The Large Local Employer That Closes
Why this matters
A plant, a hospital wing, a distribution centre or a regional call centre announces it is closing, and a shop within driving distance loses four different things on four different clocks. Most owners track only the first, because it is the only one with a name on an invoice. The other three arrive over the following quarters, look like a general slowdown, and get filed under bad luck. What makes this shock worth its own card is that its footprint is knowable in advance: you can name the ZIP codes, count your own customers inside them, and size your exposure in an afternoon out of records you already have. Almost nobody does, and that is the whole opportunity.
The four channels, and the order they arrive in
The account itself. If the employer was a commercial customer, that revenue stops on a date you can read off the announcement. This is the only channel most shops track, and in a shop with a broad residential base it is usually the smallest of the four.
The household income shock. Several hundred to several thousand paychecks stop in a tight geography. Household spending does not fall evenly across your service list: it collapses in the deferrable half and holds in the non-deferrable half, which is the split Deferrable Versus Non-Deferrable Work and What a Downturn Touches derives in full. Practically, the calls keep coming and they change character - no-heat and no-water stay, the recommended replacement becomes a repair, and the maintenance agreement does not renew.
The housing effect. People leave for work elsewhere. Listings rise, days on market lengthen, and transaction-linked work - inspection repairs, pre-sale fixes, the new owner's first-year upgrades - falls with the number of closings rather than with prices. Values soften later and matter less to you than volume does.
The second-order local economy. The diner, the two auto shops, the staffing agency and the daycare that existed on that payroll all thin out. If you carry small commercial accounts, this is where your commercial side erodes a year after the closure, long after everyone has stopped talking about it.
The phases, and why each wants a different response
The channels do not arrive together. They run roughly in this order, and the response that is right in one phase is wrong in the next.
| Phase | Typical timing | What you see | What the phase asks for |
|---|---|---|---|
| Announcement | Notice filed, work continues | Nothing in revenue. Discretionary quotes stall | Audit exposure, arrange credit while you still look strong |
| Wind-down | Between notice and last shift | Approval rates fall before call volume does | Hold price, stop adding fixed cost |
| Separation | The quarter after closure | Agreements lapse, aging stretches, mix shifts to repair | Tighten collections, re-weight marketing |
| Out-migration | Two to six quarters after | Transaction work falls, some accounts vanish without notice | Deliberate service-area work, second-market entry |
| Reset | Beyond that | A smaller, stable local base | Rebuild agreement base at the new size |
The expensive error is treating the separation phase as the whole event and cutting to fit it, then having nothing left to absorb the out-migration phase eighteen months later.
What tells you before your bank balance does
The notice filing itself, which is public. The federal WARN Act (29 U.S.C. 2101) requires employers with 100 or more employees to give 60 days' written notice of a covered plant closing or mass layoff, and state labor departments publish the filings they receive. That is the federal floor and several states layer on top of it: New York's state act reaches employers at 50 employees and requires 90 days, and a number of states add severance or longer windows. Check your own state's list, subscribe to it, and read it monthly. It costs nothing and it is the only indicator in this card that arrives before anything happens.
Your own approval rate on deferrable quotes, by ZIP. This moves in the wind-down phase, weeks before call volume does, because a household that expects a layoff defers the replacement and still calls when the water heater dies. Track quote-to-close separately for the affected ZIP codes and you will see the break before the revenue line does.
Agreement renewal rate in those ZIP codes. Maintenance agreements are the most reliably deferred recurring product a shop sells, and renewal is a decision a household makes deliberately.
Your aging, by ZIP. Customers paying slower in one geography is not a collections problem, it is a market reading. See related: The Aging Buckets and What Each One Actually Costs.
Sizing your own exposure: a worked count
Pull three counts out of the system you already run. This example uses one shop's real shape; run yours the same way.
The shop has 640 active residential customers over a trailing 24 months. The three ZIP codes that ring the plant hold 214 of them, so 214 of 640 is 33 percent of the residential base by count.
Revenue splits 78 percent residential, 22 percent commercial, and the plant itself was one commercial account carrying 9 percent of total revenue. So the account loss - the only channel most shops count - is 9 points, and the household channel sits behind a third of the residential base, which is 33 percent of 78 percent, or 26 percent of total revenue exposed to the income shock, assuming those households spend like the rest of the book, which the agreement concentration below suggests understates it. The indirect channel is nearly three times the direct one.
Now the part that changes the plan. Of 96 active maintenance agreements, 41 are in those three ZIP codes: 41 of 96 is 43 percent, ten points higher than the 33 percent those ZIP codes represent in the customer count. The agreement base is more concentrated than the customer base, because the plant's wages were better than the area's, and better-paid households buy agreements. Agreements are deferrable recurring revenue at high margin, so the shop's single most profitable product line is also its most exposed.
That finding flips one decision. Before the count, the owner was going to protect the commercial side and let residential marketing ride. After it, the priority is agreement renewal in those three ZIP codes, starting in the wind-down phase while the households still have income - an early renewal offer, at the current term, taken before the last shift rather than after it.
The levers that actually exist at this scale
You cannot change the closure. Four things are genuinely available, and only the first has a deadline:
Renew and re-contract early, before separation. Anything with a renewal date after the last shift is worth pulling forward while the household is still earning. This is the highest-return move in the list and it has a hard deadline.
Broaden the service area on purpose, not in panic. Deliberate means picking one adjacent area, sizing the drive time against your own scheduling, and committing marketing to it for a stated period. Panic means taking every call anywhere, which raises unbilled drive time and quietly lowers your revenue per hour of capacity in the exact quarter you can least afford it.
Shift the offer toward what still sells. Repair over replace, maintenance over upgrade, staged work over whole-system work. This is a mix change, not a price cut, and cutting price here is the wrong lever - see the Pricing shelf.
Reprice nothing, but re-sequence everything. Your rate is not the problem; your calendar is. Emergency and non-deferrable work is holding, so protect capacity for it rather than filling the book with discounted deferrable jobs that will not repeat.
Collections has to move first, and nobody wants to
This is the unpopular one, and it belongs in the wind-down phase rather than after it. A household that is about to lose income pays its trades last, and a small commercial account that loses the plant as its own customer will stretch you before it tells you. Two changes, both made before the closure date: shorten terms for new work in the affected ZIP codes, and take deposits on material-heavy jobs there. Do it as a standing policy for that geography, dated, rather than as a judgment call per customer, because a per-customer version becomes an argument with the people you most want to keep.
The failure mode is waiting for the aging report to prove it. By the time a 60-day bucket in those ZIP codes is visibly fatter, the money has already been spent on something else.
When an employer arrives instead
The mirror case is real and it is oversold. A new facility is announced, everyone locally starts planning for growth, and the ramp takes far longer than the press release implies: construction, then hiring, then relocation, then the household actually buying a house and calling a trade. Two to three years from announcement to a meaningful change in your residential call mix is ordinary.
What that means in practice is that the work is genuinely there and the sequencing of your response is what decides whether you profit from it. Adding fixed cost - a truck, a lease, a salaried dispatcher - on the announcement rather than on the demand is the classic way a shop turns good news into the overhead that kills it in the next downturn. See related: The Boom That Is More Dangerous Than the Slump.
Checking that you read your own numbers right
Print each figure next to the criterion it is being checked against, rather than asserting the reading is sound:
- The count-to-revenue conversion names its assumption. Residential exposure printed above: 33 percent of the base by count, which becomes 26 percent of total revenue only if those households spend like the rest of the book. The agreement concentration says they were the better-paying ones, so treat 26 as a floor rather than an estimate. A count reported without that step says nothing about cash, and a revenue share reported without naming the step is a count wearing a revenue label.
- The agreement concentration is compared like with like. Agreements in the affected ZIP codes: 43 percent. Customers in the same ZIP codes: 33 percent. Both are counts over the same geography, so the ten-point gap is real. Comparing an agreement count against a revenue share instead would have produced a number that means nothing.
- The direct channel is sized against the indirect one. Account loss 9 percent of revenue, household channel 26 percent. If your direct figure is the larger of the two, you are either genuinely commercial-weighted or you have not counted the residential channel yet.
- The leading indicator has a date on it. You either have the state WARN list subscribed or you do not. There is no partial credit on this one.
References
- U.S. Department of Labor, Worker Adjustment and Retraining Notification Act (29 U.S.C. 2101), employer coverage and notice period; state labor department WARN notice listings for state-level thresholds
- U.S. Census Bureau, County Business Patterns, for local employment concentration by industry
- See related: Deferrable Versus Non-Deferrable Work and What a Downturn Touches (owns the deferrable split), The Aging Buckets and What Each One Actually Costs, Reading an Aging Report: Who to Chase First
- See related: The Boom That Is More Dangerous Than the Slump, Supplier Concentration and What a Top Five List Hides (the same concentration test applied to the buy side)