The Hotel Owner’s Guide to CapEx, Reserves & Renovation Cycles

The reserve is a funding mechanism. It is not a plan.

A management agreement typically sets an FF&E reserve as a percentage of revenue. That number tells you what money is arriving. It tells you nothing whatsoever about what the building needs, and a capital plan built backwards from it will be short by exactly the amount nobody wanted to discuss.

And audited industry data — set out with its source on our capital-planning module page — shows hotels spending materially more, over a full cycle, than the conventional reserve provides. Roughly double. That is not a controversial finding and it is not new. It is simply one that most capital plans are constructed as though it did not exist.

The mechanism also has a specific cruelty in it. Because the reserve is revenue-linked, it shrinks in exactly the years the property is under pressure — while the chillers age on the calendar. A reserve that is a percentage of revenue funds least when it is needed most, and that is a structural property of the instrument, not bad luck.

The scope is narrower than the building, too. FF&E is what the reserve funds. Building systems and infrastructure frequently sit outside it entirely, while remaining unavoidably inside the hotel.

The gap does not disappear. It defers.

Deferral is not a decision anybody makes. It is what happens in the absence of one — and it compounds. Facilities research prices a deferred dollar at several dollars by the time it is finally spent, and the multiplier is the least interesting part.

The interesting part is that deferral never arrives as a line in the capital plan. It arrives through the business, in a specific and predictable order.

• The product degrades — quietly at first, in the places guests notice before anybody in the office does.
• Guest scores fall. The industry saw this at scale after 2020: deferred maintenance showed up directly and visibly in facilities satisfaction scores.
• The RevPAR index slips against the competitive set. You are no longer winning your fair share, and now it is measurable.
• Brand quality assurance starts failing.
• And then the spend happens anyway — as a forced PIP on a compressed timeline at full scope, or as a discount on your sale price with the buyer underwriting the renovation.

Plan against three cycles, not one

A hotel does not have a replacement cycle. It has three, and they do not align.

Soft goods run roughly six to seven years. Case goods run twelve to fifteen. Building systems — the plant, the infrastructure, the things nobody photographs — run fifteen to twenty-five and are the largest single number in the plan.

Layer those over five to ten years, price them, and compare the total to what your reserve will actually accumulate. The difference is your funding decision. Make it four years out, when it is a decision, rather than four months out, when it is an emergency.

And refresh the plan annually against condition data rather than against last year’s version of the plan. A capital programme that is updated by copying the previous one forward is a document, not a plan.

The PIP is more negotiable than owners assume

A property improvement plan is triggered by franchise renewal, a sale, a refinancing or a conversion. In every case the brand has something you want, and a moment in which you want it — which is exactly why the PIP arrives when it does.

Practitioner consensus puts a meaningful share of line items as negotiable: swapped, deferred, or value-engineered. That figure comes from people who negotiate these for a living rather than from a study, and it should be read as an estimate. But the leverage is real.

What moves: finishes and specifications where an equivalent alternate exists; sequencing and phasing; and items whose condition genuinely does not warrant replacement — which is an argument you can only make if you have the condition data to make it with.

And that is the whole game. A property that walks into a PIP with condition data, costed alternates and a phasing plan is negotiating. A property that walks in with none of those is renovating at full scope and list price on someone else’s timetable, which is the most expensive way it is possible to renovate a hotel.

Three classes of capital, three different tests

This is where a great many capital committees make an error that looks like rigour.

Return-generating projects are judged on return: IRR, payback, the usual apparatus. Compliance and asset-protection projects are judged on RISK, because they do not generate a return and were never going to. Brand-mandated projects are judged at the PIP negotiation.

Run a compliance project through an ROI threshold and it will fail the test. Correctly. Every time. And you will have deferred the sprinkler upgrade on the strength of a rigorous financial analysis of a question nobody should have asked.

Which means every business case needs the line that most business cases omit: the do-nothing consequence. It is what makes a risk project legible to a committee that thinks fluently in returns.

CapEx versus OpEx: one decision, three consequences

Improvements that better, adapt or restore an asset are capitalised. Repairs that maintain its current condition are expensed. De minimis thresholds keep small items out of the machinery.

And the classification decides more than tax. Repairs and maintenance sit above the line in the P&L, while the reserve sits below it — so a capitalised item that should have been expensed flatters your gross operating profit, and an expensed item that should have been capitalised quietly drains a reserve that will be measured for adequacy later.

Which means a misclassification does not produce one error. It produces a better-looking operating result and a worse-funded future, simultaneously, and it does so without anybody deciding to.

Be suspicious of a pattern where the judgement always lands in the direction that improves the current period. That is rarely fraud. It is the entirely human result of a judgement call made repeatedly by people whose bonus depends on GOP.

The owner–operator tension is structural, not a relationship problem

The operator’s fee rides on revenue and on brand standards. The owner carries the capital. Those incentives are not aligned, and no amount of goodwill aligns them.

Which is why the terms that actually matter in a management agreement are the unglamorous ones: annual capital-plan approval rights, project-level thresholds, and whether the reserve is genuinely cash-funded or merely notional.

An underfunded cash reserve has a particular cruelty to it. It forces an owner equity injection at exactly the moment renovation displacement has made net operating income weakest — which is to say, at the worst possible moment, by design.

And the number that makes it a portfolio decision

Across an estate, the question stops being "does this hotel need money" and becomes "where does the next dollar go". The facility condition index — deferred repair cost over current replacement value — is how that gets answered with arithmetic rather than advocacy.

Say which band scheme you are quoting, because they vary by sector. And read the trend rather than the snapshot: an FCI climbing while budgets stay flat is the deferred-maintenance spiral, visible years before it becomes a capital emergency.

The property whose deferral is compounding fastest is almost never the property complaining loudest. That is the entire value of measuring it.