NEW YORK/LONDON — Donald Trump’s tariffs were supposed to bring factories home. Brexit was supposed to restore national economic control. Washington and Beijing have spent years trying to reduce strategic dependence on each other.
Yet the global economy is still becoming more interconnected.
Not in exactly the same way.
Not through exactly the same countries.
And increasingly not without political friction.
But the data tell a very different story from the repeated predictions that globalization is dying.
Global merchandise trade is expanding.
Foreign investment has begun recovering.
International tourism and migration flows have rebounded.
Multinationals still earn huge portions of their revenues abroad.
And some of the fastest-growing products in world trade are the chips, servers, networking equipment and other components needed to build artificial-intelligence infrastructure.
In other words:
globalization has not disappeared. It has adapted.
Globalization remains near a record level
One of the strongest pieces of evidence comes from the DHL Global Connectedness Report, produced with researchers at New York University’s Stern School of Business.
The study tracks millions of data points covering four broad categories:
trade;
capital;
information;
and people.
Its 2026 edition concluded that global connectedness remains at historically high levels and has changed little since reaching a record in 2022.
The report explicitly found no broad shift from international activity back toward purely domestic activity.
That is remarkable given everything that happened in between:
U.S.-China trade wars;
Russia’s invasion of Ukraine;
Brexit;
pandemic-era supply disruptions;
Trump’s return to the White House;
higher tariffs;
and a renewed push for industrial policy.
The world economy did not deglobalize — companies rerouted
This is the crucial distinction.
If the United States imports fewer products directly from China, that does not automatically mean Americans are consuming fewer globally produced goods.
Companies can shift final assembly.
A product that was once shipped directly from Shenzhen to California might now be:
partly manufactured in China;
assembled in Vietnam;
packaged in Malaysia;
and then exported to the U.S.
Trade has not disappeared.
The route changed.
The DHL data show the U.S. and China are genuinely decoupling in their direct bilateral relationship.
U.S.-China trade fell to about 2% of total global trade, down from 2.7% only a few years earlier.
The U.S. share of imports coming directly from China has also fallen dramatically.
But analysis of the Chinese components embedded inside goods arriving from third countries shows no equally clear collapse in underlying Chinese exposure.
That means some decoupling is real—
and some is rerouting.
Trump’s tariffs changed trade more than they reduced it
President Trump’s tariff strategy has been extraordinarily aggressive.
Since returning to office in 2025, his administration has imposed or proposed tariffs involving:
China;
Mexico;
Canada;
Europe;
autos;
steel;
aluminum;
and numerous strategic goods.
Some measures have been revised through trade agreements or legal challenges.
Others remain in place.
The objective has been clear:
reduce dependence on foreign supply chains and encourage more manufacturing inside the United States.
Those tariffs have absolutely affected trade patterns.
But the evidence so far does not show global commerce collapsing.
Instead, businesses have adapted.
Countries outside the U.S.-China rivalry have often benefited.
Vietnam became one of the clearest winners
The first Trump trade war already showed how quickly supply chains can respond.
An IMF study of ASEAN found that several Southeast Asian economies experienced unusually strong export growth in products targeted by U.S.-China tariffs.
Vietnam stood out.
Foreign direct investment into tariff-exposed sectors rose particularly strongly, helping the country gain export market share.
That pattern has continued in broader form.
Companies have expanded manufacturing footprints across:
Vietnam;
India;
Malaysia;
Thailand;
Mexico;
Indonesia;
and other emerging markets.
The buzzword became “China plus one.”
The objective was not necessarily to abandon China.
It was to make sure China was no longer the only manufacturing base.
That is still globalization
This is where political language can become misleading.
Moving a factory from China to Vietnam may reduce dependence on China.
But it does not make the company less global.
It may actually make the supply chain more international.
A multinational may now source:
components from China;
semiconductors from Taiwan;
assembly in Vietnam;
software from the United States;
and sell the final product in Europe.
That is not deglobalization.
It is diversification.
Trade is now growing much faster than expected
The World Trade Organization’s latest forecast makes this especially clear.
On October 8, the WTO raised its estimate for global merchandise trade growth in 2026 to 3.9%.
Its March forecast had been only 1.9%.
The WTO now expects another 4.1% increase in 2027.
Those numbers are difficult to reconcile with the idea that world trade is collapsing.
Global GDP is expected to grow only around 2.6% this year.
That means goods trade is again expanding faster than overall economic output.
AI has become one of globalization’s biggest engines
The surprising driver is artificial intelligence.
The global AI boom is often discussed as a software story.
In reality, it requires an enormous amount of internationally traded hardware.
AI infrastructure depends on:
semiconductors;
memory;
optical networking equipment;
servers;
cooling systems;
power equipment;
and data-center components.
The WTO says trade in AI-related products surged around 67% year on year in the first half of 2026.
DHL’s latest globalization tracker goes even further.
It says AI-enabling goods accounted for 42% of global goods-trade growth in 2025 and as much as 76% in the first quarter of 2026.
That is extraordinary.
The technology now blamed for financial-market excess is simultaneously becoming one of the strongest forces supporting global trade.
Every AI query depends on a global supply chain
The reason is structural.
A U.S. AI company may build a model in California.
But the infrastructure behind that model can involve:
chips designed in the United States;
fabricated in Taiwan;
memory made in South Korea;
optical modules from China;
equipment from Japan and the Netherlands;
servers assembled in Mexico or Southeast Asia;
and data centers constructed around the world.
AI may be digitally delivered.
Its physical foundation is one of the most global supply chains ever created.
That helps explain why tariffs have not been able to reverse globalization easily.
Modern technology is too internationally interconnected.
Global goods trade had its strongest first half in years
DHL says goods trade grew faster in the first half of 2026 than in any comparable period during the previous 15 years apart from the extraordinary post-Covid rebound.
That happened despite:
high tariffs;
the U.S.-Iran conflict;
energy disruption;
shipping risks;
and geopolitical uncertainty.
Again, that does not mean protectionism is harmless.
Tariffs raise costs.
They distort investment.
They can reduce efficiency.
But businesses often respond by changing supply chains rather than abandoning cross-border commerce.
Companies would rather move than stop trading
This is perhaps globalization’s greatest source of resilience.
A company does not necessarily react to a tariff by saying:
“We will stop producing internationally.”
It often says:
“Where else can we produce this?”
That sends investment toward new countries.
Vietnam gains factories.
India gains electronics production.
Mexico benefits from nearshoring.
Southeast Asia attracts data-center and semiconductor investment.
Eastern Europe gains manufacturing.
The map changes.
The international system survives.
Trump’s trade deficit also shows how hard globalization is to unwind
One of the Trump administration’s central goals has been reducing America’s trade deficit.
Yet Reuters reported this week that the U.S. trade deficit recently widened to its highest level in 17 months, driven by record imports.
That does not mean tariffs have had no effect.
They clearly affect specific industries and bilateral flows.
But it demonstrates how difficult it is to reduce imports in a large consumer economy.
Americans continue buying:
electronics;
machinery;
clothing;
vehicles;
industrial inputs;
and consumer goods
produced through international supply chains.
Trade policy can change where those goods come from.
Changing whether consumers want them is harder.
Tariffs have also raised U.S. prices
A New York Federal Reserve study cited by Reuters estimated that Trump-era tariffs added nearly 3 percentage points to goods inflation, although the impact has begun fading.
That illustrates another difficulty with aggressive deglobalization.
Domestic production is not automatically cheaper.
Moving factories may:
increase labor costs;
require new infrastructure;
reduce scale efficiencies;
or simply shift imports to another foreign country.
Tariffs can therefore change economic geography without restoring the fully domestic manufacturing system politicians imagine.
Brexit tells a similar story
Britain left the European Union partly because supporters wanted greater national control over:
trade;
regulation;
immigration;
and economic policy.
Brexit undeniably created new barriers between Britain and its largest trading partner.
Goods exporters faced:
customs declarations;
regulatory checks;
rules-of-origin requirements;
and additional paperwork.
But Britain did not stop globalizing.
It changed its trade strategy.
The UK has since negotiated agreements with countries outside Europe, including a major new trade pact with India that entered into force in July 2026.
Britain is still deeply tied to Europe
Brexit also did not erase economic geography.
The EU remains a huge market sitting directly across the English Channel.
British companies continue trading heavily with European customers.
Official UK statistics show goods exports to the EU rose 8.5% in the three months to May 2026, while imports from the EU also increased.
In the three months to July, Britain’s combined goods-and-services exports rose by £6.1 billion.
Brexit changed the terms of trade.
It did not end trade.
UK services remain especially global
Britain’s largest international competitive strengths are increasingly in services:
finance;
insurance;
professional services;
technology;
education;
consulting;
and creative industries.
The UK continues running a huge services-trade surplus.
ONS data put that surplus at roughly £52.6 billion in the three months to July 2026.
Services can sometimes cross borders much more easily than physical goods.
A lawyer can advise a foreign client online.
A fund manager can handle global assets from London.
A designer can work for customers in several countries.
That form of globalization is less visible than container ships—
but economically enormous.
Capital is still moving across borders too
Trade is only one part of globalization.
Foreign investment is another.
UNCTAD says global foreign direct investment rose 6% in 2025 to $1.6 trillion, ending two consecutive years of decline.
Even after adjusting for unusually large flows through European financial centers, underlying investment still increased by roughly 4%.
That recovery is uneven.
Developed-country investment rose much faster than investment into developing economies.
But the core conclusion remains:
companies are not broadly retreating behind national borders.
They are still putting capital into foreign countries.
AI is reshaping FDI too
UNCTAD says a disproportionate share of new investment is going into strategic sectors.
Data centers.
Semiconductors.
Energy infrastructure.
Digital networks.
These have become magnets for cross-border capital.
The report says strategic sectors represented around 44% of global greenfield-project values, up from just 16% in 2020.
That demonstrates how globalization itself is changing.
Twenty years ago, cross-border investment was strongly associated with cheap manufacturing.
Today, it is increasingly associated with:
chips;
AI;
renewable energy;
batteries;
data centers;
and strategic infrastructure.
FDI is becoming more political
This is where the deglobalization argument does contain real truth.
Governments care much more about where investment comes from.
A Chinese acquisition of a European semiconductor company can trigger national-security reviews.
A U.S. company building technology infrastructure in China faces restrictions.
European governments increasingly screen strategic investment.
Washington limits outbound investment into sensitive Chinese technology.
So capital remains international—
but it is more politically filtered.
That is not the end of globalization.
It is the securitization of globalization.
Friendshoring is replacing unrestricted globalization
The old model prioritized efficiency.
Businesses asked:
Where is production cheapest?
The new model asks additional questions:
Is the country politically reliable?
Can sanctions disrupt supply?
Could war close the trade route?
Will tariffs suddenly change?
Is technology subject to export controls?
That has given rise to friendshoring.
Companies continue going abroad.
They just prefer countries viewed as politically safer.
Nearshoring follows the same logic
Mexico is one example.
For a U.S. company, manufacturing in Mexico can offer:
lower costs than domestic U.S. production;
geographic proximity;
trade-agreement access;
and shorter supply chains than Asia.
That is not reshoring.
It is nearshoring.
Again, globalization changes shape without disappearing.
Supply chains are becoming more redundant
For decades, corporate supply chains were optimized for efficiency.
One supplier.
One factory.
Minimum inventory.
Just-in-time logistics.
Covid changed that.
Then the Ukraine war reinforced the lesson.
Then Red Sea disruption.
Then tariffs.
Then the Iran conflict.
Businesses increasingly want backup suppliers and alternative routes.
That means more warehouses.
More inventory.
More suppliers.
More factories across different countries.
Ironically, the desire to reduce globalization risk can create more geographically diversified globalization.
“Just in case” is replacing “just in time”
The energy industry is making exactly the same transition.
Executives increasingly talk about moving from just-in-time infrastructure toward “just in case” systems.
That means:
more storage;
more pipelines;
more export terminals;
and alternative routes.
The objective is resilience.
But resilience often requires more international infrastructure, not less.
Shipping remains remarkably resilient
Global trade has also survived extraordinary logistical shocks.
The Red Sea has faced attacks.
The Strait of Hormuz has been disrupted.
Freight routes have been extended around Africa.
Yet Reuters reports European shipping and logistics companies are still expecting strong quarterly results because underlying demand for international freight remains robust.
Companies paid higher shipping costs.
They rerouted vessels.
They used premium logistics services.
They did not simply stop trading.
Globalization is expensive to unwind because companies already invested trillions
Modern supply chains took decades to build.
Factories.
Ports.
Roads.
Warehouses.
Supplier networks.
Employee expertise.
Relationships.
Standards.
Technology.
Companies cannot recreate all of that domestically overnight.
A smartphone may contain components from dozens of countries.
A car can involve thousands of internationally sourced parts.
A pharmaceutical product may depend on active ingredients from one country and manufacturing equipment from another.
Decoupling is therefore expensive.
That naturally limits how quickly governments can force it.
Consumers are another powerful force for globalization
Consumers like:
low prices;
variety;
foreign brands;
travel;
international entertainment;
and fast access to technology.
Globalization survived partly because people actually use its products.
A politician can criticize imports.
A consumer may still want:
a Korean television;
a Japanese car;
an American smartphone;
Italian luxury goods;
Chinese electronics;
or Colombian coffee.
That demand creates powerful economic incentives to keep borders commercially connected.
People flows have recovered too
Globalization is not only containers and investment.
It also involves people.
The DHL data show international people flows have fully recovered from the pandemic collapse and reached new highs in several categories.
Tourists are traveling.
Students are studying abroad.
Migrants are crossing borders.
Businesses are moving employees internationally.
That matters because human mobility reinforces:
trade;
investment;
knowledge transfer;
and cultural integration.
Information is the one area showing clearer fragmentation
There is one major exception.
Cross-border information flows are becoming more constrained.
Countries are imposing:
data-localization laws;
internet restrictions;
technology controls;
content regulation;
and cybersecurity requirements.
DHL says information globalization had produced some of the largest gains over the previous two decades but has slowed and become more volatile since 2021.
This may be where geopolitical fragmentation becomes most visible.
The internet is increasingly splitting into regulatory zones.
China has one model.
Europe another.
The United States another.
U.S.-China decoupling is real
It would therefore be wrong to say nothing has changed.
Washington and Beijing are genuinely becoming less integrated directly.
DHL calculates that since 2016, the share of U.S. international flows involving China across trade, capital, information and people has fallen about 42%.
China’s corresponding dependence on the U.S. is down about 37%.
That is significant.
The world’s two largest economies are deliberately reducing exposure to one another.
But the rest of the world has not followed them into the same degree of separation.
The surprising part is that allies are not fully splitting into blocs
One popular theory predicted a world divided into:
a U.S.-led economic bloc;
and a China-led bloc.
The data so far are less dramatic.
DHL found little comparable pattern of broad decoupling among most U.S. and Chinese allies, excluding unusual cases such as Russia.
Countries often want relationships with both sides.
Vietnam trades heavily with China and the U.S.
Saudi Arabia works with Washington and Beijing.
Southeast Asian states attract Chinese investment while exporting heavily to America.
Europe still trades significantly with China despite security concerns.
The world is becoming multipolar rather than cleanly divided.
ASEAN may be one of the biggest winners
Southeast Asia sits directly in the middle of that multipolar system.
The IMF study on ASEAN shows how tariff fragmentation can redirect trade and investment toward countries positioned between major powers.
That is why countries such as Vietnam, Malaysia, Thailand, Indonesia and Singapore have become strategically important.
They offer:
manufacturing;
ports;
growing consumer markets;
and political relationships with multiple major powers.
Global fragmentation can therefore produce regional winners even if the world economy as a whole becomes less efficient.
India is benefiting too
India has pursued its own strategy of strategic openness.
It competes with China.
It cooperates with the United States.
It buys energy from Russia.
It signs trade agreements with Europe and Britain.
That flexible positioning has made it attractive to companies seeking alternatives to China.
Apple and other electronics manufacturers have expanded production in India.
Again, the goal is not necessarily replacing one global supply chain with a domestic one.
It is building another international node.
Trade agreements are multiplying, not disappearing
Protectionism gets more headlines.
But countries continue negotiating trade agreements.
The UK-India deal is one example.
Thailand is close to completing a new trade agreement with the United States despite being hit with tariffs earlier this year.
Governments may raise barriers against some countries while simultaneously lowering barriers with others.
That is another reason “globalization versus protectionism” can be too simplistic.
The real world contains both at once.
The system is becoming less efficient — but more resilient
This is perhaps the clearest description of modern globalization.
The old model optimized cost.
The new model optimizes resilience.
Producing everything in the cheapest possible location may reduce costs.
But it creates vulnerability.
Producing in three countries may cost more.
But it protects against tariffs, wars or port closures.
Companies increasingly accept that premium.
That makes globalization more expensive.
It does not necessarily make it smaller.
There are real economic costs to fragmentation
The IMF continues warning that severe fragmentation would reduce long-term economic growth.
Its research suggests trade reallocation can benefit individual countries in the short term while producing larger aggregate losses over time.
Duplicated factories cost money.
Tariffs distort prices.
Export controls reduce efficiency.
Companies spend more managing complex supply chains.
Consumers ultimately absorb some of that cost.
Globalization’s survival does not mean fragmentation is harmless.
The world may therefore be entering “messy globalization”
That may be a better term than either globalization or deglobalization.
Companies still cross borders.
Capital still crosses borders.
People still cross borders.
Technology still crosses borders.
But now every flow is affected by politics.
Trade agreements.
National security.
Sanctions.
Subsidies.
Export controls.
Tariffs.
Domestic-content rules.
Companies must navigate governments almost as carefully as markets.
National industrial policy is back
The United States subsidizes semiconductors.
Europe supports batteries and clean technology.
China funds strategic manufacturing.
India offers production incentives.
Japan subsidizes chip factories.
Governments increasingly want domestic capacity in industries they consider essential.
That looks anti-globalization.
But the factories themselves often depend on international investors, machinery and supply chains.
A semiconductor fab built inside one country may still rely on equipment and materials from five others.
Industrial policy therefore reshapes globalization rather than eliminating it.
The semiconductor sector proves the point
Few industries are more politically sensitive.
Yet semiconductors remain extraordinarily international.
Advanced chips may involve:
American architecture;
Dutch lithography equipment;
Japanese materials;
Taiwanese manufacturing;
South Korean memory;
Chinese packaging;
and global customers.
Governments are spending billions trying to localize parts of that chain.
Nobody has yet recreated the whole ecosystem inside one country.
The economics are simply too complex.
Globalization survived because specialization remains powerful
The underlying economic logic has not disappeared.
Countries are good at different things.
Taiwan became exceptional at semiconductor manufacturing.
Germany specialized in industrial machinery.
China developed extraordinary manufacturing scale.
India built technology services.
The Gulf produces energy.
Singapore became a logistics and financial hub.
Trade allows each to sell those strengths globally.
Politics can weaken that system.
But completely eliminating specialization would impose enormous costs.
Businesses increasingly ignore political slogans and follow economics
Politicians may say:
“Bring everything home.”
Corporate executives ask:
At what cost?
Is labor available?
Is electricity reliable?
Can we obtain permits?
Are suppliers nearby?
Where are the customers?
Can we export economically?
Those questions often lead back to international production.
That is why political ambition repeatedly collides with commercial reality.
Trump may therefore succeed at changing globalization without ending it
This distinction matters.
His tariffs can:
reduce direct Chinese exports to the U.S.;
increase domestic investment;
push companies toward Mexico or Southeast Asia;
change supply-chain strategy;
and increase strategic production inside America.
Those are significant effects.
But they do not necessarily reduce total global trade.
They can instead redirect it.
The same applies to Brexit.
It changed Britain’s commercial geography and added friction.
It did not turn Britain into an isolated economy.
The biggest globalization story may now be Asia
The WTO says Asia will provide the strongest contribution to global trade growth in 2026.
That reflects:
AI manufacturing;
electronics;
industrial supply chains;
and strong intra-Asian commerce.
This is important because globalization is increasingly less centered on Western economies.
Trade between emerging economies is becoming more significant.
China trades heavily with ASEAN.
India trades increasingly with the Gulf.
Middle Eastern countries invest heavily in Asia.
South-South investment is expanding.
Globalization may become more global precisely because it becomes less U.S.-centric.
The U.S. matters less to world trade than many assume
DHL highlights a simple but important statistic.
In 2025, only about 13% of world imports were destined for the U.S., while roughly 9% of global exports originated there.
The United States is enormously important.
But most global trade does not involve America directly.
That means even sweeping U.S. tariffs cannot single-handedly stop globalization.
Other countries can trade with one another.
And increasingly, they do.
The same is true of China
China is the world’s manufacturing giant.
But it is not the entire global economy.
If China-U.S. trade weakens, companies can shift production.
China can sell more to:
Southeast Asia;
Europe;
Africa;
Latin America;
and the Middle East.
That creates new networks.
Again, fragmentation in one relationship can generate integration elsewhere.
Globalization is becoming harder to see
That may be why the perception gap is so large.
Old globalization was visible.
Container ships.
Factories moving to China.
Trade agreements.
Offshore call centers.
New globalization can be less obvious.
Cloud services.
Financial flows.
Digital subscriptions.
Remote workers.
Data centers.
AI chips.
Cross-border investment funds.
Software.
Online education.
The physical factory is no longer the only symbol.
The numbers therefore matter more than the rhetoric
Political language is moving toward nationalism.
Economic behavior remains highly international.
That contradiction may define the rest of the decade.
Governments want:
resilience;
security;
domestic jobs;
and strategic independence.
Consumers and companies still want:
low prices;
large markets;
specialization;
and international capital.
Neither side is likely to completely defeat the other.
The result is a more complicated global economy.
Globalization survived Brexit
Britain separated politically from the EU.
Its businesses did not stop selling internationally.
Instead, trade became more bureaucratic in some sectors and expanded elsewhere.
Globalization survived Trump’s first tariff war
U.S.-China direct trade weakened.
Vietnam, Mexico and others gained.
Globalization survived Covid
Supply chains broke.
Then companies rebuilt them with greater redundancy.
Globalization survived Ukraine
Energy flows changed direction.
Europe replaced Russian pipeline gas with LNG and other suppliers.
Trade rerouted again.
And globalization is now surviving another tariff wave
The latest WTO numbers may be the strongest evidence yet.
Goods trade is not merely holding steady.
It is growing much faster than economists expected.
That does not prove globalization is invincible.
A much deeper military or political rupture could still cause severe fragmentation.
But it demonstrates how adaptable the system has become.
The next phase may be globalization without trust
That is perhaps the real shift.
Countries trade with one another.
But they trust one another less.
Governments want backup suppliers.
Companies hold more inventory.
Technology is restricted.
Investment is screened.
National-security reviews are expanding.
Economic integration continues—
but political integration does not.
That can produce a more volatile world.
It may also be a more expensive world
Redundant supply chains cost more.
Tariffs cost more.
Extra inventories cost more.
Duplicated factories cost more.
Security screening costs more.
Companies eventually pass at least part of those costs to consumers.
So globalization may survive while losing some of the efficiency that made it so powerful.
That is not deglobalization.
It is higher-cost globalization.
The biggest mistake may be confusing decoupling with deglobalization
The U.S. and China are decoupling in important areas.
That is real.
The UK and EU relationship is more distant than before Brexit.
That is real.
Strategic sectors are being nationalized or friendshored.
That is real.
But those developments are happening inside a world where overall international flows remain enormous.
The pieces are being rearranged.
The puzzle has not disappeared.
The data are now challenging a decade of predictions
For years, economists, politicians and corporate executives warned that globalization had peaked.
First because of Trump.
Then Brexit.
Then Covid.
Then Ukraine.
Then U.S.-China tensions.
Then renewed tariff wars.
Yet in 2026:
global connectedness remains near record levels;
merchandise trade is forecast to grow 3.9%;
global FDI has returned to growth;
international travel has rebounded;
and AI has created an entirely new wave of hardware trade.
That does not mean nothing changed.
It means globalization proved much harder to kill than almost anyone expected.
Trump, Brexit and tariffs may have created globalization 2.0
The first era was built around maximum efficiency.
The second is being built around resilience.
Supply chains are longer in some places.
More diversified.
More politically managed.
More expensive.
But still international.
Companies are not asking whether to participate in the global economy.
They are asking how to participate without becoming dangerously dependent on one country, one factory or one shipping route.
That is a fundamentally different question.
And it explains why today’s world can simultaneously feel more protectionist—
while remaining deeply globalized.
Donald Trump changed trade policy. Brexit changed Britain’s relationship with Europe. Washington and Beijing are genuinely pulling apart.
But none of them has stopped goods, capital, technology and people from crossing borders.
Instead, globalization has done what it has repeatedly done during wars, pandemics and political upheaval:
it found another route.
