Operating costs: heavy capital, expensive labor

3.2.1 — Why focus on labor: it is the largest controllable cost

Before comparing France with its neighbors, this focus needs justifying. This paper dwells on labor cost not out of habit but because it is the heaviest cost line, and the first over which a hotelier has room to maneuver.

In hotels, payroll costs represent roughly one-third of revenue: 33.3% in 2019 for the sector as a whole, the last undisrupted year,^ᶜ and, in the RYDGE Conseil panel for 2025, from 29.7% in 3-star hotels to 41% in standard 5-star hotels.^⁶⁵ It is by far the largest operating cost. Next come overheads, which a 2016 industry benchmark puts at around 20% of revenue.^⁶⁶ Last come occupancy costs: depreciation and interest for the owner-operator, rent for a leased hotel. Within overheads, what matters is less their amount than their nature. One part is fixed and non-discretionary: insurance, taxes, subscriptions. The other can be managed: energy, water, maintenance. That second part is where technology promises the most, and an existing mechanism funds part of it (Section 7). When labor absorbs a third of revenue, any distortion in its cost flows straight through to the margin. Hence the focus.

3.2.2 — A high margin, heavy capital

The often-heard claim that margins are thin does not hold in percentage terms. Before rent and fees, the operating margin reaches 41% of revenue in 3-star hotels. After rent, the median gross operating surplus (Banque de France) of the accommodation sector still represented 20.5% of revenue in 2024. In nominal terms, hotel revenue rose 38% between 2019 and 2025, while the sector's base wage rose 16% (an index per employee, not total payroll). The constraint lies elsewhere. Accommodation is far more capital-intensive than trade, industry or transportation, with about €233,000 of operating fixed assets per employee, more than two and a half times the level in industry. Measured against that capital, its net return (6.7%) is comparable to industry (7.8%) and below trade (10.4%) or restaurants (12.1%). The issue is therefore not the size of the margin but the return on capital relative to its cost. Investors weigh hotels against other investments, starting with other real estate (4.2 and 4.3). In early 2026, the prime hotel yield (4.75%–5.25% in Paris) exceeded the cost of real-estate loans to businesses (about 3.5%): leverage was positive, but thin. We did not find any series published as of September 30 that already measures the cost of real-estate lending to businesses after this summer's rise in government bond yields (4.3); leverage must therefore be tested deal by deal, and investors now require higher returns on their equity (4.2) and higher yields on non-prime stock (4.3). For the independent core, the margin is enough to operate; what is thin is the return on the capital that renovation requires.^⁶⁷

The total hourly labor cost in accommodation and food services in France was €29.8 in 2025, the highest among Europe's major tourism markets: €27.4 in Germany, €20.8 in Italy, €17.6 in Spain and €20.9 on average across the EU.^⁶⁸ An hour of work in accommodation and food services thus costs a French employer 69% more than a Spanish one, in current euros and before any adjustment for productivity differences. The structure of that cost makes matters worse: non-wage costs account for 35.1% of the total in France, against 27.3% in Spain and 28.6% in Italy.^⁶⁹ This non-wage share funds social benefits; the average net wage in the HCR sector (hotels, cafés, restaurants) remains among the lowest in the economy.

3.2.3 — A cost that is high in level, not in growth

At the end of 2025 (before the 2026 energy shock discussed below), France had one of the lowest inflation rates in the European Union: 0.7% in December 2025, among the three lowest in the EU alongside Cyprus and Italy, with the euro area at 1.9%. Between January and December 2025, French inflation also fell more sharply than Italy's.^⁷⁰

The 2026 shock does not overturn this diagnosis. Energy prices were up 16.7% year on year in August 2026 and French inflation rose back to 2.4% (2.6% on the harmonized index), but it stayed below the euro area's 3.2%, so the difference in pace favored France through August; the provisional September estimate (3.0%, 3.4% on the harmonized index) shows that gap narrowing, with euro-area data for that month not yet published at the cut-off date.^⁷¹ For the operator, the distinction matters little. Energy is a cost that can be neither deferred nor cut in peak season: a large share of an open hotel's energy use does not depend on occupancy (heating and lighting of public areas, hot water, kitchen equipment). The shock therefore adds to an already high cost base without taking anything away from it.

The misreading to avoid, then, is to say that "costs are exploding in France." That is not true in the inflation sense: consumer prices there are rising more slowly than elsewhere in Europe. General inflation does not, however, measure a hotel's cost basket (energy, insurance, commissions, wages at the floor), for which we have not identified a public series. France's problem is not the pace of cost increases but their starting level, especially for labor. France starts from a high point and is not accelerating; its competitors start from a lower one. Over time, the gap in level persists. It is a structural disadvantage, not a cyclical one — and that is precisely what makes it hard to correct.

3.2.4 — What labor really costs, beyond the wage

First, a common misconception must be set aside: that wages are soaring. Over the past four years, the sector's base wage rose 12.2%, while prices rose 12.4% and the minimum wage 13.4%.^⁷² Pay has thus tracked prices without outpacing them. What rises faster is the statutory minimum wage (Smic): the hourly floor below which no one may be paid, raised at least once a year and more often when inflation accelerates. In a sector where the average monthly net wage is €1,979 in full-time equivalent, against €2,733 for the private sector as a whole, the consequence is mechanical: the floor moves closer to the pay grades just above it, the employer must lift the bottom of its pay scale while the top stays put, and the differential between a skilled position and an entry-level one narrows. One caveat, and a decisive one: this measure covers only permanent positions in establishments with at least ten employees. The independent core and seasonal hires are not included. That the data show no wage escalation therefore does not prove there is none on the ground: it marks the limit of the measure.

Second, the cost of labor does not stop at the payslip. It also carries two indirect costs, and neither of these two lines appears as such in an income statement.

First, in replacement costs. The sector loses employees faster than others: its employees on permanent contracts resign almost three times as often as in the economy as a whole, and more than one planned hire in two is seasonal (Section 5). These data cover accommodation and food services combined, where food services account for about three-quarters of employment (note 174): they give an order of magnitude for hotels, not a hotel-specific measure. Each departure costs recruitment, training and lost productivity while the replacement gets up to speed. The mechanism is well established; however, our research did not identify any published French measure that quantifies it for hotels. The only available orders of magnitude are American; they are reported, not transposed, in Section 5.

Second, in forgone revenue. For lack of staff, some hotels close a wing, a restaurant or a floor, or cut their opening hours. This is the costliest and least visible form: not a rising cost, but revenue that never comes in.

The employment structure that generates these costs (seasonality, turnover, housing for seasonal staff) is as much a human-capital issue as an accounting one. It is addressed in Section 5.

3.2.5 — The cost of financing: a cost in its own right

The last cost element is the cost of capital. Occupancy costs (depreciation and amortization, interest on borrowings, rent) form a block of fixed charges tied to investment and financing decisions. During a lease, rent and depreciation are barely negotiable; only the interest charge can be renegotiated or refinanced. That lever depends on the original rate: debt taken out when rates were low has nothing to gain from refinancing and will refinance at a higher cost at maturity; for other borrowers, the 2026 rise in rates shrinks the gain. All new debt will cost more and, together with labor and taxation, will squeeze the operating surplus.

For a new acquirer, the entry price comes on top. If scarce assets command high prices (3.1.4), the capital that must earn a return grows, and with it the pressure on ADR. First-half 2026 transactions give the order of magnitude at the top of the market: about €1 million per room for the Pullman Paris Tour Eiffel (430 rooms, valued at about €430 million), and more than €1.1 million per room for the Hôtel La Pérouse in Nice.^⁷³ For the Pullman, the RYDGE Conseil study names four co-acquirers: the Caisse des Dépôts, Batipart, Société Générale and QuinSpark.

3.2.6 — Pass through or scale up: the double bind of the small stand-alone hotel

Faced with a high cost, an operator in theory has two levers: pass it through to rates, or dilute it through scale. The independent core is poorly placed on both — and it is this bind, more than the level of labor cost taken alone, that threatens its survival.

Passing costs through, first, requires demand willing to pay the higher price. That pricing power is not uniform: it follows the polarization line. A premium or international property, selling an experience to guests with spending power, passes part of its costs into ADR without losing occupancy; a midscale hotel that depends on price-sensitive domestic guests (Pressure 5) has only narrow room to raise rates: in the MKG panel, it gave ground on rate in 2025 to protect occupancy. The aggregate data confirm it: in 2025, French RevPAR rose only 1.4%, driven by occupancy rather than rate (Section 2). Yet hotel prices did rise: the INSEE index increased 4.1% in 2025, while inflation was only 0.9%, and 26% since 2019, against 16% for the sector's base wage (an index per employee, not total payroll).^⁷⁴ But this index tracks constant-quality prices across the whole hotel stock; it does not show what each segment actually achieves. In the MKG panel, achieved ADR fell in 2025 in economy (−2.4%) and midscale (−1.3%), and rose only in upscale; in our reading, the growing weight of upscale is what held the national average steady.^⁷⁵ Higher prices did not translate into realized rates everywhere: economy and midscale hotels did not achieve them; they are likely the ones absorbing their own cost increases, although we have not identified a public series of their 2025 margins. The first half of 2026 shows a RevPAR recovery, at +3.0%, above French inflation.^⁷⁶ Its durability remains to be confirmed: from August 1 to 17, RevPAR fell 1.5% while inflation rose back to 2.4%, a seventeen-day window too short to judge by, even though aggregate volume is not lacking: hotel guest nights rose 2% in the first half of 2026, on a different frequency and perimeter from the monthly panel.^⁷⁷ And the pass-through is uneven: in the first half of 2026, RevPAR rose 4.5% in upscale and 3.2% in midscale, but only 1.0% in economy and 0.4% in super-economy.^⁷⁶ Guest-night data tell the same story on another perimeter: guest nights in 4–5-star hotels are rising while those in unrated hotels are falling by double digits (quarterly INSEE measure, distinct from the monthly RevPAR panel; full gradient in 3.5.5).^⁷⁸ The sector is passing costs through again; the lower end of the stock much less so, and we did not find any series showing what the independent midscale hotel achieves. The central question is therefore not only "are costs rising?" but "how far can rates follow?" The answer depends less on cost than on guests' ability and willingness to pay, which falls away as soon as the product is seen as dated or behind international standards. The extra cost is sustainable only for an asset that has already been repositioned; it chokes one that has not.

Scaling up, second, is the only alternative — and this is where labor cost stops being a mere accounting line and becomes a structural problem. Absorbing it requires pooling what a stand-alone hotel carries alone: a support function, a purchasing platform, pricing and distribution tools, a brand that reassures the lender. In other words, the weight of labor leads directly into the next pressures: the brand affiliation that provides scale (Pressure 4), the technology that makes it possible to do more with less (Section 6), and the financing without which neither is attainable (Pressure 3). Aligning with group standards is expensive and requires financing sized to the asset. The stand-alone operator is thus caught in a double bind: too small to dilute its costs, too constrained to pass them through to its rates, and without a financing instrument suited to its size to escape either.