Family Offices: Reading the Real Estate Pullback

What does the 2026 UBS pullback mean for prime resort property?#

The UBS Global Family Office Report 2026 sets the frame for that question without measuring resort property directly. Real estate held 11 per cent of the average family office's strategic asset allocation in 2025. Among the family offices now planning to change that allocation, a record 60 per cent of the 307 surveyed, the target for 2026 is 8 per cent. That is a three-point cut on a base already down from 14 per cent in 2019. The report was published 28 May 2026, drawn from a survey of 307 family offices across more than 30 markets, fielded between 22 January and 30 March 2026 (UBS Global Family Office Report 2026).

The number is a plan, not a transaction. UBS asks family offices what they intend to do with their strategic allocation over the next 12 months, not what they have already sold or bought. Eight per cent is the average target of the offices planning changes. It is not a forecast for all family offices, and not a committed reallocation of every dollar currently in real estate. Read that way, the pullback is real and it is modest. It removes three points from a book that never approached the weight family offices give to developed-market equities, now 27 per cent of the average portfolio.

For a prime resort asset, the relevant reading is narrower still. UBS reports real estate as a single line in its strategic asset allocation table, with no breakdown by property type. It does not isolate branded residences, resort property or hospitality-linked real estate from an office tower or a warehouse portfolio. A three-point cut in the aggregate says nothing directly about appetite for the segment Victaura tracks. It says something about the share of family office portfolios that segment competes for, and about the direction that share is moving. That is the question this piece takes up: not whether family offices like real estate less, but what a smaller, more selectively allocated real estate book means for the buyers a prime resort developer can realistically expect to find.

Strategic asset allocation: what the UBS number actually measures#

Strategic asset allocation is the target weighting a family office sets for each asset class inside its investment portfolio. It is the policy line, not the trading line. A family office might tactically overweight cash for a quarter without touching its strategic target; the number UBS reports is the target, the anchor the office is meant to return to. Fixed income, equities, private equity, hedge funds, real estate, infrastructure, gold and a handful of other lines sum to 100 per cent of the portfolio UBS is asking about.

The portfolio in question is the one the family office allocates, not total family wealth. UBS does not publish a detailed perimeter for it; Victaura Research reads it as excluding the operating business, which 77 per cent of surveyed families have and whose main sector is real estate for 17 per cent of them, and the property a family occupies rather than invests (Victaura Research assumption; survey profile from UBS Global Family Office Report 2026, p. 60). An allocation of 11 per cent to real estate inside that perimeter says nothing about how much of a family's total net worth sits in property once a residence, a family compound or an operating hotel is counted. Victaura has drawn this distinction before, and it holds here: the two figures measure different things, and only one of them is what UBS actually surveys (see Family Offices and Real Estate).

Historical stability is what makes the 2026 plan notable. UBS has tracked whether family offices intend to change their strategic allocation since 2020, and until now the share reporting a planned change moved in a narrow band, from 25 per cent in 2021 to 37 per cent in 2023. The 60 per cent recorded this year has no precedent in the report's history. Whatever else is true of the 2026 numbers, the willingness to state an intended change is itself the largest single move in the series.

The chart below separates reported allocations from a stated plan. From 2019 to 2025 each column is the average allocation family offices reported in that year's UBS survey, 14 per cent at the start and 11 per cent at the end, with a rise in 2022 and a low of 10 per cent in 2023. Each year's figure comes from that year's respondents, not from a fixed panel. The dark 2026 column is the average target of the 60 per cent of offices planning changes; the 40 per cent planning none are not in it.

Reallocating, not retreating: what else the 2026 plan moves#

UBS's table also shows what else the offices planning changes intend to move. Among the offices planning 2026 changes, developed-market equities hold at 27 per cent, essentially unchanged from 2025. Emerging-market equities move from 5 to 6 per cent. Infrastructure, still a small line, rises from 1 to 2 per cent. Gold, reported separately from other precious metals for the first time this year, is planned to rise from 2 per cent of 2025 allocations to 3 per cent among those changing course. Real estate is the only line UBS's text singles out as being pared back, and at three points it is the largest planned cut in the table; the only other line that falls is cash, from 9 to 8 per cent.

None of these are large moves in isolation. A one-point shift in gold or infrastructure may sit within the noise of a 307-respondent survey whose margin of error UBS does not publish. What is legible is the direction: UBS describes family offices planning changes as keeping developed markets as the anchor while tilting slightly toward emerging-market equities and selected alternatives such as infrastructure, and some describe gold as a long-term allocation or protection against geopolitical risk. Real estate, by contrast, is the asset class being trimmed; UBS does not say the cut pays for the other moves.

The reallocation reads as risk management, not disillusionment with property. UBS's own language describes a selective tilt rather than a wholesale reallocation, and the 41 per cent of the average portfolio still held in developed-market listed equities and fixed income, before counting any alternative asset, confirms family offices are not making a dramatic break from prior positioning. Real estate remains, at 11 per cent, a larger allocation than infrastructure, gold, commodities and art and antiques combined. The pullback is a rebalancing at the margin of an already-diversified book, not a signal that family offices have lost conviction in property as an asset class.

Which regions already carry the heaviest real estate weight?#

The 11 per cent global average conceals a wide regional spread, and the spread is the more useful number for assessing where institutional-grade buyers for prime property actually sit. UBS's 2025 regional cut shows United States family offices holding 20 per cent of their portfolio in real estate, the highest of any region surveyed. Switzerland follows at 12 per cent, one point above the global average. Europe, excluding Switzerland, sits at 11 per cent, in line with the global figure. Latin America and the Middle East both report 9 per cent, and Asia-Pacific, UBS's broadest catch-all region, reports 5 per cent, the lowest of the major regions surveyed. Within it, UBS reports North Asia at 6 per cent and Southeast Asia at 4 per cent.

The US figure sits far above the global average, but US family offices are only 12 per cent of respondents. Strip out the United States and the remaining regions, Latin America included, sit between 5 and 12 per cent, well below the 20 per cent US figure. UBS's commentary elsewhere in the report notes that US family offices raised the share of assets held in North America from 86 per cent in 2025 to 88 per cent in 2026, and its US spotlight shows only 21 per cent of US respondents planning to change their strategic allocation, against 60 per cent globally. Elsewhere the share planning changes runs from 43 per cent in Switzerland to 67 per cent in Europe and 82 per cent in the Middle East, the last on a base UBS itself calls low, and UBS does not say how much of those changes touches real estate. Nothing in the report suggests the 20 per cent US real estate figure is the one coming down fastest (UBS Global Family Office Report 2026, pp. 26, 46–54 and 60).

That asymmetry matters for a developer courting capital outside North America. A family office in Zurich, Milan or Dubai is, on this survey, already allocating a smaller share of its portfolio to real estate than its US counterpart, before any further cut is applied. UBS does not publish the 2026 plan broken out by region, a gap this piece flags rather than fills. The regional 2025 split and the global 2026 plan are two separate readings of the same survey, not a single number, and the two charts in this piece keep them as two series rather than blending them into one.

Does the family office fear a real estate correction?#

No. UBS asks family offices which of sixteen named risks concern them over the next 12 months and the next five years, and a real estate correction sits near the bottom of both lists. Eleven per cent of the 307 respondents name it a risk over the next 12 months, and 17 per cent over the next five years. Compare that with major geopolitical conflict, cited by 64 per cent at 12 months and 61 per cent at five years, a global trade war at 49 per cent, and a debt crisis rising from 31 per cent at 12 months to 56 per cent at five years. Real estate correction sits below climate change on the five-year list.

That ranking is the clearest evidence the pullback is not a crash call. If family offices believed prime and institutional real estate were about to reprice sharply, the correction risk would show up near the top of the list, alongside the debt-crisis and financial-market-crisis categories that do concern them. It does not. The three-point cut to the 2026 plan sits alongside a survey in which the risk attached to the asset class they are trimming ranks thirteenth of sixteen at 12 months and fifteenth of sixteen at five years.

The honest reading is that the cut is a portfolio-construction decision, not a valuation call. Family offices are, on this survey, trimming real estate while emerging-market equities, infrastructure and gold edge up, against concerns about currency, geopolitics and debt that have little directly to do with property pricing. UBS asks about a real estate correction in general, not about any property type, so the evidence shows family offices do not fear a broad correction and says nothing specific about prime resort pricing.

What does the marginal cut mean for the prime resort buyer base?#

A cut of the same number of points weighs more on a small real estate book than on a large one. The United States enters 2026 with 20 per cent of the average family office portfolio already in real estate. Switzerland, Europe, the Middle East and Asia-Pacific enter it with 12, 11, 9 and 5 per cent respectively. UBS does not publish the 2026 plan broken out by region, so the next step is an inference, not a measured fact, and it is flagged as one: if every region cut by the same three points, Asia-Pacific would lose three fifths of its real estate share and the United States less than a sixth, leaving the regions already below the global average the least room to hold, let alone add, a large-ticket commitment such as a resort residence or a branded hospitality asset.

The base outside the United States was already thin before this survey. Asia-Pacific's 5 per cent and the Middle East's 9 per cent describe portfolios where real estate, inclusive of every property type from an office building to a warehouse, is a single-digit slice; Asia-Pacific's 5 per cent is below the 8 per cent Swiss family offices hold in gold and other precious metals combined. A prime resort developer sourcing capital from family offices in these regions is already competing for a fraction of a fraction. A further pullback, even a modest three points at the global level, would narrow that fraction again if it reached these regions.

This does not mean the buyer disappears. Family office capital for prime property has never depended on the average allocation; it depends on the smaller number of offices with genuine conviction in the asset class and the operating capability to hold it well, a point Victaura Research has made elsewhere. What the regional data support is a narrower claim. Outside the United States, family offices gave real estate of every type between 5 and 12 per cent of their portfolios in 2025, against 20 per cent in the United States, and UBS does not say how much of that is resort property.

Reconciling the pullback with Victaura's operated-versus-held thesis#

Victaura Research argued in September that the family office real estate line is shrinking on the survey while its composition shifts from passively held property toward operating assets that produce income through the cycle (see Family Offices: The Move to Operated Prime). The UBS 2026 report, read here, does not contradict that argument. The 11 per cent recorded for 2025 and the 8 per cent planned for 2026 are consistent between the two readings; both pieces cite the same underlying survey table.

What this piece adds is the trend behind the single year, and the risk context around it. The earlier piece treated 11 per cent as flat against 2024. Set against 2019's 14 per cent, the line has fallen overall, with a rise in 2022, a dip to 10 per cent in 2023 and a recovery to 11 per cent in 2024 and 2025. The correction-risk data, not used in the earlier piece, shows the fall is not driven by fear of a property downturn. Together, the fuller trend and the risk ranking support a reading in which family offices are not fleeing real estate as an asset class; they are running a smaller, more selective book of it, and, on the September thesis, running more of what remains as an operating business rather than a static holding.

The two arguments point the same allocator toward the same conclusion. A shrinking, more selective family office real estate book raises the bar for what earns a place in it. An asset that produces income, is operated to institutional standard, and is underwritten as a business rather than a trophy has a better claim on a thinner allocation than one that does not.

Where competing institutional capital is heading instead#

Family offices are one buyer among several, and the wider institutional real estate market is not contracting the way the family office survey line is. Savills forecasts global real estate investment turnover to surpass US$1 trillion in 2026, up 15 per cent on 2025 and the first time the market has cleared that level since 2022, with the Americas the largest single region at a forecast US$570 billion and EMEA the fastest-growing at 22 per cent (Savills, Global Real Estate Outlook, industry estimate). On that forecast, published 10 December 2025, institutional capital broadly is expected to return to real estate at the same time the family office survey line is being trimmed at the margin.

Within the segment Victaura tracks specifically, growth has continued. Savills counted 764 branded residential schemes worldwide in December 2024 and projected 910 by the end of 2025, a 19 per cent rise in a single year and a near-tripling since 2015 (Savills, Annual Report: Branded Residences 2025/26, industry estimate). Savills notes that the 910 figure assumes all remaining 2025 pipeline projects complete. That growth is funded by a mix of developer balance sheets, hotel operators and private capital, of which family offices are one channel among several (Victaura Research assumption).

Separately, Knight Frank's Wealth Sizing Model 2026 counts 713,626 ultra-high-net-worth individuals worldwide in 2026, those with a net worth above US$30 million, up from 551,435 in 2021 (Knight Frank, Wealth Sizing Model 2026, model estimate). That count is made at the level of the individual, not the family office vehicle, and the two populations overlap only partially. Read together, the three sources describe forecast growth in institutional turnover, a growing branded residence supply and a larger UHNWI population even as the specific family office allocation line contracts, a reminder that the UBS survey describes one channel, not the whole pool a prime resort developer can draw from.

A plan, not a flow: reading the record 60 per cent#

Sixty per cent of the 307 family offices UBS surveyed say they plan to change their strategic asset allocation in the next 12 months, the highest share the report has recorded since it began asking the question in 2020. The prior high was 37 per cent, in 2023. In every prior year a majority said they did not plan changes; this year the majority flipped for the first time. That 25-point jump in the share of respondents is the largest single movement across the seven editions of the report, and it measures a different thing from the three-point shift in the real estate allocation.

A stated plan is not a completed transaction, and UBS is explicit that it surveys intent. UBS's table sets last year's plan for 2025, 10 per cent, beside this year's reported 2025 allocation, 11 per cent. The two come from different survey samples, so the gap does not prove that plans slipped, but it shows how far a stated target and a reported outcome can sit apart. A principal reading the 8 per cent figure for 2026 as a hard ceiling on family office real estate demand is reading more precision into a survey response than the survey supports.

What the 60 per cent figure does support is a genuine change in posture. Whether or not every planned reallocation executes exactly as stated, a record share of family offices actively revisiting their strategic weights, rather than holding a stable policy line, is itself informative. It signals a channel of capital more open to renegotiating its real estate exposure in 2026 than in any prior year UBS has measured, in either direction.

The weaknesses, honestly disclosed#

The central weakness of this piece is the regional inference in the buyer-base argument. UBS publishes the 2026 real estate plan only as a global figure and the regional split only for 2025. The claim that a cut of equal size in points would weigh most on already-thin regions is Victaura Research's extrapolation, not a number UBS reports. Family offices in different regions may cut, hold or raise their real estate weight by different amounts in 2026; nothing in the published survey rules out a scenario where the United States, not Europe or Asia-Pacific, carries the largest share of the reduction.

The survey itself carries sampling limits that apply to every figure cited here. Three hundred and seven respondents, drawn from UBS's own client base rather than a random sample of the world's family offices, produce a margin of error the report does not disclose; Victaura Research assumes it could be as wide as the one-to-three-point moves this piece discusses (Victaura Research assumption). A one-point shift in gold, infrastructure or emerging-market equities is plausibly within that margin. The direction across several years of reported data, from 14 per cent in 2019 to 11 per cent in 2025, is a more reliable signal than any single year's reading, and the 2026 plan points the same way.

The correction-risk data measures perception, not valuation. Eleven per cent citing a real estate correction as a 12-month risk describes what family offices say they worry about, not an independent assessment of whether prime or institutional real estate is fairly priced. A low fear reading is consistent with genuine confidence in current pricing and equally consistent with a risk respondents have not yet turned their attention to. Finally, the Savills and Knight Frank figures cited as context measure different populations, institutional investors generally and UHNW individuals, not family offices specifically, and should not be read as a direct offset to the UBS numbers.

What this means for the allocator and the principal#

The allocator's signal here is compositional and regional, not a verdict on the asset class. Family offices are not abandoning real estate. They are running a smaller, more selectively allocated book of it, trimmed by three points at the margin while developed-market equities hold at 27 per cent and emerging-market equities, infrastructure and gold edge up, against geopolitical and debt concerns that have little to do with property pricing directly. The channel is not shrinking out of fear; the risk ranking says the opposite.

For a principal assessing where family office capital for prime resort property is likely to be thinner, the regional split matters more than the global average. United States family offices carry a real estate weight roughly double the global figure and four times that of Asia-Pacific. A developer or operator courting capital outside North America should expect family offices there to give real estate a smaller share of each portfolio, a share the 2026 plan may narrow further, and should read UBS's global 8 per cent figure as a prompt for further regional scrutiny rather than as a regional forecast.

The prudent posture is to hold the plan and the flow apart. UBS's own table puts last year's plan and this year's reported allocation a point apart, measured on different samples. The 60 per cent of family offices now revisiting their strategic weights is real, and worth watching closely through 2026 and into next year's report; the 8 per cent target it points to is not yet a fact about where family office capital has gone.

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.