In this lesson · 7 sections
Almost everyone who owns a home believes they made money on it, and almost nobody has actually worked out whether they did. The reason is not stupidity. It is that housing is the only large investment most people hold where the accounting is genuinely hard: the asset also delivers a consumption service, the leverage is enormous, the costs are paid in dribs and drabs over decades, and the exit price is a single number quoted years after the entry price with no adjustment for anything in between.
This lesson builds the accounting properly, for the two distinct cases — the home you live in, and the flat you let out. The Swiss tax layer sits on top of all of it and is heavy enough to deserve its own treatment, so it is handled in the next lesson. Here the arithmetic is pre-tax, which is the right place to start because the pre-tax economics determine whether the tax treatment is worth having.
The long-run price record, before anyone gets excited
Start with the fact that most conversations skip. Over very long horizons, real house prices — that is, prices adjusted for consumer price inflation — have grown far more slowly than people assume, and in several well-documented series barely at all.
The longest continuous price record for a single housing market runs along the Herengracht canal in Amsterdam, where Eichholtz assembled repeat-sale transactions from the seventeenth century onward. Across roughly three and a half centuries, real prices ended not dramatically above where they began, having gone through enormous multi-decade swings on the way. Broader international work by Knoll, Schularick and Steger, covering fourteen advanced economies from 1870, finds real house prices roughly flat for the first century of the sample and then rising substantially in the second half of the twentieth century — and attributes most of that later rise to land prices rather than to structures. The structure depreciates; the land is what appreciates, and land appreciates where planning constraints and income growth collide.
This does not make housing a bad investment. It makes the source of housing returns clear: historically it has been dominated by rental income — the rent you receive or the rent you avoid paying — not by capital appreciation. Jordà and co-authors, assembling total returns on housing and equities across sixteen countries since 1870, find housing total returns broadly comparable to equity returns over the long run with substantially lower measured volatility, once rental yield is included. Take the rent out of that calculation and the asset looks mediocre.
So the first honest step is to stop treating price appreciation as the return. It is one component, and historically the smaller one.
The full cost stack of owning
Write down every cash flow, then decide what is investment and what is consumption.
Financing cost. Mortgage interest is a cost. Amortisation — repaying principal — is not a cost; it is a transfer from your bank account into your equity in the property. Treating amortisation as an expense is the single most common error in household property maths, and it makes owning look worse than it is. Conversely, treating interest as "money you get back" makes it look better than it is. Interest is gone.
Maintenance and capital renewal. A building wears out. Kitchens, bathrooms, boilers, façades, roofs and lifts all have finite lives, and their replacement cost is a real, recurring expense even in years when you spend nothing. A commonly used planning figure is around 1% of the property's value per year averaged over the long run; for older buildings, or for property with a shared-ownership renovation fund, the true figure can be higher. If your accounting shows zero maintenance for six years and then a large renovation in year seven, you have not saved money — you have deferred a liability.
Ancillary and ownership costs. Building insurance, administration, the sinking fund for jointly owned buildings, and the portion of utilities not borne by a tenant.
Transaction costs. Notary, land registry, agency commission, mortgage set-up and, in many cantons, a transfer tax. A round trip of several percent of value is normal. Spread over a short holding period this is brutal; over twenty years it is a rounding error. Holding period is a first-order variable in property returns and is almost never modelled.
Opportunity cost of the equity. The down payment is capital that would otherwise be invested. Leaving it out of the comparison is not conservatism, it is an error.
A worked comparison: buy or rent
Take an apartment with a purchase price of CHF 1,000,000. You put down CHF 200,000 and borrow CHF 800,000 at 2.0%, so interest is CHF 16,000 per year. An equivalent flat rents for CHF 2,500 a month, or CHF 30,000 a year — a 3.0% gross yield on the purchase price, which is a realistic order of magnitude for a good Swiss urban location.
As owner you pay interest of CHF 16,000, a maintenance and renewal reserve of CHF 10,000 (1% of value) and ancillary ownership costs of, say, CHF 3,000. Total cash outflow: CHF 29,000. Against that, you no longer pay CHF 30,000 of rent. Net cash position: plus CHF 1,000 per year.
Now the comparison. If you rent instead, you invest the CHF 200,000 elsewhere. Suppose it earns 5% nominal — CHF 10,000 a year. For ownership to match that, the property's appreciation must make up the difference: CHF 10,000 minus the CHF 1,000 cash advantage equals CHF 9,000, on a CHF 1,000,000 asset. The break-even is 0.9% nominal house price growth per year.
Add transaction costs. A 5% round trip on CHF 1,000,000 is CHF 50,000. Held for ten years, that is CHF 5,000 a year, or another 0.5% of value. The break-even rises to roughly 1.4% per year. Held for three years, the same CHF 50,000 costs CHF 16,667 a year and the break-even jumps above 2.5%.
Two things fall out of this immediately. First, the answer is extremely sensitive to the mortgage rate, because interest is the largest single line. At 4.0% instead of 2.0%, interest becomes CHF 32,000, the annual cash position swings from plus CHF 1,000 to minus CHF 15,000, and the break-even appreciation rate rises to about 2.5% before transaction costs. Second, the answer is sensitive to what you assume the alternative investment earns, which is a statement about your own portfolio, not about the property.
What leverage actually does
Now look at the same purchase from the equity holder's point of view, because that is where the drama lives.
If prices rise 2% in a year, the property gains CHF 20,000. Add the CHF 1,000 net cash advantage and the economic gain is CHF 21,000 on CHF 200,000 of equity — 10.5%, from an asset that moved 2%. The five-times leverage turned a modest price move into a large equity return.
Run it the other way. A 2% price fall produces a CHF 20,000 loss, offset by CHF 1,000 of cash advantage: minus CHF 19,000 on CHF 200,000, or minus 9.5%. A 20% fall — well within historical experience in most housing markets, including Swiss regions in the early 1990s — wipes out the entire CHF 200,000 and leaves the mortgage larger than the property is worth.
The asymmetry that matters is not mathematical, it is contractual. You cannot be margin-called on a 2% price fall in a house the way you can on a futures position, so leveraged housing feels safer than it is. But the risk reappears at refinancing: if the loan-to-value ratio has deteriorated when the mortgage rolls, the bank can require amortisation or additional collateral at exactly the moment you can least provide it. And housing downturns correlate with recessions, which correlate with job loss. The three risks arrive together.
The buy-to-let case is a different business
Letting a flat to someone else removes the consumption element and replaces it with an operating business. The accounting gets cleaner and the economics get worse in ways owners routinely underestimate.
Gross yield is the headline: annual rent divided by purchase price. Net yield is what you actually get. From CHF 30,000 of gross rent, subtract vacancy and tenant turnover — even at a modest 4% allowance that is CHF 1,200 — plus non-recoverable operating costs and letting fees of perhaps CHF 2,000, plus maintenance of CHF 10,000, and the CHF 30,000 becomes roughly CHF 16,800. On a CHF 1,000,000 price, that is a net yield of about 1.7% before any financing and before tax. Investors who compare a 3.0% gross property yield to a 3.0% bond yield are comparing a gross number to a net one.
There is also a labour input that never appears in the spreadsheet: finding tenants, chasing rent, managing repairs, dealing with the one bad tenant in ten. Delegating it to a managing agent costs a further slice of rent. Property is not a passive asset; it is a small business with a real estate balance sheet.
Finally, concentration. One flat in one building on one street let to one tenant is not diversified exposure to residential property. It is a single position with idiosyncratic risk — a bad neighbour, a structural defect, a change in the local employment base — that the market will not pay you for bearing, because it is diversifiable in principle. Listed and pooled vehicles solve this, at the cost of the tax and leverage advantages of direct ownership, and the lesson on listed real estate takes that trade-off apart.
What to take away
- Over long horizons, most of the return on residential property has come from rent — received or avoided — rather than from real price appreciation, and appreciation has come mainly from land rather than from structures, which depreciate.
- Amortisation is not an expense but a transfer into your own equity; interest, maintenance of roughly 1% of value per year, ancillary costs and transaction costs are the real cost stack, and leaving out the opportunity cost of the down payment invalidates any buy-versus-rent comparison.
- On a CHF 1,000,000 flat at 80% loan-to-value and a 2.0% mortgage rate, ownership breaks even against renting-and-investing at about 0.9% annual price growth, rising to roughly 1.4% once a 5% round trip is spread over ten years, and above 2.5% if the mortgage rate is 4.0%.
- Five-times leverage converts a 2% price move into roughly a 10% move in your equity in both directions, and the real danger point is refinancing rather than a margin call.
- A single let flat is an operating business with a net yield roughly half its gross yield, plus an unpaid management workload and undiversified, uncompensated idiosyncratic risk.
Sources
- Òscar Jordà, Katharina Knoll, Dmitry Kuvshinov, Moritz Schularick and Alan M. Taylor, "The Rate of Return on Everything, 1870–2015", Quarterly Journal of Economics, 2019.
- Katharina Knoll, Moritz Schularick and Thomas Steger, "No Price Like Home: Global House Prices, 1870–2012", American Economic Review, 2017.
- Piet M. A. Eichholtz, "A Long Run House Price Index: The Herengracht Index, 1628–1973", Real Estate Economics, 1997.
- Robert J. Shiller, Irrational Exuberance, 2nd and 3rd editions — construction and interpretation of the long-run US real home price index.
- Karl E. Case and Robert J. Shiller, "Prices of Single-Family Homes Since 1970: New Indexes for Four Cities", New England Economic Review, 1987 — the repeat-sales method.
- Bank for International Settlements, residential property price statistics and long series database.
- Swiss Federal Statistical Office (BFS/OFS), Swiss residential property price index and rent index.
- Swiss National Bank, Financial Stability Report — recurring analysis of Swiss mortgage lending standards and housing market imbalances.