Investment Strategy Outlook · Q3 2026 — Tightening Into Turbulence

On 16 September the Federal Reserve raised interest rates for the first time since July 2023, unanimously, and told us that it expects to stay at 4.1 per cent through the end of next year. Its new Chair put the reasoning plainly: inflation is too high and has been for too long, financial conditions are not restrictive, and so a dose of accommodation was removed. Sixteen of eighteen participants pencilled in another hike this year.

In May we gave our base case — higher for longer, no cuts, growth near 1½ per cent, no recession — a fifty-five per cent weight. The economy has delivered most of it and then some: payrolls re-accelerated, purchasing-manager indices sit comfortably above fifty, and the Fed has tightened rather than merely held. What we did not have in the base case was Brent at $106 seven months into a war, the Saudi East-West pipeline out of service, and the US strategic reserve at its lowest since 1982.

Markets have absorbed all of this with a composure that is either impressive or alarming. The S&P 500 is within three per cent of a record set in August. High-yield spreads sit at 276 basis points. The VIX closed the Fed decision at 17.7. Equity investors are pricing an earnings machine — second-quarter profits grew 52 per cent, the fastest since 2021 — and, on the numbers, they are right to. Bond investors are pricing a Fed that means what it says: the ten-year closed above 5 per cent for the first time since 2007.

We owe you an account of our own record, and it is below. Of the ten assessments we recorded in May, five played out as expected — two of them emphatically: emerging-market equities, which we overweighted and which have returned 24 per cent, and the view that the rates regime had shifted. Two of the five could not be held because the shelf offered no vehicle — industrial metals and listed infrastructure — and fixing that is our first priority. Three played out only in part. Two did not: we overweighted investment-grade credit into rising yields, and precious metals, where gold is lower. The supply-side view on gold recorded in the May minutes has been borne out.

Three things distinguish this edition. We show conviction as well as direction — where all five of us agree, where a consensus holds, and where, on eight of thirteen asset classes, we are divided. We set our views beside those of nine of the world’s largest managers and against the evidence of where family offices, sovereign funds and central banks are actually moving money. And we have re-specified the benchmarks against which we judge our own portfolios to match what each sleeve actually holds.

The Investment Committee · Vault Wealth · Abu Dhabi · 17 September 2026

1. 01The hike is a regime statement, not a one-off.A 12–0 vote, a median dot at 4.1% for both 2026 and 2027, and “inflation risks to the upside.” Futures price a December hike at roughly 88%. Cash and the front end now yield more than most bond indices have returned this year.
2. 02Markets are pricing an earnings machine, not an economy.S&P 500 profits grew 52% in Q2 and are forecast +32% for 2026, driven by $720–745bn of hyperscaler capex. The forward multiple has fallen to 19.1× while the index rose — the market is cheaper than in June at a higher level. That is also the concentration risk: ten stocks are 38% of the index.
3. 03The oil shock is the inflation problem.Brent is up 74% this year; gasoline is up 27% year on year; the IEA sees demand falling 2.5mb/d in 2026 — the largest supply disruption in the oil market's history, in Goldman Sachs's words. It is why the ECB hiked to 2.50% and why the Fed sees no path back to 2% without acting.
4. 04Our highest convictions are unanimous and unchanged: emerging-market equities and infrastructure.Both played out as assessed in May. BlackRock returned to an EM overweight on 14 September; State Street bought real assets a day later. Our clearest underweight is high-yield and private credit, where PIMCO now says “the credit loss cycle is upon us.”
5. 05Where we are divided, so is the field.US equities, duration and investment-grade credit split our Committee two-two-one — and split the nine houses the same way. Gold divided us in May; it no longer does. The full House View is in Part IV.

The House View at a glance — thirteen asset classes, six to twelve months forward.

Horizon

Views are tactical, six to twelve months forward, set against the strategic allocation bands in each client's Investment Policy Statement. They are not price targets.

Conviction, not just direction

Every view carries one of three tiers. High conviction — the Committee was unanimous. House view — a consensus with no member opposed. Divided — members disagreed; we show the lean and summarise the debate rather than manufacture a view.

Data

Market data as at the close of Tuesday 16 September 2026 — the first close after the Federal Reserve's decision — unless stated. Index, fund and portfolio returns are at the 31 August 2026 month-end, total return in USD, from index providers and fund issuers.

Process

The views are those of Vault's Investment Committee — five voting members, including two independent external experts — recorded in a structured poll before the quarterly meeting on 7 September and debated in it. The method is set out at the end of this page.

The year divides into three acts. A January peak and a war-driven correction; a ceasefire that let equities recover to records on an earnings boom; and, since mid-August, the lapse of that ceasefire, a re-acceleration in labour, and central banks tightening into an oil shock.

Act I · Peak, then war — January to March.

Act II · Ceasefire and recovery — April to July.

Act III · Tightening into turbulence — August to September.

Total return in US dollars, 1 January to 31 August 2026. The ranking is the story: everything that benefited from a supply shock or the AI supply chain sits at the top; everything with duration sits at the bottom.

Each assessment the Committee recorded on 19 May, set against the returns of the funds held at 31 August and market levels at 16 September. We publish this review because a house view is only worth reading if its author keeps score.

Gold prices often fall when the USD strengthens during geopolitical shocks. The Iran war has been no exception — especially as gold had become a crowded trade.

The target range rose to 3.75–4.00%, unanimously, with projections that keep policy at 4.1% through 2027. The statement dropped its reference to supply shocks and said the action “will support a timelier return” to 2%. The Fed no longer regards oil-driven inflation as something to look through.

What changed in the dots. End-2026 from 3.8% to 4.1%; end-2027 from 3.6% to 4.1% — every cut pencilled for next year removed. Sixteen of eighteen see at least one more hike this year; four see two. The longer-run dot is 3.2%.

What the market did. Two-year +7bp to 4.74%; ten-year above 5%; the dollar’s best day since June. Gold −0.7%, Brent −2.7%. S&P 500 −0.45%, Dow −1.2%, Nasdaq flat. Futures: 51% for October, roughly 88% for at least one more hike by December.

Fed funds (median)

4.1% end-2026 (June: 3.8) · 4.1% end-2027 (June: 3.6) · longer run 3.2% (June: 3.1)

Core PCE

3.4% for 2026 · 2.5% for 2027

Unemployment

4.1% for 2026 · 4.1% for 2027 · 4.2% longer run

Real GDP

2.3% for 2026 · 2.4% for 2027 · 2.0% longer run

The plain fact is that inflation is too high and has been for too long. I would be hard-pressed to describe broad financial conditions as restrictive… so we removed a dose of accommodation.

The two-year has risen 127 basis points this year to 4.74%; the thirty-year 51, to 5.35%. A curve that was 71bp steep in December is 27bp steep today — a bear-flattener: the front end is repricing the Fed, while the long end is pricing a term premium that has been rebuilding all year. The thirty-year touched 5.3% in August, its highest since 2007, before the Fed had moved at all; Treasury doubled its long-dated buybacks on 19 August in response.

The real yield is the story. Ten-year TIPS yield 2.62% — a real return on the risk-free asset that most of this decade did not offer. That is the number that drives gold’s weakness, the dollar’s strength and the equity multiple’s ceiling.

The Gulf follows the Fed by construction: SAMA and the CBUAE matched the hike within a day, taking the Saudi repo to 4.50% and the UAE base rate to 3.90%. For dollar-pegged economies running an oil windfall, that is a tightening at the wrong moment for domestic credit — and a reason GCC equities divided the Committee. Brazil, by contrast, is cutting.

August CPI printed 3.4% headline and 2.4% core. Strip out energy and the picture is close to target; include it and the Fed is a full point off. The Chair’s judgement is that a shock in its seventh month, with unemployment at 4.1% and payrolls re-accelerating, has stopped being a shock and started being an inflation rate.

• The energy pass-through is broadening. Warsh: “Too many categories are still posting increases above 3 percent, on both a 6- and 12-month basis.” Airline fares rose 2.7% in the month; shelter is still running 3%. The Fed’s fear is the second-round effect, not the first.
• Labour has re-accelerated, not softened. The 28 August benchmark revision (−79k) was widely expected to confirm a weakening trend. A week later payrolls beat consensus three times over, with upward revisions. Job openings and hours are rising; claims are at full-employment levels.
• Europe faces the same shock with less cushion. The ECB raised its deposit rate to 2.50% on 10 September — “a no-brainer,” in Lagarde’s word — with staff seeing inflation “well above target for an extended period”: 3.0% this year, 2.5% next. Euro-area growth is forecast at 0.9%.
• The household is the weak link. Consumer sentiment at 47.8 is recession-level while the hard data is not. Real wages are barely positive (earnings +3.1% against 3.4% inflation) and the saving rate is 3.0%. This is where a hiking cycle would bite first.

S&P 500 earnings grew 52% in the second quarter — the fastest since 2021 — and are forecast to grow 32% for the year. The forward multiple has fallen from 20.4× to 19.1×, below its five-year average, while the index rose. The world economy is doing something quite different: the IMF cut 2026 global growth to 3.0%, with Saudi Arabia at 1.7% on production cuts.

World

IMF real GDP growth: 3.5% in 2025 · 3.0% in 2026 · 3.4% in 2027

United States

2.1% · 2.3% · 2.2%

Euro area

1.4% · 0.9% · 1.2%

China

5.0% · 4.6% · 4.1%

India

7.7% · 6.4% · 6.7%

Saudi Arabia

4.6% · 1.7% · 5.5%

Seven months in, the Strait of Hormuz runs at roughly nine per cent of its normal traffic and more than ten million barrels a day of Gulf production is shut in. The ceasefire lapsed on 17 August, fighting resumed on the 30th, and on 10–11 September drones struck Saudi Arabia’s East-West pipeline — the route around the Strait — halting loadings at Yanbu. Two chokepoints and the land route between them are now contested at once: Hormuz and, since the Houthi seizure of Perim Island, Bab al-Mandeb, with Red Sea transits rerouted round the Cape.

Brent’s year, in five marks:

What the war means for investors based in the Gulf, as distinct from investors in oil. Production is cut, the pegs import the Fed’s hike, and two chokepoints are contested — yet sovereign balance sheets are strong, regional funds are deploying record sums abroad, and the Committee’s regional equity stance is a mild underweight, not an exit.

Twelve indicators the Committee watches, coloured for where they sit. The pattern is unusual: the real economy reads mid-expansion — PMIs above 54, leading indicators turning up, unemployment falling — while prices read late-cycle — spreads at the tights, real yields at seventeen-year highs, a central bank tightening. That combination has a name: an inflationary boom.

The Committee’s reading. This is not the late cycle of 2007 or 2019, where activity was rolling over into an inverted curve. Growth is re-accelerating on a capex boom, credit is flowing, and the curve is flat but positive. What makes it late-cycle is the price of everything: an equity risk premium near zero, credit spreads that PIMCO calls “complacency rather than strength,” and a Fed that has begun to withdraw the accommodation that got us here. Historically, expansions do not die of old age; they are ended by central banks responding to inflation. That process has started. It usually takes twelve to eighteen months to matter for earnings — which is why the Committee’s twelve-month view remains constructive on growth assets while its three-year concern is rising.

We are being paid to hold growth assets and paid to hold cash. We are not being paid to hold credit risk or long duration.

The Committee’s May probabilities carried forward, with the scenarios re-cut after the hike. What changed: “Fed holds” is no longer the base case — “Fed at a plateau” is. What did not change: the market is priced for the bull case and the macro is drifting toward the bear one.

Reading note. Probabilities are the Committee’s May weights, carried forward pending the December meeting; the scenario descriptions are updated to 16 September. A 55% base case with a 25% bear tail argues for holding growth exposure while owning explicit protection — cash, quality duration in modest size, and real assets — rather than reaching for the last leg of the credit cycle. Scenario market outcomes are illustrative, not forecasts.

The structural themes the Committee adopted in May, each marked to the 16 September close, with the investment implication the Committee draws from it.

What the largest pools of private and sovereign capital say they are doing, from their own 2026 surveys. The direction of travel matches the Committee’s two unanimous views — and runs against the asset class it has never owned. Family offices plan to take real estate from 11% to 8% of portfolios and cash from 9% to 8%, while gold goes from 2% to 3%, infrastructure from 1% to 2%, and EM equities from 5% to 6%.

The Committee read each firm’s most recent publication before forming its own view; every quotation was read at source. Where a house moved in September — BlackRock on emerging markets, State Street on gold and duration — we show the later position. These are the views of the named firms, not of Vault Wealth Limited.

The nine houses condensed to one view per asset class from their published outlooks, with the Committee in the final column. Reading across a row shows the field; reading down the last column shows us.

With the consensus. Emerging markets: six of nine houses are overweight, BlackRock having rejoined on 14 September — and we are unanimous. Gold: five houses favour it, including two that added in September; we are overweight. High yield and private credit: nobody is overweight, and the two most explicit voices — Amundi and PIMCO — are underweight, as we are.

Against it, or ahead of it. Infrastructure is where we hold the strongest view in the field: unanimous, against one house underweight and several silent. On US equities six houses are overweight; we are divided, which we regard as the honest reading of a 19× multiple set against a hiking Fed. On duration the field splits four-three-two — and so do we, two-one-two.

Six to twelve months forward. Tallies read overweight · neutral · underweight. First: equities, precious and industrial metals, and real assets.

Where the debate is sharpest. The Committee mirrors the institutional divide on duration and investment-grade credit, is clearest of all on high yield and private credit, and is most polarised on digital assets.

Changes since May. Developed ex-US: overweight to neutral. EM: strengthened to unanimous. High yield and private credit: hardened to underweight. Gold: the May split resolved to overweight. Industrial metals, IG credit, cash and digital assets held a stance in May and are now divided. Duration and GCC equities are polled for the first time.

The equity story of 2026 is a single theme — the AI capital-expenditure cycle — expressed very unevenly across regions. It made Korea the best-performing major market in the world at under five times earnings and left India and China negative on the year.

The Committee’s position. Emerging markets — overweight, unanimous. Eleven points ahead of developed markets, driven by exactly the earnings the US market is paying 22× for, at a fraction of the price. The risks are a stronger dollar on a Fed cycle and the 10 November truce deadline; both are watched. US — neutral, divided. The bull case is earnings: +32% this year, 63% of companies guiding up. The bear case is 38% of the index in ten names, an equity risk premium near zero, and a Fed that has started tightening. Equal-weight exposure at 17.8× is the Committee’s compromise. Developed ex-US — neutral. Japan is the AI supply chain’s developed-market expression; Europe faces the same oil shock as the US with a third of the growth. GCC — divided, leaning underweight. The dollar peg imports the hike; production cuts take 2026 growth to 1.7%. Regional assets are stronger than regional equities.

Stocks are supported by strong earnings due to the AI capex buildout, not by a strong economy.

Yields are the highest in a generation and spreads are among the tightest. That combination — a 5% ten-year, 80bp of investment-grade spread, 276bp of high yield — means the return on offer is almost entirely the risk-free rate. The Committee’s response is to take the rate and decline the spread.

The Committee’s position. Cash and the front end. Neutral on cash as a headline, but with a clear lean: the Fed has told you it holds at 4.1% through 2027, so one-to-three-year quality paper earns close to the policy rate with almost no duration risk. It was the only positive bond return of the year. Duration — divided, 2·1·2. The case for: a 5% ten-year and 2.6% real yield are the best entry points in seventeen years, and UBS, Julius Baer, Vanguard and PIMCO are all extending. The case against: the Fed is hiking, the term premium is rebuilding, and duration failed as a diversifier this year — BlackRock’s point. The Committee did not resolve it and says so. Investment grade — divided, 2·1·2. Quality is the one thing every house agrees on; the argument is over how much duration comes with it. High yield and private credit — underweight, house view. 276bp is the tight end of a twenty-year range; PIMCO’s shadow-default and payment-in-kind signals are the late-cycle markers; BDCs trade at discounts to NAV. We hold three private-credit funds and regard the sleeve as full.

The prospective Sharpe ratio of high quality fixed income now compares favorably with equities for the first time in many years.

Gold has record structural demand and a negative year. Copper has a record price on capex. Infrastructure is the fastest-growing allocation among sovereign funds and the Committee’s highest conviction. The common thread: in an inflationary boom, real assets tied to production have outperformed real assets tied to fear.

Gold’s year, in four marks:

The gold paradox. Central banks bought 289t in Q2, a record for a second quarter; ETFs added $18bn in August, the second-largest month ever, taking holdings to a record 4,189t; 45% of reserve managers say they will buy more. And the price is down 1% on the year and 24% from its January high. The explanation is the rates page: a 2.6% real yield and a dollar at seven-week highs. The Committee’s overweight is a view that the structural bid outlasts the cyclical headwind — held with more humility than in May.

The Committee’s position. Infrastructure — overweight, unanimous. Three members highly so. The AI, grid and energy-security capex cycle is the most durable earnings stream in the market and the allocation of choice for sovereign funds and family offices alike. Our constraint is expression: one private fund and no listed vehicle. Industrial metals — overweight, divided 3·1·1. As assessed since May; not held, for the same reason. Gold — overweight, house view. Held for what it does in a crisis, which this year it has not yet been asked to do. Private equity — neutral, divided. Structurally endorsed; tactically patient while exits are shut. Digital assets — underweight, divided. Bitcoin −14% YTD; neither a risk asset nor a hedge in 2026.

Composite returns of Vault’s twelve core profiles — VHVE, JPGL, JREG, JREM, JRUB, FLUC and JPSA (conventional); ISDW, SPUS, SPWO, ISDE, HBKU and SKUK (Shariah) — at policy weights, from verified issuer NAV total returns, measured against the stated benchmarks.

Reading the tables. Outperformance widens with equity weight this year and over one year — the active and multi-factor layers are earning their fee — and inverts over three years at the equity end, so this year repairs a deficit rather than extending a lead. The Shariah profiles finished narrowly behind at both horizons: they are fully passive, and one holding tracks a 50/50 ex-US/EM index, so true EM exposure is roughly 15% against a 5% policy.

A view is only as good as the conditions under which its author would abandon it. These are ours, mapped to the calendar.

18 Sep

Bank of Japan decision. Held at ~1.0% in July, 8–1; a hike would tighten global liquidity further.

20 Sep

Bessent–He Lifeng talks, New York. Tariff cuts on energy, agriculture and manufacturing inputs under discussion.

24 Sep

Trump–Xi summit, Washington (China unconfirmed). A truce extension past 10 Nov is the bull-case trigger.

25 Sep

Swiss National Bank. At 0.00% since June 2025.

4 Oct

OPEC+ meeting. Core-7 held October output on 6 Sep; Saudi producing 6mb/d against 12 capacity.

~10 Oct

September CPI. First print after the hike; core 2.4% in August.

28–29 Oct

FOMC. Hike priced at 51%; GSAM expects a skip for the mid-terms.

3 Nov

US mid-term elections. Fund managers see a split Congress; a Democratic sweep read as “yields up, stocks down.”

10 Nov

US–China trade truce expires. Section 301 duties of 10–12.5% would escalate without extension.

9–10 Dec

FOMC. Second hike ~88% priced; new SEP; the first test of “plateau vs cycle.”

Dec

Vault Investment Committee. Q4 meeting; probabilities re-weighted; House View refreshed.

The Committee. Vault’s Investment Committee is constituted at Vault Wealth Holdings Limited and serves the group’s regulated operating companies. It has five voting members: the Chief Executive (Chair), the Chief Operating Officer, the Investment Strategy Manager (Secretary), and two independent external experts — a former bank chief investment officer and the co-founder of a Paris asset manager. It meets quarterly; minutes are signed by the Chair and retained permanently.

The poll. Before each meeting every member records a view on thirteen asset classes on a five-point scale — highly overweight, overweight, neutral, underweight, highly underweight — relative to a representative allocation within the strategic bands of the Investment Policy Statement. Duration and credit quality are polled separately from headline fixed income. Responses are collected through a structured form, aggregated by the Secretary, and debated in the meeting; the debate can change a member’s recorded view.

Conviction tiers. High conviction — all five members on the same side. House view — a consensus with no member on the opposite side. Divided — at least one member overweight and at least one underweight; we publish the lean and the tally rather than manufacture a view. The published stance for a divided row is neutral where the split is even, and the majority’s lean otherwise.

Scorekeeping. Every stance is recorded in the minutes with the date and the market level at the time. At the following meeting each is reviewed against actual outcomes — using the returns of the funds actually held wherever possible, not index proxies — and the review is published (Part I). An assessment that played out but could not be held is recorded as such; it is an operational gap, not a forecasting success.

Benchmarks. Core portfolios are measured against fixed-weight composites of each sleeve’s stated index — the bond sleeve against 70% Bloomberg US Corporate and 30% ICE BofA 3-month T-bills, matching its actual duration; the equity sleeve against a fixed 90/10 MSCI World/EM blend, matching policy. Composites are computed at static policy weights from issuer-reported NAV total returns and are gross of Vault’s advisory fee. The framework was re-specified in September 2026 so that each sleeve is measured against a composite matching its actual composition.

Institutional benchmarking. The nine houses are chosen for scale and for publishing a dated, attributable house view. Positions are condensed from the most recent publication by rule — “constructive / favour / attractive” to overweight, “cautious / reduce” to underweight — and every quotation is read at the source document before it is used. Where a house has not expressed a view, none is inferred.

Data. Market data as at the close of 16 September 2026 unless stated; index and fund returns at the 31 August 2026 month-end. Primary sources throughout: central banks, statistical agencies, index providers, fund issuers, the IEA, IMF and World Gold Council, and the publishing institutions of the surveys cited.

This document has been prepared by Vault Wealth Limited (“Vault”), a company incorporated in the Abu Dhabi Global Market (“ADGM”) and regulated by the Financial Services Regulatory Authority (“FSRA”) with a Category 4 licence for advising on investments and credit, and arranging deals with retail endorsement (ADGM registration 000009540). It is intended for informational purposes only. It reflects the views of Vault’s Investment Committee as at the date shown and does not constitute investment advice, a personal recommendation, or an offer or solicitation to buy or sell any security, fund or financial instrument. It should not be relied upon for making investment decisions.

Views expressed are tactical, six to twelve months forward, and may change without notice. The value of investments and the income from them can fall as well as rise, and investors may not recover the amount invested. Past performance, including the portfolio and index returns shown, is not a reliable indicator of future results. Portfolio composites are computed at static policy weights from issuer-reported fund returns and are gross of Vault’s advisory fee; individual client returns will differ. Scenario outcomes are illustrative and are not forecasts. Index and market data are from the sources named alongside each figure and are believed reliable but are not guaranteed.

Where this document refers to the views of third-party asset managers, those views are the property of the firms named, are summarised or quoted from their published materials, and are not endorsed by Vault. Where it refers to surveys of allocators, the findings are those of the publishing institutions. This document is directed only at persons to whom it may lawfully be communicated and is not intended for distribution in any jurisdiction where such distribution would be unlawful. Vault adheres to the Global Investment Performance Standards (GIPS) when presenting information relating to past performance; the GIPS standards do not guarantee the accuracy or completeness of the information presented.

Next publication: Q4 2026 Investment Strategy Outlook — following the Investment Committee’s December meeting.

Reviewed by Bilal Abou-Diab, CFA