Two leaders walked very slowly through a building in Beijing this week, shook hands at a slightly uncomfortable distance, and then announced that China would buy some soybeans, some Boeing jets, and some oil. Both sides declared this a success. A new oversight body - the Board of Trade - was announced to make sure China follows through on its commitments this time. This was necessary because China made identical purchase commitments in 2020 and then didn't follow through. The solution to not honoring a committee's oversight is, apparently, a new committee.
According to the White House, Xi Jinping agreed to purchase US soybeans, oil, liquified natural gas, and 200 Boeing 737 jets. Trump and Xi met for two hours and fifteen minutes. This was hailed as progress. To understand why it isn't - or more precisely, why it can't be - you need to understand that the problem these two men are trying to solve is not a political problem at all. It is an accounting problem. And the reason it keeps not getting solved is that almost nobody in a position of power appears to understand what is actually causing it.
The background
Trade, at its most basic level, is an exchange. You make something I need; I make something you need. We swap. Both sides are better off. This idea goes back to the economist David Ricardo, who in the early 1800s argued that countries should specialize in what they produce best and trade for the rest. The entire logic of international trade is built on this premise: you sell abroad to earn foreign currency, and you use that foreign currency to buy things from abroad that you cannot produce as well yourself.
For most of the 20th century, this is broadly how trade worked. Countries ran surpluses and deficits that fluctuated, but they tended to balance out over time because everyone needed things from everyone else.
Then something changed. A group of economies - first Japan in the 1970s and 1980s, then Germany within the eurozone, then China in a much larger way - began running persistent, structural trade surpluses. A trade surplus means a country exports more than it imports, meaning it earns more foreign currency than it spends. In principle, nothing wrong with that. In practice, when it becomes a permanent feature of policy, it creates a problem for everyone else.
These surpluses didn't appear because the countries in question were just very good at making things. They appeared because of specific domestic policy choices that economists call financial repression - a set of mechanisms by which a government systematically transfers wealth away from its own households and toward manufacturers and the state. Low interest rates on savings accounts mean ordinary people earn almost nothing on their deposits. An undervalued currency makes exports cheaper and imports more expensive. Weak labor protections mean workers don't earn enough to buy back what they produce. The result is a population that makes a lot but consumes very little - and the leftover output has to go somewhere. It goes abroad.
This is not a trade story. It is a domestic income story. The trade surplus is a symptom. The cause is what happens to ordinary people inside the economy.
What is actually happening
Six months ago, journalist Robin Harding traveled through China for the Financial Times and asked every economist, technologist, and business leader he met the same question: what does China want to buy from the rest of the world? The answer he kept getting was, essentially, nothing. As Harding reported, his interlocutors consistently held that there was nothing China believed it could not make better and cheaper at home, and nothing for which it wanted to rely on foreign suppliers for a day longer than necessary. A handful went further: when pressed on what China would like to import, their answer was that foreign companies should be allowed to build factories in China - which is not an answer to the question but does reveal quite a lot about the underlying logic.
This is the trade relationship that arrived in Beijing this week. China is selling to the world at scale and buying almost nothing in return. As economist Michael Pettis, a finance professor at Peking University and senior fellow at the Carnegie Endowment, has been arguing for years, this is not the result of trade policy. It is the automatic consequence of domestic savings and investment decisions. When a government systematically prevents households from consuming, excess production accumulates. It has nowhere to go but out. The trade surplus is not a strategy - it is an accounting identity.
China's current account surplus - the broadest measure of how much more it sells to the world than it buys - has been widening again. The EU runs a trade deficit with China of roughly 1 billion euros per day, driven primarily by electric vehicles and electronics. Two-way goods trade between the US and China amounted to roughly $415 billion in 2025, down sharply from its 2022 peak of $690 billion, following years of escalating tariffs. The US put tariffs of up to 145% on Chinese goods. China retaliated with 125% on American goods, then restricted exports of rare earth minerals - the elements used in everything from electric motors to missile guidance systems - as additional leverage. Both sides eventually agreed to a 90-day truce, currently in effect and expiring in November.
The summit this week was supposed to move beyond the truce toward something more durable. What it produced was a soybean commitment, a Boeing order, and a committee.
The money trail
To understand why the beans and the Boeings don't fix anything, you have to follow the money upstream. Pettis's argument - developed over decades of writing on China's economic model - is that China's trade surplus is driven by the same mechanism that produced Japan's surplus in the 1980s and Germany's surplus within the eurozone: suppressed household consumption forces excess savings, which gets channeled into industrial investment, which eventually produces more than domestic consumers can absorb, and the excess flows out as exports.
When investment exceeds what an economy actually needs - once you have built enough high-speed rail, enough apartment blocks, enough factories - continuing to invest doesn't create economic value. It destroys it. The money is spent, the GDP numbers look impressive, and the underlying projects generate less value than they cost. This is, Pettis argues, where China is now. President Xi has reportedly spent considerable time studying Japan's lost decades - the thirty-year period of stagnation that followed its 1980s investment boom - apparently as a cautionary tale. Whether the lesson has been applied is a separate question.
The only way out of this model, structurally, is to shift from investment-led growth to consumption-led growth. Stop subsidizing factories. Start raising household incomes so that Chinese consumers can actually buy what Chinese factories produce. But governments that have built their power on controlling where capital goes are generally reluctant to hand that control to households. So the surplus continues. And the surplus needs somewhere to land.
The US is where most of it lands. Not by choice, but because the US has the deepest, most liquid financial markets in the world. When surplus countries generate excess savings - money that their own populations cannot absorb - that money flows into US bonds, US equities, US financial assets. Roughly half of the world's excess savings end up in American markets. And here is where the accounting becomes inescapable: when foreign capital flows into the United States at this scale, the US must, by definition, run a corresponding trade deficit. This is not a policy choice. It is a balance of payments identity - the mathematical relationship between a country's capital account and its current account. Foreign money buys American financial assets; Americans buy foreign goods. One side of the ledger forces the other.
The US is, as Pettis puts it, the consumer of last resort for the global economy. Not because it wants to be. Because the accounting leaves no alternative.
The bill for this arrangement is becoming visible. The US government sold 30-year debt at a 5% yield for the first time since 2007 this week, as surging energy costs tied to the Iran conflict pushed inflation expectations higher. Federal interest payments have reached an annualized $1.23 trillion in the first quarter of 2026 - roughly 4.18% of GDP - having tripled from around $400 billion in the pre-pandemic period. The national debt has climbed past $39 trillion. These are not independent problems. They are the accumulated consequence of absorbing the world's excess savings for three decades.
Running a larger fiscal deficit to stimulate the economy, which is the standard political response, doesn't fix this. It makes it worse. More deficit means more Treasury bonds. More Treasury bonds attract more foreign savings. More foreign savings flow in, the dollar strengthens, American exports become less competitive, and the trade deficit widens further. The feedback loop runs in one direction.
What people are doing about it
The response at the summit level was to announce purchases and committees. At the legal level, Trump arrived in Beijing having had his two main tariff instruments struck down by US courts. The Supreme Court invalidated his IIPA tariffs earlier this year. A federal trade court then ruled that tariffs imposed under Section 122 of the Trade Act were also illegal, on the grounds that the US is not experiencing the kind of balance of payments crisis that law was designed to address. The administration is preparing a third set of tariffs under Section 301, but those investigations won't be complete until summer. The 90-day truce expires in November, roughly coinciding with midterm elections at which tariff-driven inflation will be a factor. Trump's approval ratings currently sit at 34%, the lowest of his second term. Republican members of Congress were reportedly relieved when the Supreme Court struck down the tariffs.
China, meanwhile, is playing for time. Its primary strategic objective is to develop domestic semiconductor manufacturing and reduce dependence on Western chip technology - a process that takes years, not months. China imports roughly 40% of its oil through the Strait of Hormuz, which has been effectively closed by the US-Iran conflict, and Xi used the summit to press Trump on this directly. Rare earth export restrictions, which proved devastatingly effective as leverage during the tariff escalation, remain a tool available to Beijing.
Europe, sitting between these two blocs, is responding by preparing legislation that would require Chinese companies seeking market access to hire local workers, buy local components, and transfer technology - exactly the terms China imposed on Western multinationals a generation ago. Chinese officials have noticed the similarity, and are not finding it flattering. The EU's trade commissioner has compared Europe's dependency on Chinese minerals to its former reliance on Russian energy.
Historically, these structural imbalances have resolved in one of two ways. The 1980s version ended through the Plaza Accord - a coordinated agreement among major economies to adjust their exchange rates. The 2008 version ended in a global financial crisis. The 1920s version ended in a depression and a war. As economist Martin Wolf has noted in the FT, these clashes recur roughly every two decades and the two available outcomes are international cooperation and economic catastrophe. The historical record suggests a modest preference for the second option.
The bottom line
The Beans and Boeing Summit produced exactly what its nickname implied: symbolic purchases and performative committees. The underlying problem - China's systematic suppression of household consumption, which mechanically generates a trade surplus that the US is structurally forced to absorb - was not addressed because it cannot be addressed at a trade summit. It can only be addressed by the domestic policy choices of the Chinese government, which would require transferring economic power from the state to households. That is a political cost Beijing has consistently refused to pay. Until it does, the surplus will continue to exist. The US will continue to absorb it, accumulate debt, and pay an escalating interest bill. The next summit will produce more committees. The accounting will not care.
Timeline
- Early 2000s - First China shock: China's exports surge as it builds manufacturing capacity, driving factory closures across the US Midwest and contributing to the hollowing out of American industrial employment
- 2008 - Global financial crisis; China responds with a massive stimulus package channeled into domestic investment, widening its trade surplus and inflating a domestic property bubble
- 2018 - Trump's first term trade war begins; average US tariff on Chinese goods begins rising from 3.1%
- 2020 - Phase 1 trade deal signed; China commits to purchase targets for US goods; China misses its purchase commitments; no Phase 2 follows
- Late 2024 - Trump and Xi meet in South Korea; agree to a trade war thaw; average US tariff on Chinese goods stands at 47.5%, up from 3.1% before Trump's first term; China's average tariff on US goods at 31.9%, up from 8.4% in 2018
- Early 2025 - Tariff escalation resumes; US imposes tariffs of up to 145% on Chinese goods; China retaliates with 125% on American goods; China restricts rare earth mineral exports
- 2025 - Two-way US-China goods trade totals approximately $415 billion, down from a 2022 peak of $690 billion
- Early 2026 - US Supreme Court strikes down Trump's IIPA tariffs; 90-day truce agreed upon
- May 2026 - Federal trade court rules Section 122 tariffs illegal; US sells 30-year bonds at 5% yield for first time since 2007; US federal interest payments reach annualized $1.23 trillion
- May 14-15, 2026 - Trump-Xi summit in Beijing; Xi commits to purchase US soybeans, oil, and 200 Boeing 737 jets; Board of Trade announced; 90-day truce expected to be formalized; analysts describe outcome as stabilization, not revitalization
Summary
Who: US President Donald Trump and Chinese President Xi Jinping, with delegations including Elon Musk, Tim Cook, Jensen Huang, and Boeing CEO Kelly Ortberg.
What: A two-day summit in Beijing that produced Chinese commitments to purchase US soybeans, oil, and 200 Boeing aircraft, along with the announcement of a new Board of Trade to oversee implementation - commitments structurally similar to those made in 2020 that went unmet.
When: May 14-15, 2026.
Where: Beijing, China.
Why: Both sides are managing a 90-day trade truce set to expire in November, with the US seeking rare earth supply chain security and agricultural export commitments, and China seeking tariff relief and time to develop its domestic semiconductor industry. The deeper cause - China's suppression of household consumption creating a structural trade surplus the US is mathematically forced to absorb - was not on the agenda and will not be resolved by purchase commitments or oversight committees.