The Income Hurdle

The Divergence Nobody Is Writing

The World Gold Council's Gold Demand Trends for the second quarter of 2026, published on 30 July, contains the two numbers above in the same table, one line apart. Central banks and other official institutions took 288.9 tonnes, up from a revised 56.5 tonnes in the first quarter and 177.9 tonnes a year earlier. Exchange traded funds and similar products recorded net outflows of 44.8 tonnes, against inflows of 171.1 tonnes in the second quarter of 2025 and 62.4 tonnes in the first quarter of 2026. It is rare for the official and private legs of a market to diverge this cleanly.

The Council's own explanation for the private selling is worth reading carefully, because it is not a story about gold. Outflows were, in its words, a response to weaker gold prices and, particularly in North America, upward adjustments to both inflation and interest rate expectations alongside a strengthening US dollar. That is a description of an opportunity cost problem. Gold pays nothing. When the return available on a liquid, default remote alternative rises, the price a rational investor will pay for an asset that yields nothing has to fall. Private investors behaved exactly as portfolio theory says they should.

Central banks did not, because they are not solving the same problem. A reserve manager is not maximising a risk adjusted return over a five year horizon. They are holding an asset that carries no counterparty, cannot be frozen by a foreign jurisdiction, and does not depend on the solvency of any government. The yield forgone is the premium on an insurance policy, and the policy has become more attractive for reasons that have nothing to do with the discount rate. Reading official buying as a bullish price signal, as a great deal of commentary did through 2026, confuses an insurance decision with an investment decision.

The price itself has already registered the shift. The LBMA gold price averaged US$4,506.29 an ounce in the second quarter of 2026. That was 8% below the first quarter average of US$4,872.90, and still 37% above the second quarter of 2025. In other words, gold has had an extraordinary two years and then given back a meaningful part of the most recent leg. Anyone writing in August 2026 that gold keeps making new highs is describing January, not the present.

The consumer side confirms the same reading from a different direction. Jewellery consumption fell to 278.2 tonnes, the lowest quarterly volume since the pandemic and 17% down year on year, as high prices and broader inflationary pressure constrained affordability. Total demand including over the counter activity was flat year on year at 1,269 tonnes. This is not a market in the grip of a debasement panic. It is a market in which one specialised buyer with a non financial objective is absorbing supply that price sensitive buyers no longer want at these levels.

None of that makes gold a bad holding. It makes it a specific kind of holding, with a specific cost, in a specific regime. The regime is what most commentary gets wrong.

What the COFER Data Actually Says

The single most repeated claim in hard asset marketing is that the dollar is losing its reserve status and that this justifies owning things. The claim is testable. The International Monetary Fund publishes the Currency Composition of Official Foreign Exchange Reserves quarterly, and the data brief covering the first quarter of 2026 was released on 1 July.

The dollar's share of allocated foreign exchange reserves rose to 57.13% in the first quarter of 2026, from 56.42% in the fourth quarter of 2025. The euro's share fell to 20.03% from 20.38%. The renminbi edged up to 1.99% from 1.95%, a level that after more than a decade of internationalisation efforts remains a rounding error against the dollar. The yen recorded the largest decline of any major currency, from 5.84% to 5.44%.

The residual category is the most interesting line in the release. The share held in currencies not separately identified in COFER, which is where any genuine diversification away from the incumbent reserve currencies would first appear, fell to 6.18% from 6.25%. The IMF notes that this was the first decrease after seven consecutive quarters of increase beginning in the first quarter of 2024. The trend that de-dollarisation advocates had been pointing to for two years reversed.

The IMF also addresses, in one sentence, the statistic that generated more headlines than any other in 2025. Its words are worth quoting exactly: "In 2025, gold surpassed US Treasuries as a share of official reserves, but that development was driven almost entirely by gold price valuation effects and is not reflected in COFER's dollar share, which has remained broadly stable." The crossover was real. It was also a consequence of gold rising in price rather than of reserve managers selling Treasuries to buy bullion. The Fund adds that the dollar's mild appreciation against major currencies accounted for around half of the increase in its own share this quarter, which is an equally honest caveat in the other direction.

What survives is a narrower and more defensible proposition. De-dollarisation is a real and advancing story in payments plumbing, in bilateral local currency settlement corridors, and in the construction of alternatives to correspondent banking. It is not, on the primary evidence, a story about reserve composition.

There is a genuine tension worth naming here, because belief and measurement have diverged. UBS surveyed 307 family offices across more than 30 markets between January and March 2026, representing US$627.4 billion of family wealth, and found that 65% expect confidence in the US dollar's reserve status to weaken. That is a striking number. It is also a statement about expectations rather than an observation about holdings, and the holdings data currently points the other way. An allocator can reasonably act on the expectation. What an allocator cannot do is claim the expectation has already been validated by the data, because it has not.

One Hundred and Twenty Six Years of Gold

If the case for hard assets does not rest on an observable collapse in dollar credibility, the obvious fallback is the long run record. That record exists, it is unusually well constructed, and it is not flattering.

The UBS Global Investment Returns Yearbook, compiled by Elroy Dimson, Paul Marsh and Mike Staunton, now covers 126 years across 35 markets and is the closest thing the industry has to a definitive long run dataset. The 2026 edition reports that since 1900 the real US dollar gold price has risen 5.2 fold. Compounded over that period, that is an annualised real return of approximately 1.3%.

Set that against the alternatives from the same dataset. US equities returned 6.6% a year in real terms between 1900 and 2025. Long bonds returned 1.6% real, and Treasury bills 0.5% real. One dollar invested in US equities in 1900 became US$3,296 in real, inflation adjusted terms by the end of 2025. The same dollar in gold became US$5.20. The gap is not a matter of a decade of poor timing. It is 126 years of compounding at a rate that barely exceeds the return on short dated government paper.

The Yearbook is also direct about the inflation hedging claim, which is the specific function most hard asset arguments assign to gold. Its authors find that the relationship between gold and inflation is weak. Of the 28 years in which inflation exceeded 3%, gold returns were negative in 13 of them. That is close to a coin toss. Gold has preserved purchasing power over the very long run, but the Yearbook's own framing is that it has been an inconsistent short term inflation hedge, which is an argument for realistic expectations of defensive assets rather than an argument against holding them.

The Yearbook concedes one important qualification, and it should be stated rather than buried. Over the 54 years since the collapse of Bretton Woods, annualised real gold returns were considerably higher: 4.7% in US dollars, 5.8% in sterling and 4.3% in Swiss francs. An investor who treats the post 1971 monetary regime as the relevant sample has a materially stronger case. That view depends entirely on the regime persisting, which is a bet on politics rather than an inference from data.

The honest summary is that gold is insurance with a running cost, and the running cost is the return forgone. That is a perfectly respectable thing to own, at a sensible weight, for the reason central banks own it. It is not a growth asset, it is not a reliable inflation hedge over investment horizons that matter to a living human being, and it is not a substitute for an asset that produces income. The interesting question is what a real asset has to deliver in 2026 to be worth owning at all, and that question turns on a number that has moved further than almost anything else in the past three years.

The Hurdle Rate Has Moved

The Federal Reserve's H.15 release of 18 August 2026 records the market yield on ten year inflation indexed Treasury securities at 2.44%. That is the real yield: the return above realised US inflation, in a liquid, daily priced instrument, with no operational drag, no manager, no crew, no insurance renewal and no residual value assumption. The five year real yield was 2.13%. The thirty year real yield was 3.06%.

It is worth pausing on how unusual this is. Between January 2013 and December 2022, a full decade, the ten year real yield never once exceeded 1.74%. For most of that period it was negative. The entire architecture of the post crisis alternatives industry, including the case for illiquid real assets, was built in a world where the risk free real return was somewhere between zero and slightly negative and where any positive real yield therefore looked attractive by comparison. That world ended. On the current level, an investor doubles their real purchasing power in roughly 29 years by doing nothing more strenuous than buying an inflation linked government bond and holding it.

The nominal picture is equally striking. The same release puts the thirty year Treasury constant maturity yield at 5.31%. The last time that yield closed at or above that level was 12 June 2007, nineteen years ago. The ten year stood at 4.72% and the effective federal funds rate at 3.63%. This is not a debasement regime. It is a high real rate, sticky inflation regime, which is historically the least hospitable environment there is for assets that do not produce cash, and a demanding one for assets that do.

Every illiquid real asset now has to clear that bar twice. First it has to beat 2.44% real after every cost of ownership, which for an operating asset means after management, maintenance, insurance, crew, fuel, berthing and depreciation. Second it has to pay the holder something additional for the fact that the position cannot be exited on a Tuesday afternoon at a screen price. What that second premium should be is contested, but it is plainly not zero, and in 2021 it was frequently priced as though the first bar were zero as well.

This is where the argument for income becomes stronger rather than weaker, and where a genuinely independent institution supports it. CBRE's US Real Estate Market Outlook Midyear Review, published on 4 August 2026, revised its own forecast after the geopolitical shock of the first quarter. Having expected the ten year Treasury to fall below 4% by year end, CBRE now expects it to remain above 4%, and having expected capitalisation rates to compress, it now projects them to hold largely stable. Its conclusion is the sentence that matters: persistently higher benchmark rates will mean that income will be the primary driver of total returns.

That is the whole argument, stated by a firm with no interest in making it on anyone else's behalf. In a high real rate environment, the capital appreciation leg of the real asset case stops working, because appreciation in real assets is substantially a function of falling discount rates. What is left is the income leg. And if the income leg is all that is left, then the only real assets worth serious consideration are the ones where the income is contracted, visible and durable enough to be underwritten rather than hoped for.

The Awkward Evidence

There is a large and inconvenient problem with the argument just made, and it comes from the most sophisticated pool of private capital in the world.

The UBS Global Family Office Report 2026, published on 28 May, surveyed 307 family offices across more than 30 markets, with an average family net worth of US$2.7 billion and average assets under management of US$1.3 billion per office. For the first time, 60% of them plan changes to their strategic asset allocation over the following twelve months, the highest level UBS has ever recorded. The direction of that change is stated plainly in the release: allocations are gradually tilting towards emerging market equities and alternatives such as infrastructure, alongside reduced exposure to real estate.

Read that again in the context of the preceding section. The most established, most liquid, best understood cash flowing real asset in existence is the one this cohort is reducing. They are not reducing it because they have stopped believing in income. They are reducing it while simultaneously increasing exposure to infrastructure, which 37% of respondents identify as a target theme alongside power and resources at 37%. Infrastructure is also a cash flowing real asset. So the distinction they are drawing is not between income and no income. It is a distinction within income assets, and it is worth being precise about what it is.

Infrastructure, at institutional scale, typically offers revenue that is contracted or regulated over long horizons, frequently with explicit inflation linkage, from counterparties with public or utility grade credit, and with an operating cost base that is a modest fraction of revenue. Real estate offers revenue that is contracted over shorter horizons, with tenant credit that varies enormously, in an asset that is rate sensitive on both the financing and the valuation side, and that requires continuous capital expenditure. When the discount rate rises and stays risen, the first category holds up considerably better than the second.

An operating maritime asset sits, on most of these dimensions, closer to the second category than the first. It is operationally intensive. Its revenue is seasonal and, at the individual vessel level, concentrated. Its cost base is exposed to fuel, crew, insurance and berthing, all of which have risen. Its residual value is an assumption rather than an observable. Anyone advancing a maritime income thesis in 2026 who does not begin by conceding that the largest family office survey in the world is moving away from the category it most resembles is not arguing in good faith.

The broader luxury asset record does not rescue the position either. Knight Frank's Luxury Investment Index closed 2025 down 0.4%, stabilising after two consecutive years of decline. Over the full decade the index has risen 38.6%, which is approximately 3.3% a year in nominal terms and, against the cumulative inflation of that decade, close to flat or negative in real terms. Within it, the Liv-ex Fine Wine 100 fell 2.5% in 2025 and is down almost 25% from its 2022 peak. Ten categories of scarce, storable, globally traded luxury assets, diversified across art, watches, cars, wine, jewellery and the rest, produced roughly nothing in real terms over ten years. Scarcity by itself does not generate a return.

So the burden of proof sits squarely on the operator, not on the asset class. The correct conclusion is not that the income thesis is wrong. It is that the income thesis is only as good as the specific contractual and operating structure underneath it, and that an allocator should demand evidence at that level rather than at the level of the category.

The Test an Allocator Should Apply

What follows from all of this is a short and unglamorous checklist, and it is considerably more useful than any view on the dollar.

The first question is whether the cash flow is contracted, and if so, by whom. There is a large difference between an asset that might generate revenue in a good season and an asset whose revenue is the subject of a written obligation from an identifiable counterparty with a balance sheet. If the answer is contracted, the next question is immediately about that counterparty, because a contracted yield is worth precisely what the obligor is worth. A yield commitment made to a fund is a claim on the operator, and the credit analysis of the operator is therefore the investment.

The second question is the operating cost base, and who bears it. This is where most real asset pitches quietly fail. IYC, which describes itself as the world's largest charter fleet manager, published its first half 2026 review with an unusually candid passage: rising operating costs are influencing yacht selection, and higher fuel prices, provisions and advance provisioning allowance, together with uncertainty stemming from geopolitical tensions in the Middle East, are encouraging some clients to favour shorter, more localised itineraries and yachts with lower weekly charter rates. In the same piece, IYC notes that the global fleet of yachts over 20 metres now exceeds 2,300 vessels and is growing at approximately 6.5% a year, and that this increased supply means more competition. Rising input costs, price sensitive customers and expanding supply is a textbook description of margin pressure, and it comes from the largest participant in the market.

The third question is what the residual value assumption is doing to the return. An operating vessel is a depreciating asset. If the projected internal rate of return depends on a terminal value that assumes the asset holds its price, the analysis is really an appreciation thesis wearing an income thesis as a disguise, and section four of this argument applies to it in full.

The fourth question is arithmetic. Take the net figure after every cost, apply a realistic residual, and compare it with 2.44% real. If the margin over that is thin, the position is not being paid for its illiquidity, and the allocator is better off in the inflation linked bond. If the margin is wide, the next question is why, and the answer had better be an identifiable structural feature rather than optimism.

We built HelmShare around the view that this is the only test that matters in the current regime, and that a structure should be legible enough for an investor to run it themselves. HelmShare Prime Fund, L.P. is a Cayman Islands exempted limited partnership issuing limited partnership interests. Its general partner and investment manager is HelmShare LLC, regulated by the Dubai Financial Services Authority in the DIFC as a Category 3C firm. Interests are offered under Regulation S outside the United States, to professional, qualified and high net worth investors in the EU and EEA, the United Kingdom and the Gulf. US persons are excluded. The fleet operator, Navigare, has given a contractual 8% yield commitment to the Fund, which is a commitment to the Fund and not a promise to any investor. Returns are targeted rather than assured, capital is at risk, and the value of an investment can fall as well as rise.

That structure is one answer to the question this article poses. It is not the only one. Infrastructure, contracted energy, ground leases and long duration inflation linked credit are all rational responses to the same analysis, and several of them are more liquid. What the current regime does not tolerate is a real asset held for its scarcity, its story or its supposed protection against a debasement that the reserve data does not show. The hurdle is 2.44% real, in public, every day, and it does not care what the asset is made of.