Market News & Insights
Market News & Insights
The tariff impact: The US margin problem vs Asia’s volume problem
GO Markets
25/5/2026
•
0 min read
Share this post
Copy URL

Tariffs do not hit every company the same way.

For US retailers and consumer brands, the first pressure point is usually margin. Import costs rise, but companies may not be able to pass the full increase to customers straight away. Older inventory can also delay the impact, which means the first earnings result may look calmer than the next one.

For Asian exporters, the pressure often looks different. The issue may not be a sharp price cut. It may be fewer orders from US buyers.

Tariffs, earnings and the Asia versus US split | GO Markets

Same tariff. Different earnings hit.

That is the key split for traders watching this earnings season. The US side is mainly about margin timing. The Asia side is about demand sensitivity. Not every export sector carries the same level of US demand risk.

TL;DR

  • US companies may face margin pressure as tariffed inventory moves through earnings.
  • Asian exporters may face volume pressure if US buyers reduce orders.
  • The timing is different: US retailers may feel the impact later, while Asian exporters may see it earlier through weaker order books.
  • Textiles, apparel and basic consumer goods are likely more sensitive to US demand.
  • Semiconductors and AI hardware may be less directly exposed to US consumers, but still carry policy, capex and valuation risk.

The big picture

Tariffs are paid at the US border by importers. From there, the cost can move through the system in several ways: higher prices, weaker margins, lower supplier prices, lower demand or a mix of all four.

Research cited by the Kiel Institute and New York Fed suggests US buyers and businesses may be absorbing a significant share of the tariff burden. That matters because it changes where the earnings pressure shows up first.

For a US retailer, the problem is straightforward but uncomfortable. If the company raises prices, demand may weaken. If it absorbs the tariff cost, margins may compress. If it still has older inventory, the hit may not show up immediately.

For an Asian exporter, the pressure can arrive through a different channel. If US buyers become cautious, they may order less. The exporter may keep prices relatively stable, but factory utilisation falls, fixed costs are spread across fewer units and earnings pressure builds.

That is why this is not just a tariff story. It is an earnings timing story.

US companies: the margin problem

The US side of the tariff story is about cost absorption.

Retailers, apparel brands, consumer electronics sellers and appliance companies often rely on imported goods, components or packaging. When tariff costs rise, they may try to protect margins through price increases, supplier negotiations, sourcing changes or inventory management.

The challenge is that none of these are clean solutions.

Price increases can test consumer demand. Supplier negotiations may take time. Sourcing changes can be expensive or slow. Inventory timing can make the first result look better than the underlying cost trend.

This is why earnings calls matter. Management commentary around pricing actions, tariff mitigation, sourcing, vendor negotiations and inventory timing may reveal more than headline sales growth.

What to watch on the US side

These signals may provide useful context in upcoming earnings reports:

If margins hold while sales remain stable, companies may be managing the pressure. If sales rise but margins fall, tariff costs may not be passing through cleanly. If guidance becomes more cautious, the market may start pricing a delayed earnings impact.

Asian exporters: the volume problem

The Asia side is not always about exporters cutting prices.

In many categories, Asian suppliers operate in competitive global markets with limited pricing power. If US buyers reduce orders, exporters may feel the impact through lower volumes rather than lower unit prices.

That distinction matters.

A company can report stable prices and still face earnings pressure if factories are running below normal utilisation. Lower volumes can reduce operating leverage, delay capital expenditure and weaken guidance.

The highest-risk sectors are usually those most closely tied to US retail demand, seasonal buying cycles and low-margin production.

Which Asian sectors are most exposed?

1. Textiles and apparel +
Highest Sensitivity

Textiles and apparel are among the clearest examples of US demand exposure.

These exporters are often tied directly to US retail orders, private-label contracts and seasonal buying cycles. If US retailers turn cautious, orders can be delayed, reduced or cancelled relatively quickly.

Risk is higher because margins are often thin, production is labour-intensive and buyers may have more power in negotiations.

Relevant export markets: Vietnam, Bangladesh, India, Indonesia and parts of China.
2. Basic consumer goods +
High Sensitivity

This includes toys, household goods, furniture, simple appliances and other discretionary or semi-discretionary exports.

These categories are exposed when US retailers reduce inventory or when consumers pull back from non-essential spending. Tariffs can add pressure if buyers try to push costs back onto suppliers.

Relevant export markets: China, Vietnam, Thailand, Malaysia and Indonesia.
3. Electronics assembly +
Medium to High Sensitivity

Electronics assembly is more mixed.

Lower-end consumer electronics can be sensitive to US household demand. Higher-value components or enterprise-linked electronics may be more resilient, depending on end-market exposure.

This sector can also be harder to read because supply chains are complex. A company may look like a technology exporter, but its actual earnings sensitivity may still depend on US consumer replacement cycles.

Relevant export markets: China, Vietnam, Malaysia, Thailand, Taiwan and the Philippines.
4. Machinery and industrial goods +
Medium Sensitivity

Machinery is less directly tied to US consumer demand than apparel or household goods. The risk is more about business investment.

If US companies delay capital expenditure because tariff uncertainty rises, machinery orders may weaken. However, order books can provide some buffer, and specialised products may have more pricing power.

Relevant export markets: Japan, South Korea, China, Taiwan and Singapore.
5. Semiconductors +
Lower Direct Sensitivity

Semiconductors are less directly exposed to US retail demand than textiles or consumer goods. Demand is often tied to broader technology cycles, autos, industrials, cloud infrastructure and AI investment.

That does not make the sector risk-free. Tariffs, export controls, geopolitics and a weaker global capex cycle can still affect earnings expectations.

Relevant export markets: Taiwan, South Korea, Malaysia, Singapore and parts of China.
6. AI hardware and data-centre supply chains +
Lowest Direct Sensitivity

AI hardware is more tied to cloud capital expenditure and data-centre buildouts than day-to-day consumer spending.

The risk is different. It is less about US shoppers buying fewer goods and more about whether AI capex expectations remain realistic, whether policy restrictions expand and whether valuations already price in strong growth.

Relevant export markets: Taiwan, South Korea, Malaysia and advanced electronics supply-chain hubs.

A simple sector risk map

Sensitivity Analysis
Indicative Asian exporter sensitivity to US consumer demand
Note: This is a general framework only. Sensitivity may vary by company, customer mix, contract structure and end market exposure.

Why timing matters

The US and Asia timelines may not line up.

A US retailer may still be selling older inventory, so the tariff impact can be delayed. Margins may hold in one quarter, then weaken as new tariffed inventory becomes a larger share of the sales mix.

An Asian exporter may see the pressure earlier if US buyers reduce orders before the cost hit appears in US consumer prices.

That creates a split earnings map:

  • US side: delayed margin pressure.
  • Asia side: earlier volume pressure.
  • Policy side: tariff exemptions, pauses or escalations can change the setup quickly.

The mistake is assuming a clean and immediate tariff impact. A strong US retailer result does not automatically mean tariff pressure is gone. It may only mean older inventory is still flowing through. A stable Asian exporter margin does not automatically mean demand is healthy. Volumes may be weakening beneath the surface.

The trap in the earnings season

What to watch next

On the US side, gross margins, inventory commentary, same-store sales and second-half guidance may provide useful context.

On the Asia side, export volumes, factory utilisation, order backlogs, working capital and capital expenditure guidance may be more relevant.

Across both regions, tariff policy remains the swing factor. Exemptions, pauses or new restrictions could quickly change market expectations.

Sector charts may provide additional context on whether market pricing is aligning with the earnings narrative, but they should be read alongside company commentary and macro data from the economic calendar.

FAQ

Frequently asked questions

How do tariffs affect US companies and Asian exporters differently? +
Tariffs may affect US companies through margin pressure and Asian exporters through volume pressure. US companies may face higher import costs, while Asian exporters may face fewer orders from US buyers.
Which Asian export sectors are most exposed to US demand? +
Textiles, apparel and basic consumer goods are generally more exposed to US demand because they are closely tied to retail orders and consumer spending. Electronics assembly and machinery are moderately exposed, while semiconductors and AI hardware may be less directly exposed.
Why can tariff impacts show up later in retailer earnings? +
Retailers may still be selling older inventory purchased before tariffs applied. The impact may become more visible later as new tariffed inventory moves through sales and margins.
What should investors watch in tariff-related earnings reports? +
General signals include gross margins, inventory commentary, same-store sales, export volumes, factory utilisation, order backlogs and management commentary on pricing or sourcing.
Are semiconductors and AI hardware exposed to tariffs? +
They may be less directly exposed to US consumer demand, but they can still be affected by policy restrictions, export controls, global capex cycles and valuation expectations.
Bottom Line

The tariff story is no longer only about who pays. It is about where the earnings pressure shows up first.

By
GO Markets

Disclaimer: Articles are from GO Markets analysts and contributors and are based on their independent analysis or personal experiences. Views, opinions or trading styles expressed are their own, and should not be taken as either representative of or shared by GO Markets. Advice, if any, is of a ‘general’ nature and not based on your personal objectives, financial situation or needs. Consider how appropriate the advice, if any, is to your objectives, financial situation and needs, before acting on the advice. If the advice relates to acquiring a particular financial product, you should obtain our Disclosure Statement (DS) and other legal documents available on our website for that product before making any decisions.

Related Articles

Market insights
Trading strategies
Shares and Indices
Fundamental analysis
Forex
The tariff impact: The US margin problem vs Asia’s volume problem
Tariffs do not hit every company the same way. For US retailers and consumer brands, the first pressure point is usually margin.
GO Markets
•
May 25, 2026
Glossary
Psychology
Market insights
Trading
What is a crowded trade and why should traders understand it?
Every so often, a market move catches traders off guard. Not because the news was surprising, but because many traders were already positioned the same way.
GO Markets
•
May 24, 2026
Geopolitical events
CFDs
Psychology
Trading strategies
Strategies
How to read the VIX: Signals, scenarios and risk traps
This is the second part of the GO Markets VIX Playbook. The first piece covered the basics and explored what the VIX measures, what it does not, why traders watch it and where new traders most often misread it. If you skipped it, start there as the foundation matters.
GO Markets
•
May 11, 2026
See all News & Insights

Recent Articles

Tariffs may affect US and Asian companies in different ways. For US retailers and consumer brands, the pressure often appears through margins as import costs rise and pricing power is tested. For Asian exporters, the pressure may appear through lower order volumes if US buyers reduce demand. Textiles, apparel and basic consumer goods are generally more sensitive to US demand, while semiconductors and AI hardware may be less directly exposed to US consumers but still carry policy, capex and valuation risks.
Market insights
Trading strategies
Shares and Indices
Fundamental analysis
Forex
The tariff impact: The US margin problem vs Asia’s volume problem

Tariffs do not hit every company the same way.

For US retailers and consumer brands, the first pressure point is usually margin. Import costs rise, but companies may not be able to pass the full increase to customers straight away. Older inventory can also delay the impact, which means the first earnings result may look calmer than the next one.

For Asian exporters, the pressure often looks different. The issue may not be a sharp price cut. It may be fewer orders from US buyers.

Tariffs, earnings and the Asia versus US split | GO Markets

Same tariff. Different earnings hit.

That is the key split for traders watching this earnings season. The US side is mainly about margin timing. The Asia side is about demand sensitivity. Not every export sector carries the same level of US demand risk.

TL;DR

  • US companies may face margin pressure as tariffed inventory moves through earnings.
  • Asian exporters may face volume pressure if US buyers reduce orders.
  • The timing is different: US retailers may feel the impact later, while Asian exporters may see it earlier through weaker order books.
  • Textiles, apparel and basic consumer goods are likely more sensitive to US demand.
  • Semiconductors and AI hardware may be less directly exposed to US consumers, but still carry policy, capex and valuation risk.

The big picture

Tariffs are paid at the US border by importers. From there, the cost can move through the system in several ways: higher prices, weaker margins, lower supplier prices, lower demand or a mix of all four.

Research cited by the Kiel Institute and New York Fed suggests US buyers and businesses may be absorbing a significant share of the tariff burden. That matters because it changes where the earnings pressure shows up first.

For a US retailer, the problem is straightforward but uncomfortable. If the company raises prices, demand may weaken. If it absorbs the tariff cost, margins may compress. If it still has older inventory, the hit may not show up immediately.

For an Asian exporter, the pressure can arrive through a different channel. If US buyers become cautious, they may order less. The exporter may keep prices relatively stable, but factory utilisation falls, fixed costs are spread across fewer units and earnings pressure builds.

That is why this is not just a tariff story. It is an earnings timing story.

US companies: the margin problem

The US side of the tariff story is about cost absorption.

Retailers, apparel brands, consumer electronics sellers and appliance companies often rely on imported goods, components or packaging. When tariff costs rise, they may try to protect margins through price increases, supplier negotiations, sourcing changes or inventory management.

The challenge is that none of these are clean solutions.

Price increases can test consumer demand. Supplier negotiations may take time. Sourcing changes can be expensive or slow. Inventory timing can make the first result look better than the underlying cost trend.

This is why earnings calls matter. Management commentary around pricing actions, tariff mitigation, sourcing, vendor negotiations and inventory timing may reveal more than headline sales growth.

What to watch on the US side

These signals may provide useful context in upcoming earnings reports:

If margins hold while sales remain stable, companies may be managing the pressure. If sales rise but margins fall, tariff costs may not be passing through cleanly. If guidance becomes more cautious, the market may start pricing a delayed earnings impact.

Asian exporters: the volume problem

The Asia side is not always about exporters cutting prices.

In many categories, Asian suppliers operate in competitive global markets with limited pricing power. If US buyers reduce orders, exporters may feel the impact through lower volumes rather than lower unit prices.

That distinction matters.

A company can report stable prices and still face earnings pressure if factories are running below normal utilisation. Lower volumes can reduce operating leverage, delay capital expenditure and weaken guidance.

The highest-risk sectors are usually those most closely tied to US retail demand, seasonal buying cycles and low-margin production.

Which Asian sectors are most exposed?

1. Textiles and apparel +
Highest Sensitivity

Textiles and apparel are among the clearest examples of US demand exposure.

These exporters are often tied directly to US retail orders, private-label contracts and seasonal buying cycles. If US retailers turn cautious, orders can be delayed, reduced or cancelled relatively quickly.

Risk is higher because margins are often thin, production is labour-intensive and buyers may have more power in negotiations.

Relevant export markets: Vietnam, Bangladesh, India, Indonesia and parts of China.
2. Basic consumer goods +
High Sensitivity

This includes toys, household goods, furniture, simple appliances and other discretionary or semi-discretionary exports.

These categories are exposed when US retailers reduce inventory or when consumers pull back from non-essential spending. Tariffs can add pressure if buyers try to push costs back onto suppliers.

Relevant export markets: China, Vietnam, Thailand, Malaysia and Indonesia.
3. Electronics assembly +
Medium to High Sensitivity

Electronics assembly is more mixed.

Lower-end consumer electronics can be sensitive to US household demand. Higher-value components or enterprise-linked electronics may be more resilient, depending on end-market exposure.

This sector can also be harder to read because supply chains are complex. A company may look like a technology exporter, but its actual earnings sensitivity may still depend on US consumer replacement cycles.

Relevant export markets: China, Vietnam, Malaysia, Thailand, Taiwan and the Philippines.
4. Machinery and industrial goods +
Medium Sensitivity

Machinery is less directly tied to US consumer demand than apparel or household goods. The risk is more about business investment.

If US companies delay capital expenditure because tariff uncertainty rises, machinery orders may weaken. However, order books can provide some buffer, and specialised products may have more pricing power.

Relevant export markets: Japan, South Korea, China, Taiwan and Singapore.
5. Semiconductors +
Lower Direct Sensitivity

Semiconductors are less directly exposed to US retail demand than textiles or consumer goods. Demand is often tied to broader technology cycles, autos, industrials, cloud infrastructure and AI investment.

That does not make the sector risk-free. Tariffs, export controls, geopolitics and a weaker global capex cycle can still affect earnings expectations.

Relevant export markets: Taiwan, South Korea, Malaysia, Singapore and parts of China.
6. AI hardware and data-centre supply chains +
Lowest Direct Sensitivity

AI hardware is more tied to cloud capital expenditure and data-centre buildouts than day-to-day consumer spending.

The risk is different. It is less about US shoppers buying fewer goods and more about whether AI capex expectations remain realistic, whether policy restrictions expand and whether valuations already price in strong growth.

Relevant export markets: Taiwan, South Korea, Malaysia and advanced electronics supply-chain hubs.

A simple sector risk map

Sensitivity Analysis
Indicative Asian exporter sensitivity to US consumer demand
Note: This is a general framework only. Sensitivity may vary by company, customer mix, contract structure and end market exposure.

Why timing matters

The US and Asia timelines may not line up.

A US retailer may still be selling older inventory, so the tariff impact can be delayed. Margins may hold in one quarter, then weaken as new tariffed inventory becomes a larger share of the sales mix.

An Asian exporter may see the pressure earlier if US buyers reduce orders before the cost hit appears in US consumer prices.

That creates a split earnings map:

  • US side: delayed margin pressure.
  • Asia side: earlier volume pressure.
  • Policy side: tariff exemptions, pauses or escalations can change the setup quickly.

The mistake is assuming a clean and immediate tariff impact. A strong US retailer result does not automatically mean tariff pressure is gone. It may only mean older inventory is still flowing through. A stable Asian exporter margin does not automatically mean demand is healthy. Volumes may be weakening beneath the surface.

The trap in the earnings season

What to watch next

On the US side, gross margins, inventory commentary, same-store sales and second-half guidance may provide useful context.

On the Asia side, export volumes, factory utilisation, order backlogs, working capital and capital expenditure guidance may be more relevant.

Across both regions, tariff policy remains the swing factor. Exemptions, pauses or new restrictions could quickly change market expectations.

Sector charts may provide additional context on whether market pricing is aligning with the earnings narrative, but they should be read alongside company commentary and macro data from the economic calendar.

FAQ

Frequently asked questions

How do tariffs affect US companies and Asian exporters differently? +
Tariffs may affect US companies through margin pressure and Asian exporters through volume pressure. US companies may face higher import costs, while Asian exporters may face fewer orders from US buyers.
Which Asian export sectors are most exposed to US demand? +
Textiles, apparel and basic consumer goods are generally more exposed to US demand because they are closely tied to retail orders and consumer spending. Electronics assembly and machinery are moderately exposed, while semiconductors and AI hardware may be less directly exposed.
Why can tariff impacts show up later in retailer earnings? +
Retailers may still be selling older inventory purchased before tariffs applied. The impact may become more visible later as new tariffed inventory moves through sales and margins.
What should investors watch in tariff-related earnings reports? +
General signals include gross margins, inventory commentary, same-store sales, export volumes, factory utilisation, order backlogs and management commentary on pricing or sourcing.
Are semiconductors and AI hardware exposed to tariffs? +
They may be less directly exposed to US consumer demand, but they can still be affected by policy restrictions, export controls, global capex cycles and valuation expectations.
Bottom Line

The tariff story is no longer only about who pays. It is about where the earnings pressure shows up first.

GO Markets
•
May 25, 2026
Miniature shopping cart with sale sign and climbing stick figures, consumer rush and retail promotion concept.
Glossary
Psychology
Market insights
Trading
What is a crowded trade and why should traders understand it?

Every so often, a market move catches traders off guard. Not because the news was surprising, but because many traders were already positioned the same way.

One piece of data shifts the mood and what follows is not an orderly re-evaluation. It is a rush for the exit. Prices move faster than fundamentals alone would suggest. Stops are triggered. Margin calls follow.

That is the risk behind a crowded trade.

It can be an underestimated risk in financial markets and is a useful concept for traders to understand.

This playbook explains how crowded trades form, why they can become fragile and what traders may monitor before market conditions become difficult.

Use it as a starting point, then practise the concepts on charts, watchlists and demo tools before applying them in live conditions.
01
Part One The 101 Explainer — Building understanding

What is a crowded trade?

A crowded trade is a market position where a large number of investors or traders hold the same asset, in the same direction, for the same reason, at the same time.

Think of it like a footbridge. A few people walking across creates no problem. But if hundreds of people rush onto it at once, then all try to run back at the same time, the structure comes under extreme stress. Markets can work in a similar way.

In professional markets, crowding is viewed as a form of endogenous risk. That means the risk does not come from the asset's own fundamentals. It comes from the market's internal structure: too many participants holding the same position, with too little liquidity available to absorb them all if they try to exit at once.

This is not the same as a popular or well-researched trade. A crowded trade becomes dangerous when the collective position may be too large for the market to handle cleanly if sentiment shifts.

Why this matters to new traders

New traders often focus on whether an asset may rise or fall. Crowded-trade analysis adds a different question: what happens if everyone holding the same view tries to exit at once?

This matters for several practical reasons.

  • Price moves in crowded markets can be faster and more volatile than fundamentals alone would explain. When a catalyst triggers a mass exit, the selling or buying can feed on itself.
  • Spreads, which are the difference between the buy and sell price, can widen sharply during crowded unwinds. This can affect trading costs for contracts for difference (CFD) traders.
  • Stop-losses, which are orders designed to exit a position if the price moves too far against the trader, can be triggered in large numbers, accelerating the move further.
  • Margin calls, which can occur when a leveraged position loses more than the available margin allows, may force positions to close at the worst possible time.

For CFD traders, these dynamics matter because leverage can magnify both gains and losses. A crowded unwind can move against a position quickly and with little warning. That makes preparation and risk controls especially important.

The key terms to know

Term Plain-English explanation Why it matters to traders
Crowded trade A position where many market participants hold the same asset in the same direction for the same reason. Helps traders assess risk that is not visible in the price chart alone.
Endogenous risk Risk that comes from within the market's own structure, not from external events. Crowding is endogenous because it creates fragility before any catalyst appears. It can be harder to spot than news-driven risk, which is why it can catch traders off guard.
Liquidity How easily an asset can be bought or sold without moving the price significantly. Thin liquidity means large orders can shift prices sharply. In crowded unwinds, liquidity can change quickly, making exits more expensive.
Stop-loss An order that automatically closes a position if the price moves against the trader by a set amount. It is used to help limit losses. Stop-loss triggering during a crowded exit can amplify price moves.
Margin call A demand from a broker to deposit more funds because a leveraged position has lost value. If unmet, the broker may close the position. Margin calls can force traders out of positions during unfavourable market conditions.
Short squeeze A short squeeze occurs when traders who have sold an asset short are forced to buy it back quickly because the price rises sharply, which can accelerate the upward move. Recognising a potential short squeeze can be part of the crowded trade playbook.
COT report A weekly report published by the US Commodity Futures Trading Commission (CFTC) showing how large professional traders are positioned in futures markets. The COT report can help traders monitor whether a trade may be becoming crowded.
Days-to-cover (DTC) A ratio measuring how long it would take short sellers to repurchase all their shorted shares, based on average daily trading volume. Some market participants may view higher DTC readings as one possible sign of elevated short-squeeze risk.

How the fragility builds

A crowded trade usually begins for a legitimate reason. A new technology emerges, a commodity looks undersupplied or a currency is supported by a widening interest rate gap.

Capital flows in. Prices rise. More traders are drawn in by the momentum. Institutional funds build large positions. The trade begins to appear in a growing number of portfolios, often without participants knowing how many others are doing the same thing.

The deceptive part is that the trade often performs well during the accumulation phase. The logic holds. The price moves in the expected direction. That can be reinforcing.

Concentration risk

S&P 500 mega-cap weighting and estimated days to exit

Illustrative data
Note: Illustrates how a higher index weighting may coincide with a longer estimated exit period, based on standard days to average daily volume calculations.

The risk is that liquidity does not necessarily grow at the same rate as the collective position. At some point, the trade may become too large relative to what the market can absorb if many participants try to leave at once.

One way to monitor this is the Days-ADV metric, which estimates how many days of average daily trading volume it would take for institutional holders to fully exit their collective position.

The catalyst and the exit problem

A crowded trade does not always unwind because the original thesis was wrong. It can unwind because a negative catalyst, even a small or ambiguous one, changes the calculation for a critical mass of holders.

Enough holders decide to exit. If liquidity cannot absorb the selling, prices can fall sharply, triggering more stop-losses and margin calls.

Professionals often describe this as a non-linear price move. The market may not reprice gradually. It can move in ways that appear extreme compared with normal conditions, but occur more often in crowded markets than many traders expect.

Unwind dynamics

Volatility pattern around a market catalyst

Stylised example
Mechanism: The chart shows a prolonged low-volatility phase followed by a faster repricing after the catalyst point is reached.

The other side: under-owned assets

The crowded trade framework also has a flip side.

When capital floods into a small number of popular assets, others may become relatively under-owned. With less speculative enthusiasm built into prices, a positive structural shift, supply disruption or change in sentiment may trigger a sharp repricing as under-positioned investors move to build exposure.

This dynamic can appear in commodities such as crude oil and gold when speculative positioning becomes relatively low compared with recent or longer-term ranges. In those conditions, a supply disruption, geopolitical event or shift in demand expectations may trigger faster repositioning than the market had been pricing.

Positioning risk

COT speculative positioning and commodity price action

COT tracking
Market context: Lower speculative positioning may indicate lighter participation. If sentiment or news flow shifts, price moves can become more sensitive to changes in positioning and liquidity.

The part many new traders miss

The most common misunderstanding is confusing a strong story with a structurally safe trade. A compelling narrative about why an asset may keep rising is not the same as a well-sized, liquid, uncrowded position. In fact, the stronger the story, the more likely the trade has already attracted a large number of participants. That can make the structure more fragile.

Key insight

New traders often look at a crowded asset and see confirmation. They see that many experienced, well-resourced investors hold the same position. They interpret this as validation.

The more crowded the trade, the more carefully risk needs to be managed, because the exit problem grows with every new entrant.

The other part of the story is: who else is already here, and what happens if they all leave at once?

02
Part Two The Practical Playbook — Preparing, monitoring and managing risk

The trader's watchlist

Traders monitoring crowded trade dynamics may consider tracking these signals.

Positioning
COT positioning extremes
Historically extreme net-long or net-short readings, or rapid week-to-week changes, may suggest crowding risk.
Equities
Mega-cap index concentration
When a small number of stocks account for a disproportionate share of major index weightings, the index may be more vulnerable if those stocks come under pressure together.
Short interest
DTC ratios
Some market participants may view higher DTC readings, including those above five days, as one possible sign of elevated short-squeeze risk.
Commodities
Under-owned commodities
When speculative positioning in commodities such as Brent crude or gold falls toward the lower end of recent ranges, those assets may be relatively under-owned.
Valuations
Valuation dispersion
A widening gap between the most expensive and least expensive stocks in an index can signal that capital has crowded into a narrow set of names.
Volatility
Volatility indices
Unusually low volatility can accompany the late stages of a crowded accumulation phase. Monitor VIX and VVIX alongside positioning data.
Execution
Spread widening
During crowded unwinds, spreads can widen significantly. Watching spreads during data releases or market stress may help inform execution planning.

Practical preparation points

Before monitoring a market ▼
  • Traders may identify which markets are showing extreme COT positioning, either historically crowded long or crowded short.
  • Traders might consider checking DTC readings for assets being assessed, particularly on the short side, and consider them alongside liquidity, short interest and broader market conditions.
  • Traders may review the economic calendar to identify upcoming data releases that could act as catalysts.
  • Traders might mark key support and resistance levels on the chart, and consider where a crowded unwind could pause or accelerate.
  • Traders may review margin requirements for the instruments being monitored to understand how much adverse movement a position could absorb.
  • Traders might define in advance what would change their view. If monitoring a crowded long, they may ask what data or event would suggest the unwind is underway.
During a market move ▼
  • Traders may consider avoiding reacting to the first headline alone, as crowded trade unwinds can reverse sharply in the early stages before resuming.
  • Traders might monitor whether related markets are confirming the move. A gold sell-off confirmed by falling risk appetite across commodities could be more informative than one happening in isolation.
  • Traders may watch spreads, as widening spreads during a fast move can make execution significantly more expensive than expected.
  • Traders might avoid increasing position size during a fast move, as volatility at the start of a crowded unwind can be extreme.
  • Traders may note whether the COT report is shifting in the direction of the price move, or lagging.
After the move ▼
  • Traders may review what happened against their scenario plan, asking whether the move fit the prepared framework.
  • Traders might consider saving charts and annotating observations about speed, spread behaviour and any stop-cascade signals.
  • Traders may update their watchlist and assess whether the trade remains crowded or whether positioning has normalised.
  • Traders might review any emotional decisions made during the move, noting whether urgency or fear influenced their process.

Common mistakes to avoid

Mistake What happens How to manage the risk
Assuming popularity equals safety Many experienced investors holding the same position can feel like validation. In reality, it can also make the exit more difficult. Separate the quality of the thesis from the structural risk created by concentration.
Ignoring spread and liquidity conditions Spreads can widen significantly during crowded unwinds, making exits more expensive than expected. Factor spread costs into risk planning, especially for leveraged CFD positions.
Moving stops emotionally during fast moves Volatility during a crowded exit can feel extreme. Traders sometimes move stops wider to avoid being stopped out, which can increase risk. Define stop placement before the move, not during it.
Confusing the catalyst with the cause The news event that triggers a crowded unwind is often not the full reason for the move. The deeper cause may be structural overcrowding. After a sharp move, ask whether the catalyst alone would have caused this reaction in a less crowded market.
Forgetting that no framework works every time COT extremes can persist. Under-owned assets can stay under-owned. Crowded trades can continue working. Use crowded trade analysis as one input, not the only signal.

The emotional trap to watch

Psychology

"The trap is believing that urgency equals opportunity."

Sometimes it does. Often, it simply means the market has already moved.

The crowded-trade trap is usually a mix of fear of missing out (FOMO) and confirmation bias. FOMO draws traders into crowded positions late, when the story feels strongest and the price action looks most inviting. Confirmation bias then makes it harder to notice information that challenges the trade.

The question to ask before acting: is this urgency the result of new information, or the result of watching the price move?

The practical habit that may help: before entering a position that has already moved significantly, write down one specific reason the trade could be wrong. If one does not come to mind, that is worth taking seriously.

Beginner checklist before acting

Tick through this before any volatility-aware trade decision.

GO Markets
•
May 24, 2026
AI
Shares
Setups
Strategies
Google TPU chips and NVIDIA: What the AI chip war means for markets

For the past three years, investing in artificial intelligence (AI) infrastructure has followed a relatively simple logic: find the companies building the picks and shovels for the gold rush. At the top of that list sat one name: NVIDIA.

Its chips power many of the world’s AI models. Its software ecosystem keeps developers loyal. Its stock has been one of the most dramatic wealth-creation stories in a generation.

Then Google walked into a conference room in Las Vegas and signalled it was getting serious about making its own silicon available to the world.

Here is what happened, and why it matters for investors.

When Google Draws Its Sword — AI Chip War Analysis | GO Markets
The TPU glossary, key terms
TPU
Tensor Processing Unit. Google’s custom chip built specifically for AI mathematics, not general graphics.
GPU
Graphics Processing Unit. NVIDIA’s chip, originally built for gaming, now the dominant AI training hardware.
Inference
Running a model in the real world. Different, cheaper and increasingly important compared with training.
CUDA
NVIDIA’s software layer. The real competitive moat. Millions of developers are locked in by code, not chips.

What Google announced

At Google Cloud Next 2026 in Las Vegas, Google made two distinct announcements. It confirmed general availability of Ironwood, its seventh-generation TPU, the first purpose-built for what Google calls the “agentic era” of inference at scale. It also previewed its eighth-generation architecture: two purpose-built chips, TPU 8t for large-scale training and TPU 8i for high-speed inference, both targeting TSMC 2nm manufacturing and expected to reach general availability later in 2026.

A TPU is Google’s custom alternative to NVIDIA’s graphics processing unit (GPU). Where a GPU is a general-purpose workhorse, a TPU is a specialist chip built from the ground up for AI calculations. Google has been building them since 2016. The eighth generation is its most ambitious split yet, and the first time the company has designed separate chips for each half of the AI lifecycle.

The TPU 8t training pod reportedly delivers nearly three times the compute of an equivalent Ironwood pod, with double the performance per watt. The TPU 8i inference chip is designed to serve millions of AI agents simultaneously for enterprise customers.

That last part carries a structural implication. On a recent earnings call, CEO Sundar Pichai indicated that as TPU demand grows from AI labs, capital markets firms and high-performance computing applications, Google would begin delivering TPUs to select customers in their own data centres. Google is no longer content to keep its silicon advantage internal.

Google is no longer just a TPU user. It is becoming a TPU vendor, and its biggest customers are already signed up.

Anthropic’s compute strategy

Anthropic, the AI company behind Claude, has confirmed a major infrastructure deal with Google covering access to up to one million Ironwood TPU chips. The commitment is worth tens of billions of dollars and was formally announced by both companies.

Understanding that deal requires understanding Anthropic’s compute strategy in full.

Compute Infrastructure Strategy
Amazon Trainium
Anthropic’s primary cloud and training partner is Amazon. Project Rainier, Anthropic’s frontier model supercluster, runs on Trainium 2 chips across multiple US data centres. Anthropic has committed to up to 5 gigawatts of current and future Trainium capacity.
Google TPU
Confirmed deal for up to one million Ironwood chips, plus 3.5 gigawatts of further TPU capacity from 2027. Anthropic has used Google TPUs since 2023 and described strong price-performance efficiency as the driver for the expansion.
NVIDIA GPU
Third pillar of Anthropic’s diversified infrastructure. NVIDIA GPUs support research, specialist workloads and certain training runs. The multi-platform strategy is deliberate, with no single-vendor lock-in, optimising each dollar of compute.

The multi-platform picture matters because coverage elsewhere has occasionally characterised this Google deal as Anthropic “switching” from NVIDIA. That framing understates the deliberate architecture of Anthropic’s compute strategy. The Google deal is an expansion, not a departure from either AWS or NVIDIA.

Why this matters beyond the benchmark

On a per-chip basis, the current generation comparison is closer than the headlines suggest. Ironwood, now in general availability, delivers approximately 4.6 petaflops of FP8 computing power. NVIDIA’s Blackwell B200 delivers roughly 4.5 petaflops at FP16, although cross-precision comparisons require care, as the two figures are not measured on an identical scale.

But benchmark comparisons miss the bigger story.

At pod scale, where these chips are actually deployed, the gap widens. An Ironwood superpod of 9,216 chips delivers 42.5 exaflops. The eighth-generation TPU 8t pod, at 9,600 chips, targets 121 exaflops at FP4 precision. Google also claims near-linear scaling to one million chips inside a single logical cluster. For hyperscalers running hundreds of thousands of chips simultaneously, pod-level economics matter far more than per-chip benchmarks.

Performance benchmark
Chip comparison: compute and efficiency
Precision note: direct comparison requires caution. Ironwood compute is measured at FP8, NVIDIA B200 at FP16, TPU 8t pod figures at FP4. Halving FP4 figures provides a rough FP8 equivalent. Performance per watt is indexed to NVIDIA H100 baseline 100 and reflects Google’s published claims versus Ironwood, not independent verification. Benchmark results vary by workload and conditions.

The NVIDIA position

NVIDIA currently controls an estimated 81% of the AI data centre chip market, according to IDC. That is an extraordinary concentration of market power, and the near-term demand picture has remained resilient.

Recent analyst expectations have pointed to strong NVIDIA earnings growth, supported by elevated demand for AI infrastructure and broad adoption of the Blackwell platform. NVIDIA has itself guided for a combined US$1 trillion in Blackwell and upcoming Vera Rubin orders across 2026 and 2027.

AMD is developing rack-scale server systems and has gained meaningful ground. Estimates from analysts, including IDC, suggest AMD may now hold approximately 10% of the AI accelerator market, up from low single digits two years ago. Amazon and Google continue to expand custom chip businesses. The combined chip operations at Amazon alone, covering Trainium, Graviton and Nitro, have crossed a US$20 billion annual revenue run rate, growing at triple-digit percentages year over year, with nearly 40% sequential growth in Q1 2026.

The bull case for NVIDIA remains clear: demand has stayed strong, and NVIDIA’s ecosystem remains deeply embedded across the AI compute stack.

The longer-term question is less about near-term earnings and more about pricing power in the next upgrade cycle. Every reporting period in which Google, Amazon and Microsoft gain confidence in their own silicon is another data point in that debate. The incentive structure is powerful: these companies have every reason to reduce dependency on a single supplier, and the capital to act on it.

Market structure
AI data centre chip market share, estimated 2026
Estimated share of AI accelerator revenue. Custom silicon has grown from near zero three years ago. AMD share estimates vary across methodologies. Recent analyst estimates range from 4% to 10%.
Source note: IDC estimates, Silicon Analysis, public disclosures, company filings. Figures are approximations, subject to significant revision depending on methodology and market definition.

Stocks and sectors to watch

For NVIDIA, the near-term earnings story and the longer-term competitive story are pulling in different directions. Strong results may validate the current cycle. But the structural dynamic, where major customers build their own silicon, is unlikely to reverse.

For Alphabet, the Ironwood general availability and eighth-generation preview represent a potential monetisation opportunity well beyond advertising. Google Cloud grew 63% year over year in Q1 2026, among the fastest growth rates of any major hyperscaler. TPU-as-a-service, with confirmed anchor customers including Anthropic and Meta, could extend that runway materially if enterprise inference workloads continue migrating to Google infrastructure.

The less obvious plays are in the supply chain. TPU 8t and 8i are both targeting TSMC 2nm manufacturing, with Broadcom designing the training chip and MediaTek the inference chip. TSMC may remain a critical enabler regardless of which chip architecture gains share in each cycle, as may advanced packaging suppliers, liquid cooling companies and data centre real estate investment trusts (REITs).

Power infrastructure, liquid cooling suppliers and data centre REITs may also be exposed to sustained capital expenditure growth. Combined hyperscaler capital expenditure from the four major cloud providers is tracking toward US$700 billion or more in 2026, nearly double the US$388 billion spent in 2025. That scale of investment, sustained over multiple years, represents a different kind of macro signal.

Supply chain plays: If neither NVIDIA nor Google “wins” the chip war outright, infrastructure may still benefit. TSMC already manufactures both Ironwood and the upcoming eighth-generation chips. Advanced packaging suppliers, liquid cooling companies and data centre REITs may benefit regardless of which chip architecture gains share in each cycle.

CFD trader focus
Key instruments to watch
NASDAQ 100
The primary read-across from major NVIDIA and hyperscaler results. Earnings surprises, in either direction, tend to move the index broadly.
USD/CNH
Captures tariff and trade policy sensitivity. Ongoing uncertainty has kept spreads elevated and positioning cautious.
US10Y
The 4.5% level has served as a reference point for technology valuations. A stronger-than-expected guidance print could be a catalyst worth monitoring.
General commentary only. This is not a trading signal or personal financial advice. CFD trading involves significant risk of loss. Past performance is not indicative of future results.

Where the risks sit

Higher AI infrastructure spending does not automatically translate into stock gains. Several factors complicate a straight line from “chip war” to “buy everything”.

Valuation risk
NVIDIA’s valuation has reflected high expectations for future growth. Any guidance shortfall, margin compression or sign of slower infrastructure demand could trigger a reassessment that would be felt across the sector.
The CUDA moat
NVIDIA’s deepest competitive advantage is not just hardware. It is the software ecosystem that millions of developers build on, and a decade of investment in libraries, tooling and workflows. Google’s TorchTPU initiative is a deliberate attempt to lower that switching cost, but ecosystems are slow to move. This is the risk that is easiest to underestimate.
Execution risk
Google has announced impressive chips before. Ironwood is now generally available. The eighth-generation chips are previewed. Delivering them at scale, on time, to external enterprise customers with commercial-grade service commitments is a different challenge from announcing impressive specifications.
Market share versus revenue
As AMD, Google and Amazon gain share, NVIDIA’s percentage of a rapidly growing market may fall even as its absolute revenue continues to grow. Investors evaluating the competitive thesis should distinguish between share erosion and revenue impact. They are not the same.

What investors might take away

The AI chip war is not a story of one winner and one loser. It is a story of a market that is too large and too strategically important for any single company to own indefinitely.

NVIDIA built its lead through genuine technical excellence and a decade of software investment. That lead is real, and near-term earnings are likely to continue reflecting it.

But the challengers are no longer startups with benchmark slides. They are trillion-dollar companies with their own silicon, their own cloud infrastructure and every incentive to reduce their dependency on a single supplier, along with the capital expenditure commitments to show they are following through.

One way to frame the longer-term question is this: demand for AI compute may not be the primary variable for investors to focus on. Who captures the margin from that demand, and at what valuation multiple, may matter just as much. Those are questions each investor may need to weigh against their own risk profile and objectives.

Scenario Disclaimer: The "Next 30 days" and "Next 3 months" scenarios are illustrative "what-if" models for stress-testing a market thesis and identifying potential catalysts. They do not constitute a house view, forecast, guarantee, or prediction of future market movement. Any Brent price targets, Fed policy references, or other market benchmarks are hypothetical only. Real-world conditions are subject to volatility and unforeseen shifts.

GO Markets
•
May 20, 2026
See all News & Insights