Market Analysis

A Trader's View

In-depth market analysis from a professional trader with 25 years of institutional experience across FX, precious metals and macro — not a journalist, not an algorithm.

Latest 22 July 2026

This Is No Longer Noise

"Gold has decoupled from risk-off. For five weeks it fell with equities. It no longer does. That is not a war trade. That is a chaos trade — and it is telling you something the rest of the market has not yet priced."

Subscribe to unlock
7 July 2026

The First Domino Is Moving Again

"The oil price already has the answer. The bond market is still waiting for permission to agree. At $71.94 — down nearly $50 from its wartime peak — Brent has a 90-day appointment with the CPI data. That appointment is called August."

Read analysis
26 June 2026

The Loudest Room Is Usually the Most Dangerous One

"There is $7.92 trillion sitting in US money market funds. That is 3.6 times the entire stock of cash that accumulated during the 2008 financial crisis. The noise tells you the market is broken. The money tells you something else entirely."

Subscribe to unlock
22 June 2026

The Franc That Never Was a Haven

"The franc's weakness this week is not a one-off reaction to a single piece of news. It's multiple, independent threads pointing the same direction."

Subscribe to unlock
11 June 2026

Sat On The Fence, A Sneeze From All-Time Highs

"If all you did was read the headlines, you'd think this market should be on its knees. Instead, it's a sneeze away from all-time highs."

Subscribe to unlock
1 June 2026

Gold Confusion Unravelled

"If you trade gold as simply a measure of fear and inflation, you are likely to get very badly burned."

Subscribe to unlock

This Is No Longer Noise

The Overriding View
I can feel the tension rising. Not from one signal — from all of them at once. This is no longer noise. Take some risk off the table. Sit and watch. Let the market show you what it wants to do before you re-commit capital.
War escalating to civilian infrastructure. Oil through the highs. BOJ trapped between a fiscal crisis and a currency collapse. Three shipping channels disrupted. Gold decoupling from risk-off. Dimon won't buy. Buffett has $400bn in cash. These signals are not contradicting each other. They are pointing the same way.
The BOJ Trap
Japan carries approximately 260% debt-to-GDP. Every rate hike makes servicing that debt harder. Not hiking collapses the yen, drives inflation, and forces their hand anyway.
Damned if they do. Damned if they don't. Bloomberg reporting BOJ officials are open to moving before December — October now carries 72% probability of a hike. USD/JPY at 163. The 2024 playbook is back. But this time the setup is far more fragile.
The Gold Signal
For five or six weeks, gold fell with risk-off. That is normal — you sell what is working to cover losses elsewhere. That relationship has now broken. Gold is rising independently of the equity selloff.
We called $4,000 as the accumulation level. We continue to call it higher. But the reasons have changed. This is not a macro trade. Gold is responding to everything at once — the war, the oil, the BOJ, the shipping routes, the fiscal stress. It is not a war trade. It is a chaos trade. And that is a much bigger statement.
Read the full analysis +

I am going to be direct with you. I can feel the tension rising — not from one signal, but from all of them, simultaneously.

The global tech rout is not stopping. A war that has now explicitly targeted civilian infrastructure — bridges and power plants in Tehran — has crossed a threshold that markets have not fully priced. Trump has made clear: every attack on a ship in the Strait of Hormuz will be answered with the destruction of one bridge or power plant in Tehran. That is not a surgical military strike. That is civilian infrastructure. A completely different order of escalation.

Oil has broken through its highs again on the back of it. Brent at $94.40, up for a fourth straight session. That is no longer just a supply story — it is a global inflation story, directly feeding into the rate decisions that determine where everything else goes. September Fed hike odds have moved to 55%. The 10-year is at 4.63%, its highest since mid-May. And Iran is showing zero signs of weakness. Whatever the prevailing narrative says, the oil market does not lie. We now have three shipping channels facing varying degrees of disruption. The choked supply chain is being choked further.

Meanwhile, the BOJ is fighting for its currency with one hand and its fiscal credibility with the other. Japan carries approximately 260% debt-to-GDP. Every rate hike makes servicing that debt harder. But not hiking collapses the yen further, drives domestic inflation, and forces their hand anyway. Bloomberg confirmed this morning that BOJ officials are open to moving faster than the market expected — October now carries a 72% probability of a hike, two months ahead of prior consensus. USD/JPY at 163 is a 40-year low for the yen. August 2024 — where a single unexpected hike triggered a global selloff — is the reference point. But this time the setup is far more fragile.

This is the environment we are trading in. And in my view, markets are not pricing it.

One of the most respected figures in my career has said outright he would not buy equities or bonds at these levels. When Jamie Dimon speaks like that, I listen carefully. Warren Buffett is sitting on over $400 billion in cash. He was mocked for it. That looks very different today.

My read is straightforward: take some risk off the table. Place it to one side. Sit and watch. Let the market show you what it wants to do before you re-commit capital. This is not the moment to be a hero. This is the moment to protect what you have.

The signal that matters most to me right now is gold.

For the last five or six weeks, when risk came off, gold fell with it. That is normal behaviour — when you need liquidity, you sell what is working. But that relationship has broken down. Gold is holding. Gold is rising independently of the equity selloff. We called $4,000 as the accumulation level. We have been calling it higher ever since. But I want to be honest about something important: the reasons have changed.

The original thesis was macro — dollar softness, rate expectations, the inflation trade. Those reasons still exist. But gold is no longer rising on macro optimism. It is rising because of everything I have just described. The war. The oil. The BOJ. The shipping routes. The fiscal stress. The sheer weight of simultaneous, compounding uncertainty. Gold is the market's catch-all signal when it no longer trusts the narrative. This is not a war trade. This is a chaos trade. And that is a much bigger statement.

When gold decouples from a risk-off equity environment and begins rising on its own terms, it tells you something the rest of the market has not yet caught up with.

We have a long way to fall if these markets begin to move in earnest.

Take care of your capital first. That is always the first job.
Key data referenced — 22 July 2026
Brent: $94.40 (+3.72%) · WTI: $87.56 (+3.8%) · Gold: $4,122 · US 10Y: 4.63% · USD/JPY: 163.01 · DXY: 100.95 · S&P 500 futures: −0.3% · Nasdaq-100 futures: −0.7% · September Fed hike odds: ~55% · BOJ rate: 1.0% (highest in 31 years) · October BOJ hike probability: 72% · Buffett cash: $400bn+

GMD Trader spent 25 years as a professional trader at ABN-Amro, ANZ and JPMorgan, covering FX derivatives and precious metals markets across London, Amsterdam, Singapore, Melbourne and Asia. Retiring as Managing Director in 2016, the last decade has been spent managing a personal portfolio across equities, fixed income, funds and alternative assets.

This article represents the author's personal views and is published for informational purposes only. Nothing here constitutes financial advice or a recommendation to buy or sell any financial instrument. All trading involves risk.

© 2026 Global Macro Drivers · Premium Content

The First Domino Is Moving Again

The Overriding View
Oil was the first domino to upset the pre-war equilibrium. At $71.94 — down nearly $50 from its wartime peak — it is the first domino resetting it. The rate hike priced as near-certain is now a coin flip. The assets that absorbed the hawkish repricing haven't yet absorbed its reversal.
NFP June: +57K vs 110K expected. ISM Prices cooling on both sides simultaneously. September hike probability: ~50% (from 64%). The arithmetic of what $50 of Brent does to CPI has a 90-day delivery schedule — and the August and September prints are the ones that inform October FOMC.
Before The War
The pre-war macro setup was genuinely exceptional. Outstanding Q1 earnings. Rate cuts in train. Soft landing priced with conviction. That equilibrium was real — and it was disrupted by a single commodity.
The same commodity is now restoring it.
The Shock Was Real
Oil to $120 was a genuine inflation catalyst. Core PCE to 3.3%. ISM Prices Paid to 82.1. The Fed's hawkish response was rational.
But inflation driven by a war-disrupted supply shock is categorically different from structural inflation. The oil market — not a committee — has already priced the difference.
The Reset Is Running
Brent at $71.94 has retraced nearly the entire wartime move. IEA surplus: ~4mb/d (from 1mb/d in April). OPEC+ adding supply. JPMorgan calls sub-$60 next year.
The disinflationary impulse already set in motion will arrive in CPI data with near-mathematical certainty. It is not a forecast. It is scheduled.
What The Repricing Unlocks
When oil fades from CPI, what remains is the most rate-sensitive investment theme of the decade — $725B in AI capex backed by contracted demand, financed through long-dated bonds, held up by multiples exquisitely sensitive to the discount rate.
If long rates follow oil downward, the mathematical expansion in AI multiples does not require a new breakthrough. It requires arithmetic. Oil was the trigger in one direction. It is the trigger in the other.
"
"The oil price already has the answer.
The bond market is still waiting for permission to agree."
— GMD Trader  ·  7 July 2026
When Brent fell from approximately $120 to $71.94 — a move of nearly $50 in six weeks — it delivered the single largest disinflationary impulse to the US economy in years. CPI energy components lag spot prices by 60–90 days. The August and September inflation prints will therefore look dramatically different from anything the Fed was watching when it priced a near-certain October hike. This is not a forecast. It is arithmetic with a scheduled delivery date.
~$120
Brent crude wartime peak — the catalyst for everything that followed
Chapter 01
The Inflation That Was Real
And why the Fed's response was entirely rational — at the time
  • Brent surged from the low seventies to approximately $120 — a 55%+ move in weeks. ISM Manufacturing Prices hit 82.1 in May 2026, the highest since 2022. The inflation shock was real, broad-based, and consistent with hawkish policy.
  • Core PCE climbed to 3.3%. Wages: 3.5%. Services inflation re-accelerated. The Fed's June dot plot had 9 of 18 participants projecting at least one further hike. A September or October move was priced at ~64% probability. That was a rational market read — based on what was true at the time.
  • But there is a critical distinction between structural inflation and war-driven supply shock inflation. The oil market — not a committee — has already made this call. Its verdict is at $71.94.
Read the full analysis +

Before the war, the macro picture was genuinely exceptional. Q1 earnings delivered across the board. Rate cuts were in train. Soft landing was priced with growing conviction — and supported by the data.

When the Iran conflict disrupted Hormuz, the transmission mechanism to inflation was almost instantaneous. Brent surged from the low seventies to a wartime peak of approximately $120 — more than 55% in a matter of weeks. That surge flowed immediately into energy costs, then into input prices across manufacturing and services supply chains, then into CPI and PCE. ISM Manufacturing Prices hit 82.1 in May — the highest since 2022. Core PCE climbed to 3.3%.

The Fed, under Chair Warsh, responded as any credible inflation-targeting central bank would. The June dot plot had 9 of 18 participants projecting at least one further hike. The median funds rate projection implied 3.8% by year-end — above the current level. A hike was priced as near-certain. That was not a policy error. It was a rational response to data that was accurate at the time.

"The inflation was real. But it is important to understand its origin — because an inflation driven entirely by a war-disrupted supply shock is not the same animal as structural excess demand. The oil price, now at $71.94, has already told you which one this was."

The question for the bond market — and for every rate-sensitive asset — is whether it is pricing the inflation that was, or the inflation that is coming.

↑ Back to top
~$50
Brent's decline from peak to today — not yet reflected in CPI data
Chapter 02
The Price the CPI Hasn't Seen Yet
Arithmetic with a 90-day delivery schedule
  • Brent closed today at $71.94 — down approximately $50 from the wartime peak of ~$120. The sharpest sustained oil reversal of 2026 — driven by ceasefire progress, Hormuz normalisation, and a structural supply surplus that was building before the war began.
  • The IEA's implied global surplus has grown from 1mb/d in April to nearly 4mb/d today. OPEC+ is adding 1.4mb/d. JPMorgan calls sub-$60 next year. This is structural, not temporary.
  • CPI energy components lag spot prices by 60–90 days. The August and September CPI prints — the two data points that directly inform the October FOMC decision — will reflect this move in full. That is scheduled.
Read the full analysis +

The most important number in markets right now is not in the jobs report or the ISM. It is in the Brent crude price — $71.94 — and what that number does to CPI in 60 to 90 days.

The mechanics are straightforward. Energy components of CPI are calculated using a rolling average of spot prices over recent months. When Brent was at $120, those averages were being pulled sharply higher. At $71.94, they are being pulled sharply lower. The full magnitude of the reversal has not yet appeared in the monthly CPI prints. It will.

"The oil surplus was forming before the war. The conflict masked it. The peace is unmasking it — with force."

The IEA had already been projecting a significant supply-demand overhang for 2026 before a single shot was fired. The war temporarily disrupted that picture by choking Hormuz flows. Now those flows are recovering — 35 commercial vessels transiting daily, shipping traffic rebuilding — and the underlying surplus is reasserting itself with additional impetus from OPEC+ unwinding its voluntary cuts.

This is not a price that bounces back to $100 at the next geopolitical headline. The structural supply picture has fundamentally shifted. JPMorgan's sub-$60 call is not a tail scenario — it is a base case built on supply arithmetic.

The August CPI print lands in mid-September. The September print lands in mid-October — days before the FOMC meeting where a hike was priced as near-certain. Both prints will reflect roughly $40–50 of oil price decline. The Fed cannot unsee that data.

↑ Back to top
50%
September hike probability — down from 64% before the NFP print. Near-certain to coin flip.
Chapter 03
The Hike That Became a Coin Flip
The bond market has started moving. It hasn't committed.
  • NFP June: +57K vs +110K expected. Prior months revised down 74K combined. Labour force participation at 61.5% — lowest since March 2021. People are leaving the workforce, not finding jobs. Wages: 3.5% — sticky.
  • ISM Prices cooling on both sides simultaneously. Manufacturing: 73.0 from 82.1 — the largest single-month drop since July 2022. Services: 67.7 from 71.3 — lowest since February. Still elevated. Clearly cooling.
  • The 10-year Treasury moved just 2bps on a 53K payroll miss. The bond market knows. It is waiting for the CPI confirmation it needs before committing. When it commits, it will not move 2 basis points.
Read the full analysis +

The bond market's caution is entirely logical. Wages are 3.5%. Core PCE is 3.3%. ISM prices, while falling sharply, are still north of 65. The market has been burned repeatedly by premature inflation calls over the last four years and is not moving until the data lands in the CPI prints.

But here is what that 2-basis-point reaction to a 53,000 payroll miss is actually telling you: the bond market is not complacent. It is waiting. It can see the growth deterioration, and it can see the oil disinflationary impulse in the pipeline. It is not moving because it does not yet have the CPI confirmation it needs to justify the move.

"This is the most difficult environment I have seen for forming a rate view. Every data point is simultaneously pointing in two directions. But the oil price is not ambiguous. It has moved $50. And $50 of oil has a very clear, very scheduled impact on CPI."

Stagflation — which is what we are navigating — is the most difficult economic condition to fight with traditional policy tools. Hike and you accelerate the labour market deterioration already showing up in the NFP data. Hold and inflation stays above 3% with wages feeding it. The Fed is trapped between two legitimate concerns.

But the oil price resolves the trap — not by fixing wages or services stickiness, but by removing the largest single inflationary impulse from the data. When August CPI reflects $50 of Brent decline, the Fed's dual mandate calculus shifts materially. The hike probability that fell from 64% to 50% on NFP data alone will fall further when the CPI data arrives. And when the bond market commits — rather than simply noting — yields will move by more than 2 basis points.

↑ Back to top
$725B
Hyperscaler AI capex in 2026 — the most rate-sensitive investment cycle of the decade
Chapter 04
What Remains When Oil Fades
If the discount rate falls, the mathematics of valuation changes. Not the technology.
  • Amazon, Google, Meta and Microsoft: $725B in AI capex in 2026 — up 77% from 2025. ~$300B financed through long-dated IG bond issuance (~$360B in 10-year duration equivalents). The most interest rate-sensitive capital cycle of the decade.
  • AI equity valuations are DCF calculations on long time horizons. The denominator is the long-term risk-free rate. Google Cloud contracted backlog: $460B. Microsoft AI revenue run rate: $37B (+123% YoY). The demand is real. The sensitivity is to the discount rate — not whether the business exists.
  • Borrowing costs follow bond yields. AI equity prices follow borrowing costs. If oil disinflation flows through CPI, the Fed pauses, yields fall, and the multiple expansion on the dominant investment theme of the decade is mathematically significant.
Read the full analysis +

When we strip away the oil shock and the war-driven inflation, what remains as the dominant macro investment theme? The AI infrastructure build-out — and specifically, the question of what multiple you should pay for it given the rate environment.

The four largest technology companies are on track to spend a combined $725 billion on AI-related capital expenditure in 2026 — up 77% from $410 billion in 2025. Amazon: $200B. Google: $185B. Meta: $125B. Microsoft: $120B. Wall Street estimates approximately $300 billion of this will be financed through investment-grade bond issuance — long-dated paper, reflecting the multi-decade useful life of data centres — delivering around $360 billion in 10-year duration equivalents into a market already grappling with term premium.

"This is not a speculative build. The contracts are already signed. Google Cloud backlog: $460B. Microsoft AI revenue: $37B annual run rate, up 123% year-on-year. The bear case requires the demand to disappear. The data says it's accelerating."

But here is the rate sensitivity. AI equity valuations are discounted cash flow calculations stretched across very long time horizons — 10, 15, 20 years of earnings. The denominator is the long-term risk-free rate. A 50-basis-point decline in the 10-year yield — which is entirely plausible in a world where August CPI shows oil disinflation flowing through — produces a mathematically significant expansion in multiples on businesses already generating extraordinary returns.

This is not speculation about the technology. It is arithmetic about the discount rate. The oil price started the repricing in one direction nine months ago. It is starting it back in the other direction today. The assets most sensitive to that reversal are precisely those that were most compressed by the fear of rising rates.

That is the trade beneath the noise. And the first domino is already moving.

↑ Back to top
The View From Here · 7 July 2026
The First Domino. Again.
This is an exceptionally difficult environment. Every data point pulls in two directions. The policy response is trapped. The market is exhausted by the contradiction. But the oil price is not contradictory — it has moved nearly $50 in six weeks, and the structural surplus forming beneath it makes that move durable rather than temporary. The "near-certain" rate hike is now a coin flip. When the August CPI print confirms what Brent already has, the dominoes that have been waiting — rates, multiples, positioning — will find their direction.

Oil was the first domino to fall. It is now the first one setting them back up.
Key data referenced — 7 July 2026
Brent: $71.94 · Wartime peak: ~$120 · S&P 500: 7,483 · Nasdaq: 25,832 · VIX: 15.81 · US 10Y: ~4.46% · US 2Y: ~4.14% · Gold: $4,161 · NFP June: +57K (exp. +110K) · Labour force participation: 61.5% (lowest since March 2021) · ISM Manufacturing Prices Paid: 73.0 (prev 82.1) · ISM Services Prices Paid: 67.7 (prev 71.3) · September hike probability: ~50% (from ~64%) · Core PCE: 3.3% · IEA surplus forecast: ~4mb/d (from ~1mb/d in April) · AI hyperscaler capex 2026: $725B (+77% YoY) · Google Cloud backlog: $460B · Microsoft AI run rate: $37B (+123% YoY)
Share this analysis: Share on X

GMD Trader spent 25 years as a professional trader at ABN-Amro, ANZ and JPMorgan, covering FX derivatives and precious metals markets across London, Amsterdam, Singapore, Melbourne and Asia. Retiring as Managing Director in 2016, the last decade has been spent managing a personal portfolio across equities, fixed income, funds and alternative assets.

This article represents the author's personal views and is published for informational purposes only. Nothing here constitutes financial advice or a recommendation to buy or sell any financial instrument. All trading involves risk.

© 2026 Global Macro Drivers · Free Content

The Loudest Room Is Usually the Most Dangerous One

The Overriding View
Step back from the noise. The loudest room is almost always the most dangerous place to form a view — and right now, the room is exceptionally loud.
When a view is loud enough to dominate the conversation, it has usually been in the price for some time. The macro structure beneath the noise tells a very different story. Three things the loud room is getting wrong.
On Hyperscalers
This is a valuation debate. Not an existential crisis. The revenue is real and the contracts are already signed.
$460B Google Cloud backlog. Microsoft AI up 123% YoY. $725B capex backed by contracted demand — not speculation.
On Gold
Paused after an exceptional run. Not broken. The structural floor has softened — it has not disappeared.
Central bank demand: 863 tonnes in 2025, still well above long-run averages. Poland, China, India, the Gulf — still accumulating.
The Hidden Opportunity
$7.92 trillion in uninvested cash. 3.6× the entire 2008 crisis peak. When it rotates, it does not trickle — it moves markets.
MMFs + pension equity at 20-year lows. This capital deploys into the most liquid equities on earth — the names the loud room is writing off.
The Buffett Principle — Applied
Be greedy when others are fearful. The room right now is very loud and very fearful. Most people quote Buffett. Almost no one applies it when the moment actually arrives.
This is not a call. It is a structural observation: the macro data and the crowd sentiment are pointing in opposite directions. That gap is where the opportunity lives.
"
"Be fearful when others are greedy,
and greedy when others are fearful."
— Warren Buffett  ·  The principle most people quote, and promptly ignore when the moment actually arrives.
That quote is not a market timing system. It is a description of crowd psychology and what it does to rational decision-making. When everyone is in the same room shouting the same thing — the AI trade is dead, the rally cannot last — that room becomes the most dangerous place to take your cues from. By the time a view dominates the conversation, it has usually been in the price for some time. This week: three things the loud room is getting wrong.
$460B
Google Cloud contracted backlog — revenue already signed
Chapter 01
On the Hyperscalers
The market is asking the wrong question
  • The capex is extraordinary — and it's backed by real contracts. Four hyperscalers spending ~$725B in 2026, up 77% YoY. Microsoft AI: $37B annual run rate, up 123% year-on-year.
  • The question is what multiple to pay. Not whether these businesses have a future. Distribution, proprietary data, physical infrastructure, network effects — built over two decades. These are the most profitable businesses ever created.
  • Apple's price hikes are a supply chain story, not a demand collapse. DRAM and NAND up 80–90% as AI data centres crowd out consumer hardware. Apple raised prices because input costs surged.
Read the full analysis +

The market has been repricing Microsoft, Alphabet, Amazon, Meta and Apple. And in one sense it is right to. The four largest hyperscalers are on track to spend a combined $725 billion in 2026 — up 77% from 2025, one of the largest single-year capital mobilisations in corporate history. The sector issued $121 billion in bonds in 2025 alone, more than four times their average annual issuance from 2020 to 2024. These are genuinely significant numbers.

But these are the most profitable businesses ever created. Their competitive moats are not AI hype — they are distribution networks, proprietary data, physical infrastructure and network effects built over two decades. The question is simply what multiple you should pay for that future given current rates and elevated capex. That is a valuation debate. It is not an existential one.

"The question is not whether these businesses have a future. The question is what multiple to pay. That is a valuation debate — not an existential one."

And when you look at where the revenue is actually going, the capex case strengthens. Microsoft's AI business: $37B annual run rate, up 123% year-on-year. Google Cloud backlog: $460B — roughly double the prior year. The bears need the capex to be speculative and unearned. The data says the contracts are already signed and the revenue is arriving.

When the noise tells you the AI era is over because Apple raised Mac prices by 15–25% — ask what is actually driving that. DRAM and NAND memory prices have risen 80–90% as AI data centres buy capacity at margins three to five times conventional pricing, crowding out consumer hardware from the supply chain. That is a supply chain story. It is not evidence that the installed base of one billion Apple devices is becoming worthless.

↑ Back to top
863T
Central bank gold purchases in 2025 — still above long-run averages
Chapter 02
On Gold
Consolidation after an exceptional run is what markets do
  • The pullback is structural clearing, not a collapse. After two years of extraordinary returns, consolidation is healthy and expected.
  • Central bank demand: paused, not reversed. 863 tonnes in 2025, down 21% YoY but still well above historical averages. Poland, China, India, Kazakhstan, the Gulf — all still accumulating.
  • When the structural bid softened, price found less support. That is the market clearing — not the cycle ending.
Read the full analysis +

Gold is the talk of the room. It has had an extraordinary run, and the commentary that follows extraordinary runs is always the same: at the top, everyone finds reasons it should go higher; on the first pullback, everyone finds reasons it was always going to fall. Neither is particularly useful.

"Gold consolidating after a two-year run of extraordinary returns is healthy. It is what markets do. The noise will tell you it signals the end of the cycle. The data says central banks are still accumulating."

The structural demand picture has not reversed — it has paused. Those are very different things.

↑ Back to top
$7.92T
US money market funds — the story no one is telling
Chapter 03
The Money Wall
Building quietly beneath the noise
  • $7.92T in US MMFs. 3.6× the entire 2008 crisis level. There is no historical precedent for this level of uninvested cash in the modern era.
  • It arrived because rates made doing nothing competitive. That is changing. As the rate environment shifts, the case for holding cash deteriorates — and the capital has to go somewhere.
  • Pension equity allocations at 20-year lows. $68.3T in global pension assets, ~48% equity — down from 57% two decades ago.
Read the full analysis +

There is $7.92 trillion sitting in US money market funds right now. That peaked at $8.3 trillion earlier this year. The entire MMF stock at the peak of the 2008 financial crisis was $3.51 trillion. We are currently at 3.6 times that level.

This is not a normal level of uninvested cash. It arrived because the rate environment briefly made doing nothing competitive. That environment is changing.

"The loud room focuses on the news that is happening today. The money wall is the story building quietly beneath it."

When $7.92 trillion in cash begins to rotate — not all at once, but at the margin — it does not trickle into markets. It moves them. And when pension trustees start allocating after years of running historically low equity weights, they deploy into the most liquid, most scalable, most institutionally accessible equities on earth. Which happen to be the same hyperscalers the loud room is currently writing off.

↑ Back to top
Stepping Out of the Loudest Room
None of this is a call. Anyone who tells you they have high conviction every single day is selling something. What the data is saying: hyperscalers are undergoing a justifiable valuation adjustment backed by real contracted revenue. Gold is consolidating with structural central bank demand intact. And beneath all of it, a historically anomalous wall of uninvested capital — in money market funds and structurally underweight pension allocations — is waiting for a reason to move.

That capital does not care what the room is shouting. It cares about returns. When it moves, it will move fast. And the loudest room will be the last to see it coming.
Key data referenced — 26 June 2026
US MMFs: $7.92T (peaked $8.3T) · 2008 MMF peak: $3.51T · COVID MMF peak: $5.2T · Hyperscaler 2026 capex: ~$725B (+77% YoY) · Microsoft AI revenue run rate: $37B (+123% YoY) · Google Cloud backlog: $460B · CB gold purchases 2025: 863T · Global pension assets: $68.3T · Pension equity allocation: ~48% (near 20-year low)

GMD Trader spent 25 years as a professional trader at ABN-Amro, ANZ and JPMorgan, covering FX derivatives and precious metals markets across London, Amsterdam, Singapore, Melbourne and Asia. Retiring as Managing Director in 2016, the last decade has been spent managing a personal portfolio across equities, fixed income, funds and alternative assets.

This article represents the author's personal views and is published for informational purposes only. Nothing here constitutes financial advice or a recommendation to buy or sell any financial instrument. All trading involves risk.

© 2026 Global Macro Drivers · Premium Content · Subscriber Access Only

Sat On The Fence, A Sneeze From All-Time Highs

"If all you did was read the headlines, you'd think this market should be on its knees. Instead, it's a sneeze away from all-time highs."

That contradiction is the whole story right now. Not the ECB. Not the next data print. The contradiction itself.

The pullback that wasn't a crash

The S&P 500 closed at 7,304 on Wednesday, down from the all-time high of 7,620.9 set on 2 June — a pullback of roughly 4.3%. The DAX tells a similar story: trading around 24,240, it sits roughly 5% below the record 25,508 it printed back in January, and has spent the last month consolidating in the 24,000–25,400 range rather than breaking down. In both cases: a pullback, not a breakdown.

After the kind of run we've had — the S&P up over 11% on the year into that early-June high, nine straight weeks of gains behind it — a pullback of this size is healthy, almost mandatory. Some risk coming off the table after an outstanding stretch is exactly what you'd expect to see, and exactly what you'd want to see if you're hoping the move has further to run. What it is not is evidence of a trend change. Trend changes need a reason. Right now, all we have is a market catching its breath.

"This is just a little risk taken off the table. It makes sense — it's been an outstanding run. What's missing is a fresh catalyst."

A mega week, already written in

This week matters — but maybe not for the reason most people think.

On Thursday the ECB raised its deposit rate by 25 basis points to 2.25% — its first hike since 2023, a direct response to eurozone inflation running at 3.2% as the Iran-Hormuz oil shock pushes energy costs higher. By any normal standard, that's a headline event: the first ECB hike in three years, delivered against a live geopolitical backdrop. And the market's reaction was a shrug — the 10-year Bund yield actually moved two basis points lower, and EUR/USD and EUR/GBP were essentially flat. A hike that, in a different environment, would have been the story of the week — absorbed without a ripple.

That non-reaction tells you almost everything about where we are: the move was written in long before it happened.

The logical follow-through is that other central banks now face a higher bar for hawkish surprises too. The Fed's FOMC meets on 18 June, with the BoJ also on the calendar next week — both arriving into a market that has spent weeks pricing in rate-hike risk across the board. The question this week — and next — answers is not "will it happen" — it's "now that it has, or is about to, what does the market actually do with it?"

"It might be written in. But events still result in interesting movements. Getting through this mega week is what sets the tone for the next leg."

Reading the resilience, not the headlines

Here's what strikes me most. From where I sit, all I read is bad news. Geopolitical risk. Hawkish central bank rhetoric. Rate-hike expectations building across multiple regions. Plenty of reasons, individually and collectively, that should have put these markets clean on their backsides.

And yet — here we stand. The S&P 500 within roughly 4% of its record. The VIX at 21.6, down from over 22 the day before, sitting comfortably inside its one-year range of 13.4 to 35.3 — elevated, but nowhere near the levels of genuine stress we saw earlier in this Iran-driven episode.

"This market is resilient. There's been more than enough negative news flow to floor it. It hasn't happened."

That resilience is the signal, not the noise. When a market absorbs wave after wave of bad news and refuses to break, that tells you something about the underlying bid that no single headline can. There are still huge waves of cash sitting on the sidelines, still needing to be put to work. That capital doesn't disappear because the news cycle is unpleasant — it waits for a reason to move, and when it gets one, it tends to move fast.

The reverse indicator

Often, the best indicator is the reverse indicator. When sentiment is this consistently negative and price action this consistently stubborn, the gap between the two is where the opportunity sits.

To be clear — this isn't a call. It's a gut feel, and I'll say it tongue-in-cheek: if I were a gambling man, I'd back the resilience to keep shining through. We're in limbo, waiting for the next catalyst, and limbo can resolve in either direction. But when bad news this persistent fails to do its job, the asymmetry starts to tilt — a fresh positive catalyst could spark a sharp move higher faster than another round of bad news could meaningfully break this floor.

Where that leaves positioning

Sitting on the fence is the right approach here — but the fence has a lean to it.

This is where experience matters. A trader who has spent decades pricing risk reads "4–5% off the highs, mega week of central bank decisions, geopolitical overhang" very differently from a financial journalist who has never had to put capital behind a view. The journalist sees a wall of reasons to be cautious. The trader sees a market that has already absorbed all of those reasons and is still standing — and asks what that's telling them about supply and demand, not just sentiment.

Nobody can have a fresh, high-conviction opinion every day, all day. That's not how risk-taking works, and anyone who tells you otherwise is selling something. Most days, the right answer is to wait. This week is one where waiting has a direction attached to it.

"I sit on the fence with a gut feel that says be comfortable having risk on. That's all I'm sayin'."

Watch how markets digest this week's central bank decisions — not the decisions themselves, but the follow-through, or lack of it. That's the signal that sets the next trade pattern. Until then: comfortable holding risk, watching for the catalyst, not predicting it.

Levels referenced — 11 June 2026
S&P 500: 7,304 · All-time high: 7,620.9 (2 June 2026) · DAX: ~24,240 · VIX: 21.6 (range: 13.4–35.3) · ECB deposit rate: raised 25bp to 2.25% · Eurozone CPI: 3.2% · Next: FOMC 18 June, BoJ same week

GMD Trader spent 25 years as a professional trader at ABN-Amro, ANZ and JPMorgan, covering FX derivatives and precious metals markets across London, Amsterdam, Singapore, Melbourne and Asia. Retiring as Managing Director in 2016, the last decade has been spent managing a personal portfolio across equities, fixed income, funds and alternative assets.

This article represents the author's personal views and is published for informational purposes only. Nothing here constitutes financial advice or a recommendation to buy or sell any financial instrument. All trading involves risk.

© 2026 Global Macro Drivers · Premium Content · Subscriber Access Only

Gold Confusion Unravelled

"If you trade gold as simply a measure of fear and inflation, you are likely to get very badly burned."

Let me start there. Because everything that has happened in gold over the last two years — the extraordinary rise, the recent selloff during an active war, the confusion in between — makes complete sense once you stop looking at it the way the financial media tells you to.

The Iran test case

Active US military strikes on Iranian territory. The Strait of Hormuz effectively closed. One of the most significant geopolitical crises in a generation. And gold sold off.

Every retail trader who had been told, repeatedly, that gold is the ultimate safe-haven asset was left staring at their screen wondering what they had missed. They hadn't missed anything fundamental. They had simply misunderstood how gold actually trades.

"Gold should be viewed as a barometer for fear and inflation — however it often trades completely independently. It can do the polar opposite of what it's meant to do."

This is not a flaw in the gold market. It is the gold market.

A market smaller than you think

The first thing to understand is that the gold market is not enormous. For an asset with such cultural and financial weight, the actual tradeable market is surprisingly thin. That thinness matters — it means positioning, sentiment and flows can overwhelm fundamentals for extended periods.

Central bank physical demand has definitely been a driver. The picture that emerges from World Gold Council data is telling. Net buying across central banks globally was substantial and historically elevated through 2024 and into 2025. Full-year 2025 purchases came in at 863 tonnes — down 21% year-on-year from the +1,000 tonne pace of the preceding three years, but still well above long-run historical averages. The buying was geographically diverse: Poland was the largest single buyer in 2025, adding 102 tonnes. China, India, Kazakhstan, the Czech Republic and Qatar were also consistent accumulators. The motivation varied — some driven by strategic de-dollarisation, others by reserve diversification — but the aggregate effect was the same: sustained structural demand underpinning the gold price.

Then that buying began to moderate. Into late 2025 and early 2026, the pace slowed noticeably. Overlay that moderation against the gold chart and the correlation is striking. The structural demand floor that had supported the rally did not disappear overnight, but it quietly softened. And when it did, the price found considerably less support beneath it.

That is not coincidence. That is the market.

The FOMO machine

Here is what actually drives gold's biggest moves.

"Nobody truly understands why it's moving, but acceptance quickly grabs hold. An almost market FOMO takes over — more and more people jump on and the moves are exacerbated as a result."

Someone buys. It goes up. A narrative forms — inflation hedge, dollar weakness, geopolitical risk, whatever fits the moment. More people buy because it's going up. Momentum traders follow. Retail investors follow. The move gets exaggerated far beyond what any fundamental analysis would justify.

"Gold's enormous rise to fame in the last 24 months has been as baffling to most as the selloff recently. Anyone telling you they fully understood it in real time was either lucky or lying."

When war erupted and markets lurched violently, what happened in gold was entirely predictable — not from a fundamental standpoint, but from a positioning one. Two years of extraordinary returns had built massive long positioning across institutional and retail investors alike. When volatility spiked and margin calls hit elsewhere, gold was liquid. It got sold to cover losses in other positions. The safe-haven narrative didn't disappear. It was simply overwhelmed by forced selling from overleveraged longs who had ridden the FOMO machine too far.

A precise framework for filtering the noise

Now let me give you something you can actually use.

Look at CME September Gold ATM options. Implied volatility is currently around 23 vol. On a long-term historical basis, this is elevated — and elevated for good reason given everything described above. But it is the context around that number that matters most.

At 23 vol, the options market is implying that gold should move approximately 1.5% per day. That is simply what 23% annualised volatility translates to on a daily basis.

Under 1.5%
Noise — ignore
Gold is within its expected daily range. The options market is unmoved. Do not react.
Around 2%
Worth noting
You are outside the expected range. Note it in the context of what else is happening.
Above 3%
Significant event
Something real is happening. A significant market event — not a headline, an event.

Apply this to every gold headline you read. The financial media will report a 1.2% gold move as if it means something. It does not. Save your attention for when it genuinely matters.

The regime we are in now

Here is the context that changes how you read all of the above.

Before two years ago, gold volatility had a very different baseline. Single digits to the mid-to-high teens was the normal operating range. That world no longer exists.

The last two years have delivered a vol range of remarkable width. At the lows, around summer 2024, gold vol compressed back to 10–12%. At peak stress, it pushed into the 40s. That is an extraordinary reading for a commodity markets had long treated as relatively sedate.

At 23 today, we sit almost exactly in the middle of that two-year range of 10 to 40.

"Today we very much sit in the middle of the range. This will suggest a drift-style nature to both underlying and vol markets."

A vol at the midpoint of its range is not pricing in acceleration, nor pricing in calm. It is a drift environment. Neither vol buyers nor vol sellers have a clear structural edge. The underlying price is likely to exhibit the same characteristics — not trending aggressively in either direction, but capable of sharp moves when a genuine catalyst arrives.

For traders this means: momentum strategies will struggle. Mean-reversion will have more frequent opportunities. Neither pure option selling nor pure buying is the obvious play — structures that benefit from occasional spikes while not bleeding in a directionless tape deserve consideration.

The bottom line

Gold is not broken. It is not simple. It is an asset whose moves are driven by a layered combination of strategic central bank flows, overleveraged positioning, FOMO dynamics, and genuine macro sensitivity — and those forces rarely all point in the same direction at the same time.

Trade it with that complexity in mind. Use the options market as your noise filter. Watch positioning, not just headlines. And never, ever trade it as a simple fear gauge.

You will get very badly burned.

GMD Trader spent 25 years as a professional trader at ABN-Amro, ANZ and JPMorgan, covering FX derivatives and precious metals markets across London, Amsterdam, Singapore, Melbourne and Asia. Retiring as Managing Director in 2016, the last decade has been spent managing a personal portfolio across equities, fixed income, funds and alternative assets.

This article represents the author's personal views and is published for informational purposes only. Nothing here constitutes financial advice or a recommendation to buy or sell any financial instrument. All trading involves risk.

© 2026 Global Macro Drivers · Premium Content · Subscriber Access Only

The Franc That Never Was a Haven

This morning brought the kind of headline that should have hurt sterling. Keir Starmer has resigned, and Andy Burnham — a figure markets have learned to treat with caution — is the overwhelming favourite to become Britain's seventh prime minister in a decade. By the textbook, GBP should be under pressure.

It isn't. Cable is firm, and GBP/CHF is actually pushing higher on the day. The explanation isn't a UK story at all. It's a Swiss one — and it's one we flagged here well before it became consensus.

A call made early

Readers of this column will recall the long GBP/CHF case laid out in detail on 11 June: a 19-year technical downtrend breaking to the upside, an exceptional 375bp carry differential, and — critically — a Swiss National Bank trapped at the zero bound, structurally unable to let its currency strengthen without risking outright deflation. The framing then was explicit: SNB intervention and any Middle East de-escalation were not threats to the trade, they were entry opportunities, because the franc's safe-haven status in this particular conflict looked thinner than the conventional narrative assumed.

That call has now been made independently, and rather more loudly, by the wider market. FXStreet published a piece on 19 June — "Swiss Franc sinks by design, not by peace" — that amounts to the same argument restated: the franc's slide isn't a haven trade unwinding, it's a central bank's policy finally being allowed to show through once the geopolitical noise clears. We were making that case in print a full week earlier.

The Franc that never showed up to its own war

That FXStreet piece cuts against the lazy narrative that's been attached to the Swiss franc since March. The framing goes: Middle East war breaks out, safe-haven flows hit CHF, franc strengthens; war de-escalates, safe-haven flows reverse, franc weakens. Tidy story. Wrong story.

Look back at what actually happened. In the opening days of the conflict in early March, the franc did spike — briefly touching its strongest level against the euro since 2015. But the move was short-lived. By late March, analysts were openly describing the franc as a war loser rather than a refuge, as the SNB pushed back against the strength almost as soon as it appeared.

That distinction matters enormously for how we should read this week's move. A haven-unwind story is fragile — it can reverse the moment headlines turn again. A central-bank-by-design story is durable, because it doesn't depend on the news cycle. It depends on the SNB's policy stance, which isn't going anywhere.

What the SNB has actually said — on the record

This isn't speculation about central bank motives. The SNB's verbal intervention campaign is unusually well documented, and it's worth setting out the actual record rather than relying on the general impression that's built up around it.

2 March 2026 — In an unsolicited statement to media, the SNB said: "In view of international developments, we are increasingly prepared to intervene in the foreign exchange market... We are ready to intervene in the foreign exchange market to curb a rapid and excessive appreciation of the Swiss franc, which would jeopardize price stability in Switzerland." Bloomberg/SWI reporting noted this marked a deliberate hardening of tone, departing from the SNB's standard, softer formulation.

24 April 2026 — SNB President Martin Schlegel told Bloomberg: "We have unrestricted room for manoeuvre with regard to the SNB policy rate and foreign exchange market interventions." The same report noted that UBS economists estimated the SNB had bought around CHF 2.5 billion of foreign exchange in March specifically to weaken the franc.

18 June 2026 — At the latest policy meeting, the SNB held its rate at zero and restated its readiness to intervene "if necessary." Schlegel told CNBC: "A strong and rapid appreciation of the Swiss franc could endanger price stability in Switzerland. Therefore, we still have this increased willingness to intervene in the FX market."

That last data point comes with a genuine nuance worth flagging. Commentary on the June decision noted that the shift from the harder March wording to the softer "if necessary" framing was itself read by markets as dovish — the SNB signalling the intervention tool exists without confirming it's loaded and ready to fire. Schlegel declined to explain the wording change when pressed by reporters. If anything, that reinforces rather than undermines the core point: the franc sold off sharply within minutes of that statement, on the mere suggestion that SNB support might be less active than previously priced. The market doesn't need the SNB to actually sell francs to weaken the currency — it just needs to believe the central bank is less inclined to defend it.

The technical confirmation

USD/CHF has done real technical damage to its 2026 downtrend, closing decisively above its 200-day moving average for the first time this year. The immediate ceiling is the year's high near 0.8100; a sustained push through that opens clear air toward 0.8150 — territory untraveled all year. Momentum indicators are stretched, so a pullback to retest the breakout zone wouldn't be unusual, but the structural break itself looks real.

Read-through to the live position

This is where the macro story and the trade converge. The 11 June note built the long GBP/CHF case on exactly the three legs now being independently validated: a genuine technical breakout, a carry differential that the BoE/SNB policy paths only widen further, and a Swiss central bank constrained by its own zero bound from doing anything about franc weakness beyond talk. Today's price action — GBP/CHF firmer despite a UK political shock that should, on paper, have weighed on sterling — is the clearest single-day evidence yet that the CHF leg of that thesis is now doing more work than the GBP leg.

The read-through for currency markets generally

"The franc's weakness this week is not a one-off reaction to a single piece of news. It's multiple, independent threads pointing the same direction."

For anyone watching the broader CHF crosses, the message is consistent across the board. The franc's weakness this week is not a one-off reaction to a single piece of news. It's multiple, independent threads pointing the same direction: a central bank with an explicit, recently-hardened-then-softened verbal intervention stance; a currency that demonstrably failed to perform its traditional haven function even during an actual war; and a technical breakdown that's confirming what the macro picture has been suggesting since March. Put a UK political transition next to that backdrop and ask which story is doing more work in currency markets this morning. The franc's troubles are homegrown. Sterling's political theatre is, for now, just noise sitting on top of it.

GMD Trader spent 25 years as a professional trader at ABN-Amro, ANZ and JPMorgan, covering FX derivatives and precious metals markets across London, Amsterdam, Singapore, Melbourne and Asia. Retiring as Managing Director in 2016, the last decade has been spent managing a personal portfolio across equities, fixed income, funds and alternative assets.

This article represents the author's personal views and is published for informational purposes only. Nothing here constitutes financial advice or a recommendation to buy or sell any financial instrument. All trading involves risk.

© 2026 Global Macro Drivers · Premium Content · Subscriber Access Only
Premium only

Unlock A Trader's View

Every article is written by a professional trader with 25 years of institutional experience. No fluff, no filler — just direct market analysis from someone who has managed real money.

Published when the market warrants it, not on a schedule.

Subscribe — £29/month · Less than £1/day
Cancel anytime · Instant access · No lock-in
Also included with Premium
Full Macro Heat Map
Daily briefs to your inbox
Trade Ideas

Disclaimer: All content published on Global Macro Drivers is for informational and educational purposes only. Nothing on this website constitutes financial advice, investment advice, trading advice, or any other form of regulated financial guidance. Global Macro Drivers does not recommend the purchase or sale of any financial instrument, security, or asset. Past market commentary does not guarantee future results. Markets are volatile and all trading and investment activity carries risk, including the risk of total loss of capital. You should always conduct your own research and consult a qualified independent financial adviser before making any investment or trading decisions. Global Macro Drivers accepts no liability for any losses or damages arising from reliance on the content published on this site.