Geneva, 24 apartments: the facility that we untangled first

Few places in Switzerland are shorter of homes than Geneva. On 1 June 2026, according to the Federal Statistical Office's annual count published on 10 September, 0.31% of the canton's dwellings stood empty, the second-lowest rate of any canton after Zug and a third of the national figure of 0.93%. The canton's own government had already declared a shortage in every apartment category for the whole of 2026, by a decree of 3 December 2025 that cites a vacancy rate of 0.34% the year before, against the 2% level the decree treats as its benchmark. A residential scheme in central Geneva starts with demand on its side. This anatomy is about one that still took almost two years to finance: case file 2 in our published book, what the developer faced, what had to be put right before the file could go to investors, and how the facility was built.

A scheme the market wanted, and a search that wouldn't close

The project itself read well from the first page. A prime city-centre site, 24 apartments to be built on it, and a sponsor putting substantial equity of their own into the deal, so much that the facility would sit at a loan-to-value of 25%. For a development loan that is unusually conservative: the investors' claim would cover a quarter of the value the file was assessed against, with the remaining three-quarters beneath it as cushion. In a canton with 0.31% vacancy, the question of whether finished apartments would find occupants hardly arose.

Even so, the developer had been searching for close to two years when the file reached us. It came through the contractors who were set to build the scheme, people who had watched the project wait from the closest seat. When a site and equity of that quality go unfunded for that long, the cause is usually somewhere in the documents, and in this file it was.

What had to be resolved first

Our record of the file notes internal and communal challenges that had to be resolved before the deal could be presented compliantly to investors. The specifics stay confidential, and an anonymised case file is the wrong place to reconstruct them. The shape of the problem is general enough to describe, though. A credit process that meets an open issue in the documents tends to stop there: the issue goes back to the borrower as a question, and weeks later the conversation restarts from the beginning. Repeated across several banks over two years, that cycle produces a project nobody has formally declined and nobody has funded either.

We took it the other way round. The first steps of our nine-step diligence are an initial project assessment and legal due diligence, in which Swiss and local law firms vet the borrower and the structure, and here they were used to map exactly what stood between the project and an investable file. Then the issues were worked through with the developer until each one was resolved. Only after that was the deal shown to investors, as a clean file with the history of its resolution documented inside it. A term sheet followed within 1.5 months of first contact, after almost two years elsewhere.

One facility, two layers of security, a quarter of the value

Acquisition and construction sat in a single CHF 12'000'000 facility over 36 months, so buying the site and building the apartments ran under one set of terms from the start of the project to the end of the build. The security is the pair recorded on every one of our twelve published case files: a first-rank mortgage on the property in the investors' favour, registered as a condition of their funding, and a personal guarantee from the borrower standing behind it. The loan-to-value of 25% reflects the two things the deal record names, heavy sponsor equity and the prime location. The facility has been funded.

For investors, the order of those steps is what counted. A file sent out with issues still open hands the unfinished diligence to whoever reads it. Each investor here received one in which the open questions already carried answers, read it, and decided on what was in front of them.

How this connects to Watson

This is our own deal, so the link needs no stretching. Our deal box is published (EUR 1M to EUR 50M+, a loan-to-value of up to 75%, terms of 6 to 72 months), and a borrower can check a project against it in a minute. What the box can't show is the structuring that some files need before they fit anywhere: two of our published case files, Geneva and a Barrow-in-Furness land deal, came to us after 18 months to two years of searching elsewhere, and the Oslo restoration arrived with an ownership and documentation trail that had to be untangled first. Introductions keep bringing these files in, too: the Geneva contractors brought this one, as a partner in Spain brought the Oslo file. A developer weighing a similar project can read the criteria on the borrowers page. Investor and borrower material is available on request.

Watson AM, Basel. This article is general market commentary for professional and qualified investors and for real-estate developers. It isn't investment, legal, or tax advice, or an offer of any financial product.

Sources

• Federal Statistical Office (FSO), Logements vacants au 1er juin 2026, press release, 10 September 2026 (national vacancy 0.93%; Zug 0.20%, Geneva 0.31%, Obwalden 0.38%): admin.ch.
• Republic and Canton of Geneva, Arrêté déterminant les catégories de logements où sévit la pénurie (RSG I 4 45.07), adopted 3 December 2025, in force 1 January 2026 (average vacancy 0.34%, 858 of 252'906 apartments at 1 June 2025): silgeneve.ch.
• Watson AM deal record, case file 2 of the published deal book (anonymised).

Back to The Casebook