Cost of Capital: An Underwriting Input for Prime

Revision notes#

This article was corrected on 27 September 2026. A check of its figures and attributions against the cited sources found errors, which have been corrected in the text. The full list, with what the article said and what it says now, is in the corrections section at the end.

This article was corrected on 29 September 2026. An earlier version of the disclosure at the end described Victaura's positions beyond its public project pages. In Ras Al Khaimah, Victaura holds a capital position in off-plan units built by a local developer, not a development of its own. The disclosure now follows those pages.

The rate is not the signal#

The instinct to read a central-bank rate as a buy or sell signal for prime property is the most common category error in the current market. It rests on a hidden assumption: that the policy rate prices the buyer who clears the marginal prime transaction. Through the first half of 2026 that assumption is running the wrong way. Across the three central banks that matter most for European and Gulf prime, the move being priced at the margin is a hike, not a cut. The European Central Bank raised its three key rates by 25 basis points on 11 June 2026, its first increase since 2023, and again by 25 basis points on 10 September 2026, taking the deposit facility rate to 2.50 per cent. The Federal Reserve held in July against three dissents to raise, then raised its target range by a quarter point to 3.75 to 4.00 per cent on 16 September 2026 on a 12 to 0 vote. The Bank of England held Bank Rate at 3.75 per cent on 17 September 2026 on a 6 to 3 vote, the three minority members voting to raise (all as of 17 September 2026). The cutting cycle the timing narrative assumes is not the cycle in front of us.

The ECB frames the proximate driver as the energy shock from the conflict in the Middle East, and on its own account that shock has not cleared. In its 23 July 2026 statement the ECB Governing Council held the deposit facility at 2.25 per cent, noting that the energy-price outlook, while highly volatile, stood close to the baseline of the June Eurosystem projections and well above the levels recorded prior to the conflict in the Middle East, and that the full inflationary impact of the shock had yet to play out. The June staff projections placed headline euro-area inflation at 3.0 per cent in 2026, 2.3 per cent in 2027 and 2.0 per cent in 2028. The Council explicitly declined to pre-commit to a rate path. It was a hawkish hold, conditioned on a supply shock, not a pause on the way down: on 10 September 2026 the Council raised rates again, stating that the conflict in the Middle East continues to generate inflation pressures, with the September staff projections at 3.0 per cent for 2026, 2.5 per cent for 2027 and 2.1 per cent for 2028.

The temptation is to convert this into a market call, and both directions of that call are wrong for the prime tier. If rates are higher for longer, the reasoning runs, prime must de-rate; if a cut is coming, prime must be bought ahead of it. Each half fails for the same structural reason: the buyer who sets the clearing price of a statutory-scarcity prime asset is not the leveraged buyer the policy rate prices. Reading the rate as a signal to time the prime market confuses the cost of debt with the reservation price of equity, two different numbers moving on two different clocks.

Two prices, not one#

A prime asset has two prices, and only one of them moves with the policy rate. The first is the price a leveraged buyer can pay, which is a function of the cost of the debt that funds the purchase. The second is the reservation price of an equity buyer, which is a function of scarcity, mandate and horizon. In mainstream residential the two prices are close, because the mortgage-dependent marginal buyer lets the debt price set the clearing price; in the prime and super-prime tiers they diverge, because a large share of the buyer pool is not leveraged at all.

The borrowing cost is the policy rate plus a credit spread, and it moves in lockstep with the cycle. When the cost of debt rises, the maximum a leveraged buyer can bid for a given income stream falls, mechanically, because more of the asset's cash flow is consumed by debt service. That is the channel that repriced mainstream housing across developed markets as rates rose from 2022, and the one the trade press has in mind when it treats a hawkish central bank as bearish for property.

The equity buyer's reservation price is set by the irreplaceability of the asset, not by the overnight rate. For a family office allocating its own capital to a lakefront villa on Como inside the 300-metre landscape band under Decreto Legislativo 42/2004 art. 142, the binding input is that the asset is hard to reproduce at any interest rate. Article 142 does not prohibit building: it makes the band a protected landscape area, where any intervention that modifies it needs a landscape authorisation under art. 146. A cut does not unlock that authorisation and a hike does not withdraw it, so new lakefront supply is constrained by the permitting regime rather than by the cost of construction finance (the degree of constraint is a Victaura Research assessment, not a measured figure). The policy rate enters that calculation as the opportunity cost of the equity, not as the price of the asset.

The two prices converge in mainstream property and diverge in the prime tier, and the size of the divergence is the size of the equity share. The more of a market that clears without leverage, the less the policy rate transmits to the clearing price and the more it reaches only the opportunity cost of the committed capital. This is why a hawkish turn can widen mainstream yields while leaving prime values broadly intact, a pattern consistent with the H1 2026 data below. In the segment where the marginal transaction is equity-funded, the rate signals the leveraged bid, not the price.

The marginal buyer is leveraged; the prime buyer is not#

The mainstream market clears on leverage; the prime tier increasingly does not, and the cash share rises with the price band. Where the marginal buyer is mortgage-dependent the policy rate is the price, because affordability is a function of the monthly payment and the payment is a function of the rate. One tier up that force fades. Knight Frank's Wealth Report 2026 notes that nearly 50 per cent of purchases in prime central London are unleveraged. That a similar or larger share of super-prime transactions elsewhere clears with limited or no mortgage leverage, funded from liquidity, from a family-office balance sheet or from the proceeds of a prior sale, is a Victaura Research assumption, not an audited statistic; a buyer who is not borrowing is not priced by the borrowing rate.

The consequence is that a higher policy rate reprices the leveraged bid and leaves the equity bid largely intact. When rates rise, the leveraged buyer's maximum price falls and, in a market where that buyer is marginal, the clearing price falls with it. Where the marginal buyer is an equity allocator, the clearing price holds, and the rate instead raises the hurdle the equity has to clear. The asset does not become cheaper; the bar it must clear becomes higher. Only the first is what timing the rate assumes.

None of this argues that rates are irrelevant to prime. It argues about which channel transmits. The policy rate reaches equity-funded prime through the opportunity cost of the capital, through refinancing risk on any debt actually used, and through the sentiment of a thinner buyer pool at exit. It does not reach it through the affordability of a mortgage that most of the marginal buyers are not taking. That distinction is why the rate should be underwritten rather than timed.

Cap rate is a relationship, not a yield promise#

The capitalisation rate is the most misread number in prime real-estate marketing, because it is quoted as a promise of yield when it is a relationship. A cap rate is net operating income divided by asset value. On its own it says nothing about whether an investor should expect that return, and nothing about whether leverage would add to it or subtract from it. The institutional use of the cap rate is comparative, against the cost of the money that would fund the asset.

The relationship that matters is the cap rate against the borrowing cost, because it determines whether leverage works. When the cap rate sits above the all-in cost of debt, leverage is accretive: each borrowed euro earns more in the asset than it costs in interest, and gearing lifts the equity return. When it sits below the cost of debt, leverage is dilutive, and the more debt is added the worse the outcome. With policy rates running 2.50 to 4.00 per cent across the three banks as of 17 September 2026 and a credit spread stacked on top, the all-in cost of cross-border acquisition debt sits above prime residential cap rates in the tightest, most supply-constrained markets (Victaura Research assumption, reasoned from those policy rates plus a typical cross-border acquisition-debt spread, not a quoted figure). In those markets, leverage is dilutive.

This is precisely why an equity-funded structure is the coherent one for statutory-scarcity prime. An asset whose cap rate sits below the cost of debt cannot be levered accretively, which means the return has to come from the scarcity and the value-add rather than from financial engineering. The market where the running yield is lowest, because scarcity has bid the capital value highest, is exactly the market where debt does least work and equity does most. Como is the archetype: a low running yield, a statutory landscape constraint on new lakefront building, and a return case built on irreplaceability and restoration, not on the spread between a cap rate and a mortgage.

The cap rate is therefore an input to a relationship, never a promise of yield, and it is treated as such here. We do not publish a target yield on any prime asset, because the honest number is a relationship between an income stream, a cost of capital and a horizon, each of which has to be underwritten. A brochure that leads with a cap rate as a return is answering a different question from the one an allocator is asking. The cap rate signals when leverage helps. It does not fix what the asset will earn.

What the prime value and rent data actually shows#

The prime value and rent data through the first half of 2026 is quietly consistent with the equity-anchored reading, and it is a pre-hike baseline. Savills, in its Prime Residential Index for World Cities covering the first half of 2026, reported average capital values up 0.6 per cent and average prime rents up 1.1 per cent across the 30 tracked cities, with 60 per cent of them recording positive capital-value growth. The H1 window closes days after the ECB's 11 June increase, so it records the prime market as it entered the hawkish turn, not its behaviour through the months that follow. In a market priced by leverage, capital values would be expected to fall into the turn. On the Savills average they rose, if modestly, and with rents rising slightly faster than values, average yields did not compress.

Yields that hold as a rate cycle turns are the signature of an equity-anchored market, and the H1 baseline is consistent with that reading. Where the marginal buyer borrows, a rising cost of debt forces yields wider, because the price a leveraged buyer can pay falls and the income stream is unchanged. Where the marginal buyer is an equity allocator, the yield holds because the reservation price holds, and the rate is absorbed as opportunity cost rather than expressed as a repricing. The Savills reading is what the second case looks like on the eve of the turn; the post-hike prints are not yet in the data.

Lake Como sits toward the strong end of this behaviour, and the reason, on this reading, is statutory. Knight Frank's Prime International Residential Index placed Lake Como at plus 6.5 per cent in 2025, 19th of the 100 markets tracked, against an index average of plus 3.2 per cent (The Wealth Report 2026). The mechanism, on the Victaura Research reading, is the 300-metre landscape band under D.Lgs 42/2004 art. 142, which does not prohibit building but subjects any intervention that modifies the protected shoreline to landscape authorisation under art. 146, constraining new freestanding lakefront supply and making restoration of existing stock the usual path. A supply curve that does not respond to the rate is what lets a price enter a hawkish turn without the mechanical de-rating that reprices leveraged markets. That is not resilience as a slogan. It is the arithmetic of constrained supply meeting equity-funded demand. Past index performance is not a reliable indicator of future results.

The cost of capital as an underwriting input#

The correct treatment of the cost of capital is to name it and price it as an underwriting input of equal rank to the title, the tenure and the scarcity. It is not an externality to be waved away with an assertion that prime always holds, nor a market-timing lever to be pulled when the forward curve looks favourable. It belongs in the underwriting with the same explicitness as the landscape authorisation or the tenure ceiling.

Named, the cost of capital enters the underwriting in three specific places. It is the discount rate applied to the equity case, which sets what the future cash flows and the exit are worth today. It is the refinancing assumption on any debt actually used, construction finance or a bridge, which is where a hawkish turn bites directly. And it is the opportunity cost against which the committed equity is measured, the return the same capital could earn risk-free instead. Each rises when the policy rate rises. None of them is the asset's price.

Priced, the cost of capital disciplines the entry rather than vetoing it. A higher cost of capital raises the hurdle the scarcity has to clear, which is a reason to underwrite harder, not a reason to stand aside. An irreplaceable asset that clears a higher hurdle is still an allocation; an ordinary asset that only worked at a lower cost of capital never had a scarcity case to begin with. The rate environment sorts the two, which is the point of pricing the cost of capital honestly, and why the marketing line tends to omit it.

The weaknesses, honestly disclosed#

The equity-anchored argument has real limits, and the institutional reader is owed them with the same precision as the strengths. The claim is not that the cost of capital does not matter to prime, but that it matters through different channels and on a different clock. Four qualifications discipline it.

First, the compression the sell side forecast has not arrived, and the reason is the rate. CBRE, in its U.S. Real Estate Market Outlook 2026, forecast cap-rate compression of roughly 5 to 15 basis points across most property types. Its own midyear review then walked the forecast back, judging that cap rates would hold steady through 2026 because benchmark interest rates stayed higher than expected, with incremental compression now expected in 2027. The higher-for-longer rate has deferred a repricing the industry expected, and an honest account records the forecast that did not come true, not only the thesis that did.

Second, that prime buys with equity is a tendency, not a law. Where debt is used, construction finance, a bridge, or partial leverage on a stabilised asset, a higher cost of capital bites directly, and refinancing risk is real precisely at a hawkish turn, when maturing facilities reprice into a higher rate. A debt-funded position transmits the rate directly into the return, whatever the operator asserts about prime.

Third, the equity that funds prime carries an opportunity cost that rises with the rate. With a near-risk-free anchor at 3.75 to 4.00 per cent in dollars and 3.75 per cent in sterling as of 17 September 2026, the equity case has to beat a higher bar than it did in the zero-rate era. Knight Frank's index put Lake Como prime prices at +6.5 per cent in 2025, above that bar for that year; one year of index data is not a forward promise, and a scarce asset that returns below the bar is still a poor allocation, however irreplaceable. Scarcity protects value; it does not exempt the capital from an opportunity cost.

Fourth, illiquidity compounds at exactly the wrong moment. Statutory-scarcity prime is illiquid at exit, and a hawkish rate environment thins the leveraged buyer pool that would otherwise deepen the bid. The same constraint that protects the asset on the way in lengthens the sale on the way out. The friction is the price of the scarcity, and it has to be underwritten as a cost, not wished away as a virtue.

The operator advantage#

The operator advantage in a higher-cost-of-capital environment is structural, not promotional. An equity-aligned, ring-fenced special purpose vehicle with the principal co-invested does not face the forced-deleveraging channel that repriced over-levered developers in prior cycles: a position not built on a maturing debt stack cannot be margin-called into a distressed sale when rates rise. That is not a claim of superior returns; it is a statement about which failure modes are absent from the structure.

No margin call runs through an equity-funded position, and that changes the behaviour under stress. A developer funded with construction debt must refinance into whatever rate prevails at maturity, and a hawkish turn can convert a viable project into a distressed one at that date. An equity-funded SPV holds to its horizon on its own terms; a delay costs the principal's own capital first, disciplining the timeline from the inside rather than through a lender's covenant.

The cost of capital belongs in the subscription materials, not buried in an assumption of falling rates. The discount rate applied to the case, the funding structure, and the sensitivity of the return to a further hike are the numbers a credible operator publishes, because they are the numbers a credible allocator will ask for. An operator that can only make the case work by assuming the cutting cycle the forward curve no longer prices is answering the wrong question.

What this means for the investor and their advisor#

For the family office or private-bank allocator, the practical read is to stop treating the rate as a timing signal and start treating it as an underwriting input. The question is not whether to buy prime ahead of a cut or sell it ahead of a hike. It is whether a specific asset, in a specific tenure and jurisdiction, clears its hurdle at today's cost of capital, held to a realistic horizon and funded the way it is actually funded. That question has an answer in any rate environment, without forecasting the central bank.

The composition follows the funding structure, not the cycle. An equity-funded position in a statutory-scarcity market, a Como lakefront under the 300-metre band, absorbs a higher cost of capital as opportunity cost and holds its clearing price. A market where supply is expanding rather than statutorily constrained, such as the designated-freehold Al Marjan corridor in Ras Al Khaimah, transmits the same rate through a different channel, the cost of development debt and refinancing risk on new pipeline, not the reservation price of equity against a fixed stock. A levered, cycle-driven position transmits it most directly of all. Sizing the three differently, to the friction and the funding of each, is the allocation decision; reading a single policy rate as a verdict on all is the error.

The cost of capital should be named, priced and stress-tested to a further hike, not assumed away by a forecast of cuts the forward curve no longer supports. As of 17 September 2026 the ECB has raised twice in this cycle, the Fed has raised once, and the Bank of England is holding against three votes to raise. An underwriting that survives another quarter point is worth committing to. An underwriting that requires the next move to be down is a bet dressed as an analysis. This article is not a personal recommendation, and it is not an offer to sell or a solicitation to buy any security or interest in any vehicle. Any investment decision should be taken on the basis of formal subscription documentation, independent professional advice, and a documented assessment of suitability for the investor's specific circumstances.

Skin in the game disclosure. Victaura, through its parent Greystone B.V. (Netherlands), develops projects on Lake Como (Italy), in Nungwi, Zanzibar (Tanzania), on Gili Air and in Uluwatu, Bali (Indonesia), and holds an off-plan capital position on Al Marjan Island, Ras Al Khaimah (UAE). Readers should assume commentary may be influenced by, or benefit, these positions. This document is classified as marketing material under MiFID II Article 24(3). It is not investment advice.

Corrections, 27 September 2026#

What the article said and what it says now. The published version described the Federal Reserve and the Bank of England as holding and the European Central Bank as having raised once; it now records that the ECB raised again by 25 basis points on 10 September 2026 to a 2.50 per cent deposit rate (ECB, 10 September 2026), that the Federal Reserve raised its target range to 3.75 to 4.00 per cent on 16 September 2026 on a 12 to 0 vote (Federal Reserve, 16 September 2026), and that the Bank of England held at 3.75 per cent on 17 September 2026 on a 6 to 3 vote with three members voting to raise (Bank of England, September 2026); the ECB stat, the rates table, the policy-rate ranges and the risk-free anchor have been updated accordingly. The published version stated that Lake Como prime prices rose 54 per cent over five years; that figure is not supported by Knight Frank's Wealth Report 2026, whose five-year PIRI 100 table does not include Lake Como among the markets above 50.5 per cent, and it has been removed; the article now states that Lake Como rose 6.5 per cent in 2025, 19th of 100 markets against an index average of 3.2 per cent (Knight Frank, The Wealth Report 2026). The published version stated that Savills reported broadly stable prime yields in H1 2026 and that about 60 per cent of cities were stable or positive; the yield statement could not be verified and has been removed, and the article now reports average capital values up 0.6 per cent, rents up 1.1 per cent and 60 per cent of cities with positive capital-value growth (Savills World Cities H1 2026, as reported by PNI News). The published version stated that art. 142 of D.Lgs 42/2004 forecloses or fixes new lakefront supply; the provision makes the 300-metre band a protected landscape area in which interventions require authorisation under art. 146, and the article now describes it as a constraint rather than a ban (D.Lgs 42/2004, art. 142). Statements that super-prime buyers are predominantly equity-funded and that debt costs exceed prime cap rates are now labelled as Victaura Research assumptions, with Knight Frank's figure that nearly 50 per cent of prime central London purchases are unleveraged. The central argument, that the cost of capital is an underwriting input rather than a timing signal, is unchanged, but the Lake Como supply case now rests on a permitting constraint rather than a statutory prohibition.

A further review on the same day refined some of the corrections above and fixed further points. The article stated that CBRE deferred its forecast of 5 to 15 basis points of cap-rate compression to 2027; CBRE's Midyear Review 2026 in fact projects cap rates to hold largely stable for the remainder of 2026 and expects only incremental compression in 2027, without a figure, and the stat, the takeaway and the relevant paragraph now say so (CBRE, U.S. Real Estate Market Outlook Midyear Review 2026). The article described the Savills World Cities figures for H1 2026 as prime yield data; Savills, as reported by PNI News, gives average capital values up 0.6 per cent and rents up 1.1 per cent, not yields, and the section now describes them as value and rent data, with the yield reading presented as an inference from those two figures (Savills World Cities H1 2026, as reported by PNI News). The central argument is unchanged.