Lumix Lens — 01/06 to 05/06

Summary Table: Gold Macro Driver

Levels as at the close of the week ending 5 June 2026. The direction column reflects the trailing ninety-day change except where a thirty-day window is indicated. Source: Federal Reserve Economic Data (FRED) and COMEX. Compiled by the Lumix quantitative model.

News overview

3 June ADP National Employment reported private payrolls rose by 122,000 in May, beating the consensus of 110,000.

4 June President Trump was reported to want to maintain the ceasefire with Iran unless American troops are killed, easing fears of escalation.

5 June President Putin rejected President Zelenskyy's offer of a face-to-face meeting, calling it "boorish".

5 June Nonfarm payrolls: the United States economy added 172,000 jobs in May, exceeding the consensus estimate of 105,000.

Micro drivers

Gold

Gold's cyclical drivers turned against it throughout the week. The ten year real yield, measured by DFII10, stood at 2.11% on 4 June, close to its 52-week high of 2.18% set a year earlier, and has been grinding higher for three consecutive months, steadily eroding the carrying cost advantage of a non-yielding asset. The broad trade weighted dollar at 118.88 was modestly firmer over the month within a narrow 118 to 119.5 range. The Federal Reserve balance sheet, at US$6.711 trillion, expanded modestly after a month of quantitative tightening, a mild counterweight that is supportive at the margin but small relative to the size of the balance sheet. Policy remains restrictive, with rates at 3.50 to 3.75% against inflation near 2.4%, and the pause in cuts removes the primary bull case for gold that ran through the rate path.

Inflation expectations offered no offsetting support. The ten-year breakeven rate, T10YIE, at 2.36%, has declined steadily from a peak of 2.50% on 4 May, and the retreat over the month reduces the inflation hedge bid for the metal. Five years forward inflation, T5YIFR, at 2.24%, remains anchored in a 2.2 to 2.4% band, offering no panic driven premium.

The employment report then triggered a sweeping repricing of Federal Reserve expectations. Gold suffered its largest single session decline in months, falling 3.1% on Friday and ending the week near US$4,337 an ounce, down 4.90% over the five sessions. The selloff was amplified by a surging dollar, spiking Treasury yields and a collapse in rate cut expectations, with futures markets now pricing roughly a 58% probability of a rate hike by year end. The move raised the opportunity cost of holding a non-yielding asset and represents the textbook headwind reasserting itself over a shorter horizon. The structural offset that has defined the metal for two years, sustained official sector accumulation that is largely insensitive to price and rate, remains intact and continues to provide a demand floor beneath cyclical selloffs. The week is best read as the cyclical headwind overpowering the structural bid at the margin rather than a reversal of it.

Crude oil

The single most dominant factor for crude remains the Iran war, which Barron's has described as the worst supply shock in modern history. OPEC+ ministers met on Sunday 7 June to consider higher production quotas to cap prices, but the war has effectively crippled the organization's ability to influence the market, because the supply shock originates from a chokepoint it cannot control while its spare capacity sits in Saudi Arabia, the United Arab Emirates and Iraq. The meeting is better read as a nonevent than as a catalyst in either direction, since higher quotas on paper translate to nothing while the physical logistics of delivery remain paralysed. Prices pared recent gains after President Trump said that talks between the United States and Iran were ongoing, a reminder that any credible diplomatic resolution could trigger a rapid unwind of the geopolitical risk premium.

The supply side is unusually exposed. The United States Strategic Petroleum Reserve is on pace to reach its lowest level since the early 1980s later this month, with a further 8 million barrels withdrawn last week. The reserve has historically been the federal government's primary mechanism for capping price spikes during supply disruptions, so any additional shock, however minor, will now land in a market with no federal buffer beneath it and will amplify the price impact in a way not seen in four decades. Commercial crude inventories compound the problem, falling toward critical minimum operating levels as a sharp rise in export volumes depletes domestic stocks, which leaves the country without either its emergency reserve or the commercial cushion that would normally absorb a short-term shortfall.

Two market signals are worth tracking as forward indicators. Energy equities including Patterson UTI, ProPetro, Atlas Energy and Kosmos Energy sold off in the latest session despite elevated prices, which reads as profit taking after prior euphoria rather than a fundamental exit, rotation rather than capitulation, and should not be mistaken for a bearish structural signal. More telling is the tanker shipping fund BWET, which has surged more than 1,645% over the year, a leveraged bet on the Strait of Hormuz remaining disrupted for an extended period and the clearest real time barometer of how aggressively the market is pricing Hormuz risk. A reversal in BWET, which could occur within hours of a credible diplomatic development, would likely precede weakness in the front month's crude contract by days rather than weeks. The canary has not yet turned, but it bears watching closely.

United States equities: reversal or drawdown

The repricing met an equity market priced for perfection, and the collision produced the sharpest session of the year. As the week closed, the Nasdaq Composite fell 4.18% to 25,709, its worst day since the tariff disruption of April 2025, the S&P 500 fell 2.64% to 7,384, and the technology sector dropped more than 5%. The proximate trigger was a guidance disappointment from Broadcom, whose third quarter artificial intelligence revenue outlook of US$16 billion undershot the US$17.2 billion expected and whose refusal to lift its full year forecast was read as the first crack in the capital expenditure narrative. The stock fell 14% even while trading near 37 times forward earnings. A simultaneous spike in Treasury yields after the employment report compounded the move, since a higher discount rate weighs most heavily on the long duration of cash flows that dominate the technology complex.

Whether this is a temporary reversal driven by Federal Reserve sentiment or the opening of a larger drawdown is the central question for the third quarter, and the evidence supports both readings. The case for resilience is that the rally has been earnings led rather than purely multiple driven, that corporate profits remain firm, and that the underlying artificial intelligence capital expenditure cycle continues to expand. The case for a deeper correction rests on valuation and on the change in the monetary regime. The S&P 500 trades near 24 times forward earnings and the cyclically adjusted price to earnings ratio has breached 40, a level last seen at the peak of the dot com era, while the index at 7,384 already sits close to the median yearend target near 7,550 set by firms such as Goldman Sachs, leaving minimal cushion. The liquidity tailwind that powered 2025 was the Federal Reserve easing cycle, and a pivot from cuts toward a hold or a hike removes that support precisely as the market leans on the most demanding valuations in a generation. Our reading is that the path of least resistance has shifted toward a higher volatility, ranging to a corrective regime in which the artificial intelligence leadership is the epicentre of fragility, rather than a disorderly unwind.

Private credit and corporate financing

The same repricing transmits into the fastest growing and least transparent segment of corporate finance. Private credit direct lending assets are expected to exceed US$2 trillion in 2026 and to approach US$4 trillion by the end of the decade, and the asset class is built almost entirely on floating rate loans. The default rate is difficult to measure precisely because of the opacity of the structures, but the proxy ranges from roughly 1.6% excluding distressed exchanges to approximately 4.7% once those soft credit events are included, a band that itself signals how much stress may have been deferred rather than resolved. Around 65% of corporate defaults in 2025 were distressed restructurings rather than outright failures, and more than one in three of those ultimately ends in a hard default or a repeat event.

The link to the employment report is direct. The benign 2026 default outlook that strategists at Bank of America had put forward, easing toward 4.5% from 5%, was predicated on the Federal Reserve cutting rates. Floating rate borrowers do not benefit if the central bank holds or hikes, and a sustained restrictive stance keeps coupons elevated for companies whose interest coverage is already thin. The Moody's baseline places growth near 1.5%, just above the stall speed below which defaults tend to accelerate, so the margin of safety is narrow. The sector is also heavily tilted toward technology and services, the areas most exposed to artificial intelligence disruption, and Morgan Stanley has warned that direct lending defaults could surge toward 8% in an adverse scenario against a historical norm of 2 to 2.5%. The systemic concern is not the headline default rate today but the channels of contagion, since banks finance the funds and insurers and pension funds hold the paper, and the collapses of First Brands and Tricolor last year demonstrated how quickly opacity can convert an idiosyncratic failure into a broader confidence shock. For now, the equilibrium holds, but it is fragile, and a higher-for-higher rate path is the most plausible catalyst for tipping it.

The artificial intelligence complex and the 2026 listing wave

The thread connecting the equity selloff, the credit risk and the macro pivot is the artificial intelligence build out, and the second half of 2026 brings the moment of truth in the form of an unprecedented cluster of listings. SpaceX filed confidentially on 1 April for a flotation that targets a valuation above US$2 trillion and a rise of up to US$75 billion, the largest in history, with a window in the middle of June. Anthropic is reported to be targeting October at a valuation near US$900 billion on a raise of US$30 to US$50 billion, and OpenAI is contemplating a listing later in the fourth quarter near US$1 trillion. The combined pipeline could absorb close to US$200 billions of equity, more than the entire United States listing market raised in 2025, and the concentration of supply matters as much as the valuations. A crowding out effect, in which mega offerings drain liquidity from existing equities, has historically accompanied listing waves of this scale, and it would arrive alongside a record technology weighting in the S&P 500 and a Federal Reserve that no longer adds liquidity.

The valuations themselves encode the duration risk that the rate of move has activated. SpaceX is priced near 95 times revenue and OpenAI near 40 to 60 times against a Microsoft multiple closer to 15, and both OpenAI and Anthropic remain loss making, with profitability targeted for 2030 and 2028 respectively. A higher discount rate compresses the present value of the most distant cash flows most severely, which is why the most speculative end of the complex is the most exposed to a hawkish Federal Reserve. The financing is also reflexive. Model developers raise capital and commit it to compute, paying cloud and chip providers whose booked revenue then validates their own valuations, and the agreement under which Anthropic will pay SpaceX roughly US$1.25 billion a month for capacity through 2029 is one node in a circular structure that functions while capital is abundant and reverses when it is not. There is a genuine read through to public equities regardless of how the offerings are received, because the listings will mark to market the stakes held by Microsoft, SoftBank, Amazon and Alphabet and will reprice the infrastructure providers Nvidia, Oracle and CoreWeave against the first public benchmarks for the sector. The sale of US$6.6 billion of OpenAI stock by current and former employees in the secondary market is a reminder that informed holders are using the wave to transfer risk, and the burden of proof now sits with sustained profitability rather than narrative.

Hong Kong

Hong Kong entered June as one of the best performing major markets in the world and left it confronting a convergence of external and domestic pressure. The Hang Seng Index tracked a year-to-date advance near 28.7% through late May, gained 0.64% on the first session of the month to close near 25,343, and then reversed. It fell 1.5% on the Thursday to close at 25,253 and extended losses into the Friday, leaving the benchmark lower over the week and back toward the 25,000 level. Reading the move requires separating three forces, because the index is being driven at once by a domestic capital story, by the exhaustion of an artificial intelligence rally, and by contagion from the United States, each of which carries a different implication for what comes next.

Capital controls and the southbound bid

The most consequential development for Hong Kong this week was domestic. The structural engine of the 2025 and 2026 rally has been the southbound channel of the Stock Connect, through which mainland investors purchased a record sum of Hong Kong equities last year, estimated by Goldman Sachs at around US$160 billion, and which has at times accounted for between a fifth and a half of daily turnover on the exchange. That bid is now under threat. On 22 May the authorities tightened capital controls restricting mainland residents from accessing offshore accounts, and the China Securities Regulatory Commission imposed more than US$330 million of fines on online brokers it accused of operating without licenses. Hong Kong banks have responded by tightening scrutiny of mainland clients opening investment accounts, the Shanghai branch of Bank of East Asia suspended such openings, and regulators urged lenders to verify sources of funds and close accounts with suspicious documentation.

The market impact was concentrated in the financial names most exposed to cross border flows. On the Friday AIA fell more than 3% in Hong Kong and as much as 6.8% in its London quotation, its worst session since March, HSBC fell around 2% in Hong Kong and as much as 6% in London, Standard Chartered slid roughly 3% in Hong Kong and up to 7.6% in London, and Prudential dropped between 6.5% and 8% to an eight month low. The deeper significance lies in the threat to the flow itself. China International Capital Corporation had already noted that mainland mutual funds were approaching the 50% ceiling for Hong Kong equity positions and were running low on fresh capital, so a regulatory squeeze on cross border access compounds a bid that was already maturing. This is a local fundamental, distinct from anything happening in the United States, and it is a significant variable for the structural case on Hong Kong.

The artificial intelligence tigers and the reversal narrative

The second force is the unwinding of the artificial intelligence rating that drove the market for eighteen months. The DeepSeek breakthrough in early 2025 ignited a revaluation of Chinese technology that lifted the MSCI China Index roughly 31% on the year and carried Alibaba and Tencent to multi year highs, and it produced a wave of listings by the so called artificial intelligence tigers, the leading Chinese model developers, the first of which, Zhipu, debuted in Hong Kong in January. That narrative began to reverse well before this week. The Hang Seng Tech Index entered bear market territory in February, falling more than 20% from its October peak on fears of a value added tax increase on internet services and broader policy risk, and by late May Alibaba sat roughly 34% below its high for the year even as the index it most resembles, the Nasdaq 100, traded higher.

The debate now is whether the reversal is a healthy consolidation within a structural re rating or the end of a liquidity driven move. The constructive argument rests on valuation and policy, since the Hang Seng Tech Index trades near 24 times earnings against the Nasdaq at roughly 31, Beijing continues to back technological self-reliance, and capital is rotating from consumer platforms toward the hardware and semiconductor names that benefit from domestic chip development. The cautious argument is that the rally was a capital flow event vulnerable to the same flow reversing, and that earnings have lagged the capital expenditure they were meant to justify, with cloud revenue trailing infrastructure spend by two to four quarters. The resolution will be visible in artificial intelligence listing volumes, in exchange for traded fund flows, and in the southbound data discussed above.

Contagion from the United States

The third force is contagion, and it is worth being precise about the channels through which a selloff in New York transmits to Hong Kong, because the relationship is more conditional than it appears. Over 2026 the two markets had in fact decoupled, with the Hang Seng Tech Index down around 26% on the year while the Nasdaq 100 rose around 14%, which implies a low and even negative unconditional correlation as Chinese technology traded on domestic catalysts. On acute risk off days that correlation snaps positive, and this week demonstrated the mechanism. The first channel is the semiconductor supply chain, where the chip names carry the highest sensitivity to the global cycle. When the Philadelphia Semiconductor Index sold off on the Broadcom guidance miss, Semiconductor Manufacturing International fell more than 7% on the Friday and Hua Hong fell more than 4%, a single day move that implied a beta near 1.7 to the Nasdaq decline of 4.18%, far above the broad index whose fall of around 1.1% implied a beta closer to 0.25.

The second channel is dual listing arbitrage, since the largest Chinese platforms trade as American depositary receipts whose overnight moves in New York are arbitraged into the Hong Kong shares at the open. The third is the discount rate, because the spike in United States Treasury yields raises the global cost of capital that prices long duration growth assets everywhere, and the Hong Kong dollar peg transmits Federal Reserve policy mechanically through the Hong Kong Monetary Authority. The fourth is positioning, since the global funds that hold both United States and Chinese artificial intelligence names as a single thematic basket de gross both together, a dynamic visible in the US$10 billion of foreign outflows from Korean equities during the same episode. The platform names such as Tencent, which fell between 1.3 and 1.6%, and Xiaomi, which fell around 2%, moved far less than the chip names, confirming that the contagion is concentrated in the high beta semiconductor sub sector rather than spread evenly across the market.

Where Hong Kong is heading

The direction of Hong Kong from here depends far more on its domestic variables than on its beta to the United States. Our reading of the United States trajectory is cautious, since a Federal Reserve removing its easing bias into the most stretched valuations in a generation, an artificial intelligence capital expenditure narrative showing its first cracks, and a listing calendar set to drain liquidity in the second half together argue for higher volatility in American equities rather than a continuation of the trend. That environment exports episodic, tail correlated pressure to Hong Kong, concentrated in the semiconductor complex, but it does not set the direction of the broad index, whose unconditional correlation to the United States has been low this year.

The greater part of Hong Kong's path will be decided locally: by whether the capital control regime chokes the southbound bid that has been the structural pillar of the rally, by whether the artificial intelligence re rating resumes or exhausts, and by whether Beijing delivers the policy support implied by its 4.5 to 5% growth objective. The constructive case is that Hong Kong trades at a discount to the United States, is supported by domestic policy, and is less exposed to a global artificial intelligence de rating on an unconditional basis. The cautious case is that its two domestic pillars, the southbound flow and the technology re rating, are weakening at the same time. On balance the market looks more vulnerable to its own fundamentals than to imported volatility, and the southbound flow data is the single variable worth monitoring most closely. None of this constitutes a recommendation, and the range of outcomes is wide.

Conclusion: going forward with Lumix

The coming week is unusually dense. The Federal Reserve decision on 16 and 17 June, the first under Chair Kevin Warsh and the first to carry his projections, an OPEC+ meeting whose influence the Iran war has hollowed out, the SpaceX listing window, and the unresolved status of the United States and Iran framework all fall within days of one another. The May employment report has tilted the rates of debate from cuts toward a hold or a hike, and the consequences have radiated across every asset discussed in this note. We will continue to place the commodity complex at the centre of our coverage, refined against every new data print through our in-house model.

We therefore encourage you not to miss our next weekly update. For everything you need to stay informed, visit us at lumixtraders.com.au.

Disclaimer: The content published by Lumix Traders Pty Ltd (ABN 35 678 305 738) is intended for educational and informational purposes only. It discusses general market activity, macroeconomic conditions, industry trends, and related commentary and should not be construed as financial product advice, investment advice, or a recommendation to buy, sell, or hold any financial product or security. Lumix Traders Pty Ltd is not licensed to provide financial product advice. We do not hold an Australian Financial Services Licence (AFSL) under the Corporations Act 2001 (Cth). Nothing contained in this material constitutes financial advice, and it should not be relied upon as such. Readers are strongly urged to consult with a licensed and qualified financial adviser before making any investment or financial decision.

The views and opinions expressed in this material are current as of the date of publication and are subject to change without notice. Lumix Traders Pty Ltd has no obligation to update or revise any information following publication. Economic and market commentary presented herein reflects a series of assumptions and judgments as of the date of publication. Actual outcomes will vary. Forecasts and analysis are subject to significant uncertainty and should be viewed as one of a broad range of possible outcomes, not as predictive statements of fact.

Past performance is not indicative of future results. The value of financial assets can go down as well as up. Capital is at risk. This material is not intended for distribution to, or use by, any person in any jurisdiction where such distribution or use would be contrary to applicable law or regulation.