Guadalupe: San Miguel de Allende's Arts District, Tested

US$315k – 688k

Asking range (no closed-sale data exists)

MXN 35k – 60k

Asking price per m²

Guadalupe is the colonia most often named when someone in San Miguel de Allende wants to point at value. The description is always similar: the arts district, murals, Fábrica La Aurora, flat streets, a few minutes’ walk to the Jardin, and prices below Centro. It is a persuasive picture, and this publisher has repeated a version of it before.

This page tests it. The conclusion is that roughly half the thesis holds and roughly half is narrative wearing the clothes of analysis. The half that holds is about location, terrain and relative price. The half that does not hold is the part that sounds most like an investment case: appreciation and rental yield. Neither of those is measured for Guadalupe by anyone.

What is actually established about Guadalupe

Three brokerages describe Guadalupe in mutually consistent terms: colourful, with a strong artistic community, home to the Fábrica La Aurora arts and design hub, and gentrifying. They are commercial sources with a commission motive, and the descriptions are undated. Consistency across independent firms is weak corroboration rather than evidence, but it is what exists.

The one load-bearing detail in that material is terrain. Guadalupe is flat, as are San Antonio and Guadiana. Atascadero, Ojo de Agua and Balcones sit on the hillside. Slope drives construction cost and it decides whether a stated walking time is real. That distinction is worth more to a buyer than any of the adjectives around it.

On distance, The Clark Group describes Guadalupe and San Antonio together as a “flat few minutes walk to the center”. That is a selling agent’s figure on an undated page, and the same category of claim conflicts elsewhere: Guadiana is variously described as a 10-minute and a 15-minute walk by two different firms. No source publishes a measured distance in metres or a verified walking time from any San Miguel colonia to the Jardin. Walk it yourself, at the time of day you would actually walk it.

Supply: Guadalupe is a thin submarket

The closest thing to neighbourhood-level market data in San Miguel is one brokerage’s count of its own active listings by colonia.

Read this as one firm’s listing book, not as a market census. It draws on the shared local MLS but it is not audited, it mixes houses, condos and land, and a subsequent check found some rows on the same page no longer matching the figures recorded here. The shape is still informative. Centro alone carries about a quarter of the inventory in this book. Guadalupe carries 18 listings.

Eighteen is not a comparable set. It means that in any given month a buyer in Guadalupe may be choosing between a handful of properties that share a colonia name and almost nothing else, and that a seller has no local price signal to work from. Thin submarkets produce wide bid-ask gaps and long, unrecorded negotiations. That is a structural feature of buying here, and it cuts against the idea that Guadalupe behaves like a tracked asset class.

Price: the entry-point case is the strongest part of the thesis

Guadalupe’s whole-property range of MXN 5.5m to 12m converts to roughly US$315,000 to US$688,000. Against Centro’s MXN 8.5m to 20m, the entry point is genuinely lower, and the price-per-square-metre gap points the same way.

Four caveats belong with that table, and they are not decorative.

These are asking prices scraped from listing portals, published by a content aggregator that sells property investment products. They are not recorded sales. The publisher does not disclose sample sizes, and the underlying portals block verification.

They are asking prices in a market with 29.6 months of inventory averaged across the first five months of 2026, up from 17.6 months a year earlier, against a local norm the main brokerage puts at 15 to 18 months. In a market that slow, asking prices sit systematically above clearing prices, and the gap is not published.

The aggregator’s citywide average of about US$390,000 sits far below the same period’s closed-sale average of US$568,829 reported by the local brokerage. The two datasets are measuring different markets: the portals capture the peso-denominated domestic segment, the brokerage MLS captures the dollar-denominated foreign-buyer segment. They should never be blended, which is exactly what happens on most neighbourhood pages.

And every figure above spans different products. A range that runs from MXN 5.5m to 12m for a whole colonia is a statement about the mix of what happens to be listed, not a price band for comparable homes.

The appreciation claim does not survive contact with the transaction data

One aggregator published a 12-month appreciation estimate on 19 June 2026 putting Guadalupe at 9 to 12 percent, San Antonio at 10 to 13 percent, Centro Histórico at 9 to 12 percent and the city at 8 to 10 percent.

That set of numbers has no stated measurement method, no disclosed sample and no repeat-sales or hedonic basis. Set it beside what was actually transacted, as reported by the brokerage whose MLS view the whole market relies on:

Average sale price fell US$81,633, about 12.6 percent. Active listings rose from 593 in May 2025 to 860 in May 2026, an increase of 45 percent, before falling back to 745 in June 2026 on withdrawals and expirations rather than absorption. The brokerage’s March 2026 update states that in several price ranges months of inventory are as high as they have been since the period following the 2008 financial crisis.

The average price decline is partly a mix effect rather than a pure price effect, since the luxury segment contracted from 19 sales to 12. It should not be published as minus 12.6 percent appreciation. But it is the opposite direction from a 9 to 12 percent gain, and the aggregator lists that same brokerage among its own sources, so it had access to the contradicting data.

The honest position for a publisher with no commission at stake is the flat one: measured appreciation for Guadalupe does not exist. There is no repeat-sales index, no hedonic index and no MLS time series broken out by colonia in the public record. The same applies to days on market. The one citywide days-on-market figure in circulation, around 150 days, cites a brokerage market update that contains no days-on-market data at all. Guadalupe is named in the “60 to 120 days for well-priced homes” version of that claim, which inherits the same defect.

Everything on this page that touches transactions traces to a single brokerage’s own MLS view. There is no public registry in San Miguel de Allende, and deed values recorded at the Registro Público de la Propiedad are not published in aggregate and are commonly understated for tax reasons.

The rental yield claim cannot be checked at all

The strongest version of the Guadalupe thesis is that it produces the best short-term rental gross yield in the city. That claim is not testable. No source publishes short-term rental performance by neighbourhood for San Miguel de Allende. AirROI names eight San Miguel neighbourhoods and keeps the per-neighbourhood metrics behind its paid tier; Guadalupe is not one of the eight it names. Airbtics gates its breakdowns the same way. There is no AMPI, MLS or notary-association dataset of any kind, and every occupancy and rate figure available comes from a listing scraper.

What is published is citywide, and it is sobering.

The median listing grosses roughly US$13,150 a year. The mean of US$17,033 is about 29 percent higher because the top decile drags it up, and quoting the mean alone overstates the typical owner’s outcome by that margin.

An illustration, and it is an illustration rather than a measurement: US$13,150 against Guadalupe’s published asking range of US$315,000 to US$688,000 is between roughly 4 percent and 2 percent gross. That arithmetic joins a citywide median revenue figure to a colonia asking-price range from a different publisher measuring a different segment. It is not a Guadalupe yield. It is the order of magnitude a buyer should be arguing against, before any cost is deducted.

The costs are substantial and mostly documented. A named local brokerage publishes a 20 percent short-term rental management fee; the wider quoted range runs 20 to 30 percent. Guanajuato levies a 4 percent lodging tax under Art. 55 of the Ley de Hacienda, and because the host’s relief is conditional on the platform actually remitting, and the state publicly asserts the platform has not been remitting properly, an owner should assume that 4 percent is their own liability. Above that sits the federal line most often left out of these lists: a furnished letting loses the residential IVA exemption under Art. 20 fracción II of the Ley del Impuesto al Valor Agregado, so 16 percent applies whatever the lease length. That is statute rather than an enforcement position, though whether it is in practice collected from individual foreign owners is undocumented, and no scraper discloses whether the revenue figures above are quoted before or after it. San Miguel de Allende charges MXN 45,000 per fiscal year for the specific land-use permit for lodging, confirmed in Art. 24 fracc. XIV of the 2026 Ley de Ingresos, plus a one-time MXN 11,248.64 for the change of land use. At the ECB reference rate of 17.46 MXN/USD on 23 July 2026, the annual permit alone is about US$2,577, which is roughly a fifth of the median listing’s entire gross revenue.

Two further datapoints deserve to sit next to any yield conversation. Airbtics grades San Miguel de Allende in the lowest 3 percent nationally for short-term rental yield, with revenue up 5.9 percent year on year but down 11.5 percent over three years. And the four scrapers disagree sharply on occupancy: AirROI 29.9 percent, AirDNA 38 percent, GuestFavorites 41 percent, Airbtics 44 percent. None publishes its method for handling blocked calendars, which is almost certainly the source of the divergence. Publish the range, never a point.

Gentrification: measured at the wrong resolution

The best evidence on who buys where in San Miguel is peer-reviewed and carries no commission motive. Navarrete Escobedo’s 2022 study of transnational gentrification found that the seven census tracts inside the UNESCO World Heritage polygon held 1,803 foreign residents, 42 percent of the municipality’s entire foreign-born population of 4,312. Foreigners made up 8 percent of the 22,405 people living in those tracts. The single most concentrated tract was 15 percent foreign; a peripheral south-eastern tract reached 29 percent.

That is real data, and it is at the wrong resolution for this page. Census tracts are not colonias, and the study does not name Guadalupe. It tells you that foreign settlement is concentrated in and immediately around the heritage core and that concentration falls off with distance. It does not tell you which side of a given Guadalupe street has turned.

This matters because the uneven-street-to-street quality of Guadalupe is the single most repeated qualitative observation about it, and there is no dataset in existence that resolves it. A colonia name in San Miguel de Allende is not a market unit. With 18 active listings in one brokerage’s book and no sub-colonia data of any kind, the only instrument available to a buyer is a walk, at night as well as in daylight, on the specific block.

The unflattering items

Water. Caminos de Agua’s monitoring map, as reported in February 2022, places Guadalupe in the middle band, described as moderately contaminated, alongside Obraje, San Antonio and Providencia. Centro sits in the most-contaminated group and Guadiana in the cleanest. That report is four years old, gives no concentrations, no thresholds and no well counts, so treat it as a prompt for a question rather than a finding. The underlying problem is not in dispute: the Cuenca Alta del Río Laja aquifer runs a deficit of about 61.98 million cubic metres a year on CONAGUA’s 2024 determination, with explicitly zero volume available for new concessions, and arsenic and fluoride concentrations rise as the water table falls. Ask what the property is connected to, and ask for a current test.

Security. San Miguel is not insulated from Guanajuato. The municipality recorded 88 homicides between September 2024 and August 2025, a rate of 49.3 per 100,000, ranking it Mexico’s 50th most violent municipality. The trend since has been sharply down, with 3 homicides recorded in the municipality in June 2026 against 45 in León and 32 in Salamanca, and Guanajuato is the state with the steepest decline in the country. But in February 2026 cartel reprisal violence reached roughly 23 Guanajuato municipalities including San Miguel de Allende, with an arson incident on 22 February and a body found on 25 February. Any page telling you San Miguel is unaffected is not describing this place.

Carrying cost. The 2026 municipal revenue law sets predial on urban built property at marginal rates from 0.234 percent up to 0.354 percent of cadastral value. Cadastral values sit far below market, so the effective annual carrying cost is very low by North American standards. What cannot be worked out is the amount, because the municipal tabla de valores unitarios de suelo y construcción, which converts those rates into a per-colonia figure, is not published in the statute and could not be located.

What survives

The location case survives. Guadalupe is flat, it is close to the historic core on every account that exists, and its asking prices sit below Centro’s and Guadiana’s on the only comparative table anybody publishes. For a buyer who wants to walk to the Jardin and cannot or will not pay Centro prices, that is a real proposition.

The character case survives, with the note that it is qualitative. The Fábrica La Aurora cluster and the artistic community are described consistently by independent firms and are visible on the ground. Whether they translate into price is a hypothesis.

The investment case does not survive, because it was never constructed. There is no measured appreciation figure for Guadalupe, no days-on-market figure, no recorded sale price, and no rental performance data at neighbourhood level. The gross yield claim rests on numbers nobody has published. The appreciation claim rests on a modelled estimate that runs opposite to the only closed-transaction series available.

That is not an argument against buying in Guadalupe. It is an argument against buying it as a spreadsheet. The honest framing is that Guadalupe is a plausible place to own a house, priced below the heritage core, in a market where nobody, including this publisher, can tell you what it will be worth.

All conversions at the ECB reference rate of 17.46 MXN/USD on 23 July 2026. Spot rates and annual average rates are different measures and are not interchangeable.