Depreciation: what drives it and what slows it

Depreciation is often described as an annual share, and that is the most misleading simplification on the subject. The path is front-loaded: the largest amount falls in the first year, after which the annual reduction shrinks, until eventually the curve flattens so far that a car is worth roughly the same for years.

In practice: buying a young car means buying the steep part of the curve with it. Buying in the fourth or fifth year leaves that part to the previous owner and puts you in a stretch where the annual loss is considerably smaller. That is the real argument for buying used, and it is an arithmetical one rather than an ideological one.

At the far end of the curve something happens that surprises many people: the value stops falling because it has reached its use value. An old but mechanically sound car is no longer priced from what it cost new but from what someone will pay to be mobile. From that point depreciation has almost vanished as a cost — and repair costs have taken its place.

A car's new price is not a market price but a list price with structure. It contains the dealer's margin, the cost of delivery and of registration, and it is set by the manufacturer. The used market knows none of that: there the price emerges from supply and demand for one particular car on one particular day.

The crossing between the two worlds is the first registration, which is where the drop sits. It has little to do with the condition of the car — mechanically, one with a few hundred kilometres is barely different from a factory-fresh example. What changes is the category it is sold in.

The same mechanism explains why discounts on new cars move the later depreciation figure. A car sold with a heavy discount is worth exactly the same on the used market as any other example of its kind — the gap to the list price was never real. So anyone calculating their own loss starts from the price actually paid, not from the list price.

Two cars of the same model and year fetch very different prices, and the difference traces almost entirely to a short list. That list is also the list of what an owner can influence — which makes it the most useful one here.

Mileage weighs heaviest, but not linearly: around the round numbers buyers search on, prices step down abruptly. Then come the history and the evidence for it. A complete service book is no formality but the only way to prove the state of the mechanicals to a stranger — and what cannot be proved gets deducted from the price.

Accident damage is the special case on this list. Damage properly repaired can be technically irrelevant and still costs value, because it shrinks the pool of interested buyers. The deduction depends less on the quality of the repair than on whether load-bearing parts were involved and whether the work is documented.

Beyond the individual car, characteristics that apply to a whole model do their work. The most important one is mundane: how many there are. A car built in large numbers has cheap parts, plenty of workshops that know it and a wide pool of buyers. Each of those supports the price, and all three are missing on a rare model.

The second is how repairable the engineering is. Elaborate systems — complex gearboxes, air suspension, extensive sensors in bumpers and windscreens, large displays — raise the expected repair bills of an ageing car, and that expectation is exactly what a buyer deducts from the price. This is not a prejudice against technology but a calculation about future invoices.

The third is how many are for sale at the moment you sell. Cars from large fleets and from leases arrive on the market in waves, often in similar trim and of similar age. Selling into such a wave means selling against many near-identical offers; buying into one means buying cheaply. The waves are foreseeable, because the contracts producing them run to fixed terms.

Optional equipment is paid for in full when the car is new and only partly recovered on resale — that is the rule. The exceptions follow a recognisable pattern: what gives a second buyer a capability they would otherwise lack holds its value better than what merely makes things nicer.

A tow bar is the clearest example: someone who needs to tow simply cannot use a car without one, and retrofitting is involved. Gearbox type, drivetrain and an auxiliary heater work similarly where they are needed. Conversely, everything that ages — infotainment, navigation, screen sizes — loses value faster than the rest of the car.

Two items matter more than their price suggests. Paint colour narrows or widens the pool of interested buyers, and a smaller pool means a longer time on the market and a weaker negotiating position. And wheel size works twice over: large wheels cost more to buy, to replace and to wear out, and they make the car less comfortable on poor roads — which shows up in demand at resale.

Part of depreciation has nothing to do with the car. It arises outside and hits whole categories at once — and because it comes from outside, no amount of care, documentation or gentle driving prevents it. Knowing about it at least lets you price it in.

The strongest forces of this kind are rules and prices. A change to city access rules, a new emissions requirement, a shift in fuel or electricity prices — each of them moves demand for an entire drivetrain type, and prices follow within months. What was the sensible compromise yesterday can be hard to sell the day after tomorrow without anything on the car having changed.

Across borders another lever appears: the same car does not cost the same everywhere. Different tax systems, different preferences in body style and drivetrain, and the exchange rate together mean a model can be plentiful and cheap in one market and scarce and dear in the next. For buyers that is an opportunity; for sellers it is a reason not to let the pool of interested parties stop at the national boundary.

On cars with combustion engines the value path is well established, because the market has known it for decades. The uncertainty here lies less in the engineering than in regulation: access restrictions and emissions categories affect older cars first, and they affect them differently from region to region.

On electric cars an additional factor applies that has no counterpart elsewhere: the state of the traction battery. It is the single most expensive component, its condition is invisible from outside, and its remaining warranty has a date on it. So the price of a used electric car depends more on a verifiable finding than on other drivetrains — and a documented condition report is worth hard cash accordingly.

What holds for every drivetrain: uncertainty itself costs money. A buyer who cannot judge a future development demands a discount for it — and the newer a technology, the larger that discount. It disappears as reliable experience accumulates, and that is a process of years rather than months.

The practical conclusion is unspectacular and effective: when you get in matters more than which model you choose. A car whose steep stretch is behind its previous owner and which is still well short of the phase where larger repairs become common is, over the years you keep it, almost always the cheapest option.

The second conclusion concerns how long you keep the car. Keeping it a long time spreads a depreciation you paid once across more years and more kilometres. Changing every three years means paying for the steepest stretch of the curve again and again. Both are legitimate — it is just worth knowing that the decision about holding period is a decision about cost and not a matter of taste.

And the third concerns place: where a model is plentiful, it is cheaper. Comparing prices across the European markets shows those differences immediately, and for a car currently sitting in a return wave in one market, the gap is larger than anything you could negotiate.

On the selling side depreciation cannot be undone, but the price achievable is regularly well above what an unprepared sale produces. The levers are the ones from the list above read from the other side: whatever a buyer cannot verify, they deduct — so make it verifiable.

Timing works too, in two ways. Seasonally, convertibles are in demand in spring and four-wheel-drive cars in autumn — not a rule, but a reliable observation. Structurally, it pays not to sell into the same wave in which a fleet operator is disposing of the same model.

In the end the price you get is a function of how wide a pool your listing reaches. A car visible only regionally gets valued regionally; one visible across the markets finds demand where it is strongest. Carvexia shows listings and prices across those markets, charges no sales commission and no per-listing fee, and is not a party to the sale itself — the contract arises between buyer and seller.