CASE-03 · Banking & credit risk
Increasing office-building loans by 30% in city «M»
73
Banking
Banking
It isn't diversification: it's concentration in disguise.
In one line Avoided a concentration disguised as diversification in a declining market, before the next downturn.
The head of commercial real-estate lending, with a portfolio growth target set by the board and a relationship team bringing new applications every week.
A regional bank with a traditionally conservative commercial property portfolio. City M is a mid-sized office hub that saw a construction boom in the pre-pandemic decade. Relationships with the main local developers go back twenty years.
The mid-tier office market in city M shows contradictory signals: nominal occupancy rates hold up thanks to multi-year leases not yet up for renewal, but active search for new space is at a historic low. Remote work has reduced effective space demand per employee by 30–40% at medium-sized services firms — exactly the typical tenant of B and C grade offices. Prime buildings with high energy ratings and flexible layouts hold or appreciate; mid-tier has no such defense. The peak of rental contract expirations is clustered in 2025–2027, meaning price correction could materialize precisely as the bank needs to roll its exposure.
The office portfolio yields more than average: margins are attractive, collateral looks solid, clients are known. The issue is not the quality of individual loans — it is the structure of the portfolio. Thirty loans on different buildings in the same city in the same market segment are not thirty risks: they are one risk with thirty exposures. If the mid-tier market in M corrects — and the forces driving it down are structural, not cyclical — the collateral collapses together, clients struggle to refinance together, and the 2026–27 maturity wall arrives at exactly the worst moment.
Indexed transaction values for mid-tier offices in city M over the past six years — the post-pandemic correction has not been absorbed (illustrative data).
It looks like diversification, it's the opposite. Mid-tier offices in that city are in structural decline: remote work empties them, and those who remain pick only the best buildings. The point is that all the collateral would collapse together, for the same reason, at the same time — and most of the loans all mature in 2026-27. You don't have thirty different risks: you have one, repeated thirty times.
Provenance: transaction prices in the M market · current rent roll · loan maturity schedule · red-team base.
Composite cases, in the method of the Harvard Business Review: reconstructions based on real, recurring situations in each sector, merged and anonymized to protect confidentiality. The decision dynamics are authentic; names, figures and details are altered and not traceable to any single client or case. The «provenance» notes describe the type of evidence the engine cites with traceability in production. The Δ-CSI values illustrate the intensity of the pressure the contradiction put on the assumptions.