How much should an HOA have in reserves?

There is no universal dollar figure, because "enough" depends entirely on what your community owns and how old it is. But the question has a real answer for your association, and it's computable: enough that every component can be replaced on schedule without a special assessment. That's measured not in dollars but in percent funded — and, since 2026, condo communities also face one hard number from the mortgage market, which is about to get harder.

The wrong question and the right one

"How much money should be in the account?" is the wrong question, because the same balance means opposite things in different communities. $300,000 is luxurious for a townhome community with little shared property and alarming for a 40-year-old mid-rise with an original roof and two elevators. The balance only acquires meaning against what the community owns — which is why the reserve study and the reserve fund have to be read together.

The right question is: what fraction of our components' consumed life have we actually saved? That's percent funded — your balance divided by the fully funded balance your reserve study calculates. It's the number that lets a brand-new community and a 40-year-old one be judged on the same scale.

A community at 70%+ funds its projects from reserves and keeps dues predictable. Below 30%, special assessments are a matter of when, not if. In between is the uncomfortable majority: fine in good years, one surprise away from a shortfall.

The one hard number: what lenders now require

For condominium communities, the abstract question acquired a concrete floor. Fannie Mae's eligibility rules require a condo project's budget to allocate at least 10% of its annual assessment income to replacement reserves (Selling Guide B4-2.2-02, verified September 2026). A community that budgets less can render its units ineligible for the most common mortgages — which buyers' lenders check when your owners try to sell.

That floor is rising. Under Lender Letter LL-2026-03 (March 2026), the minimum climbs to 15% for loan applications dated on or after January 4, 2027 (CAI's summary of the changes, verified September 2026). For a community collecting $500,000 a year in assessments, that's the reserve line moving from $50,000 to $75,000 — a jump many boards will meet with a dues increase.

But note the exemption, because it's the whole point: the 15% allocation is not required if the association has a reserve study conducted or updated within the last three years and is funding at the study's highest recommended level (baseline funding explicitly doesn't qualify). In other words, the lending market's answer to "how much is enough" is the same as the reserve profession's: whatever your current study says, actually funded. The blunt percentage is the penalty for not knowing your real number.

What "enough" costs per month

The real number comes from component arithmetic, and it's worth seeing the scale. Take a 100-unit community whose study inventories $6,000,000 in components with an average useful life of 30 years. Straight-line, replacement alone consumes $6,000,000 ÷ 30 = $200,000 a year — $167 per unit per month before a dollar of operating costs, and before inflation widens it. A community of the same size owning only a monument sign and a mailbox kiosk might genuinely need $10 per unit. Neither number is derivable from a rule of thumb; both fall straight out of a component inventory.

That's also why the reserve share of dues varies so widely — commonly somewhere between a quarter and a third of the total, but legitimately far outside that range in either direction depending on what the community owns.

Why rules of thumb fail

Every rule of thumb in circulation fails for the same reason: it ignores the building's age and inventory.

• "10% of the budget" treats a new community and an old one identically. The new one is banking years of cushion; the old one is falling further behind every year the percentage stands still. (Fannie Mae's own rules acknowledge this — the study-based path exists precisely because the flat percentage is crude.)
• "$X per door" imports another community's inventory into yours. Elevators, private roads, and a heated pool don't cost what siding and a fence cost.
• "100% funded or bust" overshoots in the other direction. Full funding is the most conservative goal, but a community can be financially sound below it if its funding plan keeps the projection safely above zero through the big replacement years. What matters is the trajectory, not a badge.

Some states add their own floor — California mandates the planning (a study every three years, Civ. Code §5550), and Florida now prohibits waiving structural reserves for taller condo buildings (SB 4-D) — both verified September 2026. But no rule of thumb, statutory or otherwise, replaces the inventory-based answer.

Getting from here to enough

If your percent funded is low, the path is a plan, not a panic. Boards close the gap with some combination of stepped contribution increases, re-sequencing projects by risk, and phasing large replacements — modeled over the projection so owners can see the trade-offs, because the cost of standing still compounds faster than any of the alternatives.

Rough out the scale before the modeling: at 40% funded against an $800,000 fully funded balance, the gap is $480,000. Closed over eight years, that's on the order of $60,000 a year on top of the normal contribution — and in practice somewhat more, since the fully funded balance keeps growing as components age. Your preparer's projection turns that order-of-magnitude into a real schedule, but even the rough number tells you which conversation you're having: a dues adjustment, or a phased plan with harder choices. Put the target in the annual budget as a stated goal ("reach 70% funded by 2032") and report progress against it every year: a board that names its number builds more trust than one that hopes nobody asks.

One more audience should ask this question: buyers. If you're purchasing into a community rather than governing one, ask the resale package for the current reserve study and the percent funded figure. A low number isn't necessarily a dealbreaker — but it is a deferred bill with your name on it, and you deserve to price it before closing.

And if you can't answer "what does our study recommend?" — that's the actual gap. Enough starts with knowing the number.