack, so the direction of travel is visible without anyone reading a year of field notes. This is what the board needs in order to make a funding decision and what an outside reader uses to judge whether the program functions.

The board’s own decision record. Minutes showing what the board was told, when, and what it resolved to do. Notices sent to residents. The approved scope. This half of the file is the board’s to keep, not the contractor’s, and it is regularly the half that is missing.

Reporting should be written for the person who has to act on it. The managing agent needs operational detail and the list of what is still open. The board needs the trend, the risk position, and a clear statement of what it is being asked to approve and why. One document written to serve both audiences typically serves neither, and the usual consequence is a necessary repair sitting unfunded for a full budget cycle.

What proof should a board expect to receive?

Enough to satisfy a director who was not there, a shareholder who is skeptical, and a reader who is hostile — and it should arrive routinely rather than on request.

At minimum, a board should expect photographic evidence of conditions found and of the same locations after correction. Before-and-after images of a sealed penetration are unglamorous and they are the single most persuasive item in any board packet, because they show a physical change to the building rather than an activity that was performed.

It should expect a monitor map that matches the building as it currently stands, and monitor readings recorded by location rather than summarized. It should expect findings written in language a layperson can follow, since the people making the decisions are not pest professionals and should not need to be. It should expect the awkward findings to be included — the refused unit, the condition the building has not funded, the source that turns out to be a neighboring property — because a sanitised report is worthless at precisely the moment it is needed.

And it should expect a plain statement of what has not been resolved and what it would take. A program that only ever reports success is not reporting.

We do not publish case studies of other buildings, and a board should be a little wary of contractors who do. What is portable between buildings is method, not outcome; every structure has its own construction, its own refuse arrangement and its own history of alterations. What a board can reasonably ask for is the standard of evidence it will receive about its own building, which is a fairer test and a more useful one.

What goes wrong, and why?

The failure modes are consistent enough that a board can check itself against them.

Treating only the units that complain. The most common failure and the source of most of the others. It reduces counts where someone was looking and redistributes the rest, and the file it produces reads as a spreading problem under continuous service.

Sealing before reducing. The second most common. It closes the exits, concentrates the population inside voids and produces a worse problem, sometimes with an odor attached to it.

A corrective register that nobody funds. The survey is done, the findings are written, the items have no owner and no date, and eighteen months later the same conditions are re-recorded verbatim. At that point the file has become documentary evidence that the building knew.

Alterations that reopen the work. A shareholder renovation opens a kitchen wall, the chase is exposed and left unsealed, and a stack that had been quiet for two years starts reporting again. Boards can close this loop cheaply by writing penetration sealing and a post-work inspection into the alteration agreement.

Losing the institutional memory. Boards turn over annually and supers move on. A program that lives in one director’s head or one super’s routine dies with them. The defense is documentary: the baseline, the register and the trend belong in the building’s permanent records, not in an email chain.

Refuse discipline regressing. Nearly every building’s program drifts here first, because refuse is a daily practice and practices decay. The interval between a bag leaving an apartment and the material leaving the property is the number to watch, along with the state of the compactor room floor and whether its drain holds a seal.

Buying on price alone. A low per-visit price buys a fixed number of minutes on site, which buys attendance rather than diagnosis. The building still has the same chases. It now also has a service log that documents visits and finds nothing, which is a genuinely poor document to be holding.

Expecting resolution in one season. Population reduction is fast. Structural correction runs on a construction timeline and a budget cycle. A board that expects both within a quarter concludes the program has failed at exactly the point it is starting to work.

How should a board start?

With a survey and a written finding, before any contract is discussed.

Bring the history if it exists: previous service records, the complaint log, dates of any renovation or capital work, which lines and stacks generate reports, and anything the super has noticed but never formally reported — that last category is frequently the most useful thing in the room. We would rather walk the building, tell the board what is actually happening inside it and what correcting it involves, and let the board decide from there, including deciding that the honest answer is narrower than feared.

Managing agents running several buildings will find the portfolio version of this on our property management page, and the wider framework for buildings under compliance obligations is on the commercial pest management hub. The sealing discipline that underpins all of it is described in full under structural exclusion, and the pest-specific detail for the three problems boards deal with most sits under cockroach control, rodent control and bed bug treatment. Towns and neighborhoods we cover are listed under locations, and Ryan Katz sets out how the diagnostic side of this work is approached.

To begin, contact us with the building — its age, its construction, how the refuse is handled, and which apartments keep coming back.