What Airbnb Actually Pays in San Miguel de Allende The Full Distribution

The number most agents quote you when you ask about Airbnb income in San Miguel de Allende is the number that closes deals. It is not the number that describes what a typical property owner actually earns. Those two numbers are very different, and confusing them leads to investment decisions built on projections that most properties will never meet.

This analysis publishes the complete revenue distribution for the market, from the top decile to the bottom quartile, and then says plainly which of the things you will be told about that distribution are measured and which are asserted. No best-case framing. No anecdote about the one property that had a good year. The full picture, including the parts of it that nobody has published.

The revenue distribution

AirROI publishes the only quartile breakdown available for this market, covering the twelve months to June 2026. The annualised column is simply the monthly figure multiplied by twelve.

The median listing grosses $13,152 a year. Not $60,000. Not $40,000. $13,152, before a single cost is deducted.

The mean sits about 29 percent above it, at $17,033, and that gap is the whole problem with how this market is sold. It is not noise. It is the arithmetic signature of a market where a professionalised minority captures most of the revenue, and it is why a single average quoted as though it described a typical outcome materially overstates what a new entrant should expect. A listing in the top decile grosses nearly twelve times one in the bottom quartile. When somebody quotes you a figure in the sixties, they are quoting the ninetieth percentile without saying so.

Two things about that table are worth holding onto. Every figure in it is gross, before management, tax, the municipal permit and predial. And every figure in it comes from a listing scraper. There is no booking-verified dataset for San Miguel de Allende: no MLS, no AMPI chapter and no notarial body publishes rental performance, and the platforms publish nothing at market level. Airbtics, working a different window, reports a median far above AirROI’s, and the two cannot be reconciled from anything either of them publishes.

The occupancy figure, and the occupancy figure nobody has

Citywide occupancy is measured by four vendors, and they do not agree.

Fourteen percentage points separate the extremes, on the number that determines whether a listing is viable at all. None of the four publishes how it handles blocked calendars or dormant listings, which is almost certainly what drives the divergence, because excluding unavailable nights from the denominator raises apparent occupancy. The honest position is a band of roughly 30 to 44 percent, not a point estimate. The AirDNA row could not be retrieved directly, since that page returns HTTP 403, and its figures come from search snippets rather than a page we read.

Now the part that matters more, because it is what most Airbnb income articles get wrong. Occupancy broken out by revenue tier is not published by anyone. The distribution above exists for revenue only. No vendor publishes what the top decile achieves on occupancy, or the bottom quartile, or the median. If you have been shown a table claiming the top ten percent of San Miguel listings run at seventy percent or better, that number is not sourced to a published dataset, because no such dataset exists at any price. It was either modelled from assumptions the author did not state or invented outright.

That absence is why this article is organised around revenue rather than occupancy. Revenue is published. Occupancy by tier is not, and a structure built on it would be a structure built on nothing.

The incumbent field you would be joining

A 46-day lead time marks San Miguel as a planned-trip destination rather than a spontaneous one, which matters for how quickly a new listing can fill a calendar. Half the listings hold Superhost status. This is not an amateur field, and a new entrant should not assume it reaches even the market average in year one.

On supply, the counts themselves conflict. AirROI records 2,285 active listings, Airbtics 2,149, GuestFavorites 2,107, and AirDNA roughly 4,004 to 4,200 for the same market. The Guanajuato state government counted 1,843 properties in June 2026, which is not measuring the same thing at all: that is a count of properties the state can identify and attribute, which describes the enforcement gap rather than the market. The president of the San Miguel hotel and motel association estimates the city holds roughly 3,500 hotel rooms and that platform supply is comparable. That is a judgement from a direct commercial competitor rather than a count, and it should be read as one.

What determines which tier you land in

This is where the published record thins out sharply, and it is worth being explicit about that rather than filling the gap with confident prose. The distribution is measured. The causes behind it are not, at least not for this market. What follows is what can actually be said about each of the levers you will hear named.

Location

The premium on walkability is the most repeated claim in San Miguel property marketing and the least documented. No vendor publishes short-term rental performance by colonia here. AirROI names eight San Miguel neighbourhoods in its free view and keeps the metrics for all eight behind its paid tier. The walking times themselves are agent-stated, undated and mutually inconsistent, with the same colonia variously described as a ten and a fifteen minute walk to the Jardin by two different brokerages, and no measured distance in metres from any colonia to the Jardin has been published by anyone.

So the position this article takes is narrow: location is very likely to matter, the market plainly prices it, and nothing published for San Miguel de Allende quantifies how much. Anyone who hands you a per-neighbourhood daily rate or occupancy figure for this city is not reading it off a source.

Management

Here there is a number. One named local operator, a brokerage that also sells into this market, publishes 20 percent of the rental income stream. A commission-adjacent publisher gives a prevailing range of 20 to 30 percent for short-term rentals against 8 to 12 percent for long-term, and a narrower 18 to 25 percent elsewhere on its own site. The 20 percent floor is corroborated by a named publisher. The 30 percent ceiling is not corroborated by anyone.

Model 20 percent as the floor and treat anything above it as a quote to be checked rather than a market rate. On the median listing, 20 percent is about $2,630 a year, the second-largest single deduction after IVA.

Seasonality

Seasonality in San Miguel is an occupancy phenomenon rather than a pricing one, and this is measured.

Occupancy swings by a factor of 1.8 across the year while the rate moves by only 1.2. An owner who assumes they can price their way through the summer trough is working against the shape of the demand curve. AirROI groups February, March and December as peak season, averaging 38.4 percent occupancy, and May, June and September as low, averaging 26.2 percent.

Which month actually leads is disputed. AirROI puts February first; GuestFavorites identifies March and April as peak and September and October as low. Both cannot be fully right. Semana Santa fell between 22 March and 8 April in 2026, and its movement between the two months year to year probably explains part of the disagreement. The local claim that March is unambiguously the peak month is not what the one dataset publishing monthly figures shows.

One specific correction, because this figure circulates widely. No source publishes a December average daily rate for San Miguel de Allende. Only February’s $222 and June’s $183 are available at month level. Any December rate quoted to you is either an inference or a misread of the $210 annual market average.

The neighbourhood yield table that does not exist

There is no published short-term rental gross yield for San Miguel de Allende, by neighbourhood or citywide. This is not an omission from this article. It is an absence in the public record, and it has two independent causes.

The revenue side averages across every active listing in the market, including part-time and badly run ones. The price side has no defensible denominator at all: every San Miguel transaction figure available traces back to a single brokerage’s own view of the local MLS, and no public registry publishes sales here. Dividing one by the other produces a figure accurate to four significant digits and wrong in ways a reader cannot see.

A previous version of this article carried a per-neighbourhood yield table. It has been removed because it could not be traced to a source, and no replacement for it exists to publish. What can be published is the arithmetic: readers who want a yield can divide $13,152 by their own acquisition price and will at least see exactly what they are dividing.

The regulatory reality you need to know

A property used for lodging in San Miguel de Allende requires a specific land-use permit, a change from residential to tourist commercial use. The annual municipal fee runs 45,000 Mexican pesos per fiscal year under Article 24 fraction XIV of the 2026 Ley de Ingresos, which is about $2,577 at the European Central Bank reference rate of 17.4601 pesos to the dollar on 23 July 2026. A separate one-time charge of 11,248.64 pesos applies to authorising the change of use itself, so a converting owner pays both in year one. Some press coverage cites 40,000 pesos; the statute says 45,000 and the statute governs.

The municipal permit is not the whole public cost, and the two charges above it are the ones most often left out of an income model. Guanajuato levies a 4 percent lodging tax under Article 55 of its Ley de Hacienda, and because the host’s relief from remitting it depends on the platform actually paying it over, an owner should assume the liability is their own. Federally, a furnished letting loses the residential IVA exemption: Article 20 fraction II of the Ley del Impuesto al Valor Agregado excludes property supplied furnished, or destined for or used as lodging, so 16 percent applies whatever the lease length. That is statute rather than an enforcement position. Whether it is in practice collected from individual foreign owners is undocumented, and none of the rental data platforms discloses whether its revenue figures are quoted before or after the tax.

On enforcement, the record is thinner than either side of the argument suggests. The best hard evidence is four years old: in all of 2022 the municipality collected 175,442 pesos from the land-use charge, from approximately 243 registered properties, against a listing count in the thousands on every scraper. The fee then rose from 10,000 to 45,000 pesos for 2023 and has not moved since, including for 2026. In July 2026 the state tax authority sent roughly 5,000 invitation letters to platform hosts, describing that first phase as friendly with no sanctions and flagging audits as a possible second phase, and publicly accused Airbnb of three and a half years of non-compliance. No 2026 compliance rate is published by anyone, which is arguably the single most important regulatory number in this market and it does not exist.

The direction of travel is collection intensity rather than rate changes. The municipal fee has been flat for three years, and the 4 percent state tax survived the November 2025 reform of the state finance law unchanged. Budget for the permit, the 4 percent and the 16 percent IVA as recurring costs rather than contingencies, and model tightening enforcement as the variable most likely to move.

The calculation most buyers get wrong

The typical first-time investor here models the purchase on top-tier revenue, applies a 20 percent management fee, deducts nothing else, and arrives at a projected return that looks compelling. Then they buy and spend the first year discovering their property is performing at the median.

Start from the median instead: $13,152 of gross. Take out 16 percent IVA on the furnished letting, 20 percent management, the 4 percent state lodging tax, federal income tax withholding, the 45,000-peso permit and predial. On a property with a fiscal value of 5 million pesos, those lines come to roughly $9,170, leaving about $3,982 before utilities, cleaning consumables, linen, insurance, repairs, furniture replacement, accounting, or any period the owner uses the property personally.

The permit is the line that behaves worst, because it is flat. It does not scale down with revenue, so it takes a far larger share of a small listing’s gross than of a large one.

A compliant bottom-quartile listing hands nearly half its gross to the municipality before it pays anyone to clean it. This is the mechanism by which the gap between the best case and the median is wider after costs than before them, and it is why the honest question is not what the market earns but what your specific position in the distribution earns after the stack.

Ask whether the investment makes sense at the median. If the answer is yes, any outperformance is genuine upside. If the answer is no, you are relying on best-case assumptions to justify a purchase, which is a different kind of decision.

Model your short-term rental investment on the median. Optimise for the top quartile. And be clear-eyed that the variables separating them are, in this market, asserted far more often than they are measured.

The SMA Wealth Intelligence Report contains the complete short-term rental analysis, including the four-way occupancy disagreement, the full statutory cost stack, and a modelled median listing worked line by line. Download it free on this site.

---

Data sources: AirROI, Airbtics, AirDNA and GuestFavorites, all listing scrapers rather than booking-verified datasets. All revenue figures USD and gross. Peso conversions at 17.4601 MXN per USD, the European Central Bank reference rate for 23 July 2026. Past performance does not guarantee future returns.