U.S. and China: How Growth Becomes Economic Power

The central argument

Two economies, different sources of strength

The United States and China have developed different sources of economic strength. China has expanded the volume of goods and services it produces, while the United States maintains a larger economy at market exchange rates and substantially deeper public equity markets. The strategic question is how each country converts these strengths into sustained prosperity, productive investment and influence abroad.

This distinction becomes clear over the past decade. Between 2015 and 2025, China's real output increased approximately 72%, compared with 27% in the United States. Yet China's GDP in dollars moved from 61.7% to only 63.4% of the U.S. total. The difference can be reconciled through domestic prices and currency translation. It also exposes the limits of evaluating either country through a single growth rate.

For business leaders, these measures have practical implications. Local purchasing power informs demand. Currency movements affect translated revenue and costs. Financial depth influences access to capital. Production networks shape the cost and resilience of supply. A useful comparison connects these measures to the decisions they inform.

At a glance | Eight comparable measures

The U.S. leads in income and finance; China leads in volume and production

Two economies, different sources of strength

The U.S. remains larger in current-dollar output, output per person and listed equity value. China's economy is larger at purchasing-power parity, real output per person is growing faster, and it has a much larger share of electrical-goods exports. These distinctions matter for a company choosing a demand market, supply base, reporting currency or source of capital.

Why these leads cannot be added together

PPP adjusts domestic prices, while dollar GDP translates output at market exchange rates. A stock-market value reflects sector mix, earnings expectations and valuations. Official reserves and military budgets measure separate capacities. The chart is a menu of relevant comparisons, not a ranking of national power.

At a glance | Where the models are exposed

The U.S. faces financing pressure; China faces a demand and debt test

U.S. exposure: finance the gap

The IMF estimated general-government debt at 123.9% of GDP in 2025. WDI puts the current-account balance at -3.6% of GDP. Dollar funding and deep capital markets provide flexibility, yet the fiscal path and external financing requirement are material to the durability of that advantage. The current-account figure is an accounting balance; it does not establish a near-term crisis.

China exposure: convert production into household demand

Household consumption was 40.0% of GDP in the latest common WDI observation (2024). The IMF said private demand remained weak and that property adjustment and local-government financing posed downside risks. Its wider augmented debt estimate reached 126.6% of GDP in 2025. That perimeter includes local-government financing vehicle activity and cannot be placed on the same scale as U.S. general-government debt without further reconciliation.

The decision test

In the U.S., test capital costs and demand under tighter financing. In China, test domestic demand and returns if property and local debt remain a constraint. These are conditional exposures, not forecasts.

2026 update | Verified 26 September 2026

Current releases add a different time horizon

Current momentum

In the second quarter of 2026, real GDP rose 0.4% from the preceding quarter in the U.S. and 0.9% in China, using each agency's seasonally adjusted measure. The familiar U.S. headline of 1.5% is an annualized rate. China's 4.3% headline is year-on-year. Placing those two headlines beside each other would create a misleading comparison; the chart uses the same time basis.

Current financial and trade observations

The latest WFE monthly table reviewed reports July 2026 domestic market capitalization. It shows approximately $42.5 trillion on Nasdaq U.S., $33.6 trillion on the NYSE, $10.0 trillion in Shanghai and $6.3 trillion in Shenzhen. These exchange observations supplement the harmonized annual country comparison. Census reports a U.S. goods deficit with China of $91.2 billion for January-July 2026; that partial year should be compared with the corresponding seven months, not all of 2025. [4,11]

Keep the observation date visible

The annual exhibits use the latest common observations in the retrieved WDI series, including 2025 GDP and 2024 expenditure shares. Current quarterly and monthly releases appear here without being spliced into annual histories. The World Bank's July 2026 China update also identifies continuing property adjustment and cautious consumers; its 4.4% full-year growth figure is a projection, not an observed 2026 result. [12]

01 | Reconciling the growth gap

Faster real growth met two offsets

The arithmetic

Starting from 61.7% in 2015, China's relative real-output growth raises the ratio to 83.5%. Applying relative changes in GDP prices reduces it to 73.2%. Translating at the changed currency conversion rate brings it to the observed 63.4% in 2025. The net increase is 1.7 percentage points. Each contribution is an accounting effect in the stated order; changing that order reallocates the interaction effects.

The mechanism

Over the decade, the implicit GDP price level increased about 16.0% in China and 32.5% in the U.S. The national-accounts currency conversion rate increased about 15.5% in yuan per dollar, reducing China's translated dollar output. Consumer inflation can help describe household conditions, but the GDP deflator is the relevant measure for reconciling real and nominal GDP.

The implication

A business selling locally and a business reporting in dollars may therefore experience different results in the same economy. This calculation does not establish deliberate currency manipulation or prove that one growth model is superior. It shows why a real-growth advantage does not automatically translate into an equal gain in dollar economic weight.

02 | Household prosperity

The income gap is large. The growth gap runs the other way.

Levels and rates answer different questions

At market exchange rates, U.S. output per person was approximately $90,000 in 2025, compared with $13,900 in China, a gap of 6.5 times. Adjusting for local purchasing power reduces that gap to about 3.1 times: $90,000 versus $29,300. China's faster real growth per person, 5.1% versus 1.6%, is consistent with convergence from a substantially lower starting point.

What an average conceals

GDP per person measures average production. It does not establish the purchasing power of a typical household, the distribution of income, or the share available for consumption. A consumer-business assessment should also examine household disposable income, consumption per person, regional differences, and the distribution of spending. A country-level average is an initial screen for market opportunity.

What sustains convergence

Long-term improvement depends on labor participation, hours worked, skills, capital efficiency and productivity. Per-person output growth is not identical to productivity growth per worker or per hour. These distinctions matter as populations age and technology changes the relationship between employment and production. A credible long-term view should test the contribution from each source rather than extrapolating a recent growth rate.

03 | The growth model

China allocates much more of its economy to investment

Different starting points for the next phase

Household consumption represented 68% of U.S. GDP and 40% of China's in 2024. Gross capital formation was approximately 22% and 40%, respectively. These allocations illustrate why a similar increase in total output can have different consequences for domestic consumers, producers and exporters.

The return on investment matters

A high investment share can build infrastructure and expand capacity. Its value ultimately depends on the additional income and cash flow generated by those assets. The World Bank's July 2026 China update describes continuing property adjustment and cautious consumption. The IMF's latest completed China Article IV assessment also connects domestic demand, property adjustment and local-government debt. Together they identify the channels that a financial-risk comparison should test. [2,12]

A useful extension of the comparison

Real estate belongs in this analysis because it connects household wealth, construction, collateral and local finances. The next exhibit compares house-price changes, residential investment and household debt, with China's local-government financing exposure shown separately. Manufacturing scale is also evaluated alongside margins, utilization and returns on capital. Production volume alone cannot establish the quality of the investment that created it.

04 | Financial strength

Large financial assets create both capacity and exposure

The distinction between size and value creation

The U.S. listed equity market was approximately $68.9 trillion in 2025, compared with $15.5 trillion in China. This difference is relevant to capital raising and the value investors assign to listed earnings. It also incorporates industry composition, valuations, overseas earnings and the scope of companies included. It does not establish that U.S. assets offer higher prospective returns.

A balance-sheet perspective

Changes in wealth can arise from new investment or from changing prices of existing assets. A useful comparison separates those sources, then examines the obligations financed against them. Equity values should be assessed against earnings and cash flow; property values against rents, household income and financing costs. Government debt requires its own perimeter and debt-service analysis.

The financial tests

For equities, compare total returns in a common currency, dividend reinvestment, valuation multiples and sector concentration. For sovereign debt, use the same IMF general-government perimeter and vintage for both countries, then disclose China's augmented liabilities as an additional view. For real estate, examine prices, leverage and debt service.

05 | Trade and production networks

A smaller direct trade gap does not measure the full exposure

What changed

The U.S. goods deficit with China declined from $297.0 billion in 2024 to $202.7 billion in 2025. This is a clear change in direct bilateral trade. It does not, by itself, establish how much Chinese production remains embedded in imports from other countries, or how Chinese exporters have redirected their sales.

What the research adds

The latest September 2026 discussion-paper abstract by Antràs, Kulesza and Presbitero distinguishes export competition, import competition, consumer gains and sourcing gains. It cautions that simple trade-basket similarity can mischaracterize exposure because it omits several economically relevant factors. Only the public abstract of the September version was reviewed here. [5]

The business question

A manufacturer sourcing from a third country should examine the origin of critical inputs, ownership of suppliers, substitution lead times and exposure to tariffs or export restrictions. A competing producer should separately examine Chinese supply entering its own sales markets. Lower input prices can help a downstream business while putting pressure on an upstream competitor. National trade totals conceal that difference.

05A | Industry export strength

Export strength is concentrated in different industries

A more useful comparison than manufacturing totals

China accounted for 27.9% of the panel's global electrical-equipment exports and 22.1% of machinery and computer exports in 2024. The corresponding U.S. shares were 6.8% and 9.7%. The U.S. held larger shares in mineral fuels and oils, optical and medical instruments, and precious metals and jewelry. These are export positions, not measures of total domestic output or company profitability.

What the distinction changes

A procurement team needs to identify the products for which alternative supply is limited. A producer needs to identify where Chinese suppliers are competing in its own sales markets. Broad national rankings do not answer either question; industry and product comparisons begin to do so.

Scope matters

The panel excludes services, unspecified commodities and discrepancy categories. It therefore does not describe the full economy, and confidential or unallocated trade can obscure industries such as aerospace. China is reported separately from Hong Kong. Export location does not establish ownership of intellectual property or the domestic value added in a product.

05B | How competition changed

China gained in batteries and cars; computers followed a different path

The evidence does not support one national trajectory

China's export share in storage batteries rose from 28.0% to 46.8%, while its passenger-vehicle share rose from 0.7% to 9.6%. In computers and units, China's share fell from 44.9% to 30.9%. U.S. medical-instrument share declined from 24.3% to 21.5%, but remained above China's 6.7% in 2024.

Translate the comparison into a decision

For battery sourcing, examine supplier qualification, alternative capacity and landed cost under different tariff assumptions. In passenger vehicles, compare the product mix and destination markets before inferring pricing pressure. In medical instruments, test whether certification, service and distribution support returns despite changes in export share. These are decision implications, not causal findings from the trade data.

Avoid overstating what the categories show

Storage batteries include technologies beyond lithium-ion; passenger vehicles include more than electric vehicles. Computer exports do not capture all software, design or cloud value. Nominal export-share changes can reflect prices, product mix and production location as well as physical output.

05C | Competitive overlap

Export baskets became modestly more similar

Export composition moved closer; specialization is a separate test

The export-basket similarity index increased from 44.2 in 2015 to 47.8 in 2024, a 3.6-point rise. It was broadly flat to lower through 2022, before rising in 2023 and 2024. The result supports modest convergence in the composition of goods exports; it does not establish convergence in technology, wages or profitability.

Specialization is a relative measure

A product has revealed comparative advantage, or RCA, of at least one when it carries as much or more weight in a country's export basket than in the world basket. On this consistent basis, the number of products where both countries met that threshold rose from 121 to 127. A country can export a large dollar amount without meeting the threshold, and counts give small products the same weight as large ones.

The business implication and the remaining test

National complementarity and product-level rivalry can coexist. A strategy based only on aggregate trade balances can miss that distinction. The next decision-specific test is whether overlap occurs in the same destinations, product qualities and customer segments. This analysis does not measure supply-chain dependence, domestic value added or the causal effect of tariffs.

06 | Strategic capacity

Spending, currency use and reputation require separate tests

Defense and international finance

The U.S. spent about 2.8 times as much as China on its military in nominal dollars in 2025. Procurement costs, force readiness, geographic commitments and allied capacity also determine what those resources can accomplish. Global reserve-currency use is another form of influence: the dollar represented 57.1% of official foreign-exchange reserves in 2026 Q1, compared with 2.0% for the renminbi. A country's own reserve holdings and international use of its currency measure different things.

Technology requires an output test

Stanford's 2026 AI Index reports $285.9 billion of U.S. private AI investment in 2025 versus $12.4 billion in China, while also describing a much narrower model-performance gap. The investment measure excludes parts of China's state financing. The combination supports a focused question about how efficiently capital becomes capability; it does not support treating dollars invested as a direct measure of technological superiority. [9]

Soft power requires judgment

Brand Finance's 2026 survey puts the U.S. at 74.9 and China at 73.5. These are weighted measures of perceptions. They should be considered alongside observable indicators such as international students, research collaboration, tourism and cultural exports. A small score difference is insufficient to establish a precise strategic advantage.

06A | Defense resources and economic burden

China narrowed the spending gap without a rising GDP burden

Resources and burden moved differently

From 2015 to 2025, China's military spending rose approximately 71% in real terms, versus 11% for the U.S. The U.S./China spending ratio narrowed from 4.3x to 2.8x on this constant-dollar basis. China's GDP burden was broadly unchanged at about 1.7%; the U.S. share fell from 3.5% to 3.1%.

Keep the dollar bases separate

The 2025 endpoints here are $929 billion for the U.S. and $335 billion for China in constant 2024 dollars. The preceding snapshot reports approximately $954 billion and $336 billion in current 2025 dollars. These are different measures, not conflicting estimates.

What a spending comparison cannot establish

Military expenditure measures inputs. Procurement costs, readiness, allies and geography affect the capabilities those inputs deliver. Economy-wide PPP is not a defense procurement price index. Evaluate program economics and supplier capacity before treating national spending as an industry opportunity.

07 | What would change the outcome?

The next decade depends on more than real growth

A conditional outlook

If China's real-output growth factor exceeded the U.S. factor by 2% annually for ten years, with neutral relative prices and FX, its dollar GDP would reach approximately 77% of the U.S. level in 2035. A 2% annual headwind from the combined price and currency factor would leave the ratio near 63%. The same-sized tailwind would raise it to approximately 94%. These cases illustrate sensitivity; they do not assign probabilities.

What management should do

For market expansion, underwrite household demand in local purchasing-power terms and test translation into the reporting currency. For sourcing, map critical upstream inputs and the cost of substitution. For capital allocation, distinguish operating performance from valuation expansion and test refinancing exposure. For technology, evaluate adoption, operating results and supply constraints alongside research and investment spending.

08 | PUBLIC DEBT, MATCHED PERIMETER

Government debt on one IMF definition

Both ratios rise through 2031; the level gap narrows.

One definition, one vintage

Exhibit 9 uses the IMF's April 2026 Fiscal Monitor for both countries: general-government gross debt as a share of GDP. In 2025 the ratio was 123.9% in the U.S. and 99.2% in China. China's ratio has risen every year since 2019, by about 39 percentage points in total. The U.S. ratio is above its 2019 level but below its 2020 peak.

The projection is a baseline, not an outcome

The IMF's baseline places the U.S. at 142.1% and China at 126.8% in 2031. Both paths assume current policies. The U.S. staff report says stabilizing debt would require a general-government primary surplus of about 1% of GDP, roughly 4 percentage points of adjustment from the baseline.

Why one definition matters

U.S. Treasury dollar totals and an IMF ratio for China are each correct, but together they cannot show relative burden. This exhibit gives a like-for-like baseline. The wider China measures follow in the next section.

08A | CHINA'S DEBT PERIMETERS

China's debt depends on where the state ends

The same 2025 economy shows debt from 68% to 127% of GDP.

Three published measures

The IMF publishes three views of China's public debt for 2025. On-budget general-government debt in the Article IV report is 68.4% of GDP. The WEO and Fiscal Monitor measure is 99.2%. The augmented measure, which adds local-government financing vehicles (LGFVs) and government-guided funds, is 126.6%.

Why the U.S. bar sits apart

The U.S. figure of 123.9% shares a definition with China's middle bar only. It is shown below the dividing line for reference. The similarity of 123.9% and 126.6% is a coincidence of levels, not a like-for-like result.

What the widest measure captures

LGFVs borrowed to build infrastructure and prepare land, often repaid from land sales. The IMF recommends restructuring unviable LGFVs through insolvency frameworks, alongside a plan to contain spillovers to banks. The augmented perimeter matters because it is where the property cycle meets public finance.

08B | DEFICITS AND INTEREST

Deficits persist; interest is the U.S. pressure point

At these debt levels, rates matter as much as growth.

Deficits on the matched basis

The Fiscal Monitor puts the 2025 overall balance at -6.8% of GDP in the U.S. and -7.9% in China, with both between -7.4% and -8.4% in every projected year through 2031. China's augmented deficit was -14.3% in 2025 and narrows only to -12.5% by 2030 in the IMF baseline.

The interest bill

CBO projects federal net interest above $1.0 trillion in fiscal 2026, 3.3% of GDP, rising to 4.6% by 2036, after $970 billion in 2025. A comparable published interest series for China's general government was not found in the releases reviewed; it is not estimated here.

A simple rate test

If the full debt stock repriced 1 percentage point higher, annual interest would rise by about 1.24% of GDP in the U.S., 0.99% on China's WEO measure and 1.27% on its augmented measure. This is an upper bound, because debt reprices gradually as it matures.

09 | PRIVATE BALANCE SHEETS

U.S. households deleveraged; Chinese firms did not

China's private debt service is rising while America's falls.

Households

BIS data show U.S. household credit falling from 77% to 67% of GDP between late 2015 and early 2026. Chinese household credit rose from 38% to 58%, peaking near 61% in 2020 and easing slightly since.

Firms

Chinese non-financial corporate credit was 147% of GDP in 2026-Q1, about twice the U.S. level of 72%. It fell to 123% in 2021 and has risen by 24 points since. LGFV borrowing is classified as corporate rather than government debt in the BIS data.

Debt service

The BIS private non-financial debt-service ratio, interest and principal as a share of income, rose to 19.1% in China and fell to 13.8% in the U.S. China's burden has risen as debt grew faster than income. The BIS publishes a China household ratio only within the private-sector total.

10 | PROPERTY

China's house prices have fallen for five years

Real Chinese home prices are 25% below their 2021 peak.

Two cycles moving in opposite directions

In inflation-adjusted terms, Chinese residential prices peaked in 2021-Q3 and have fallen 25.3% since, to 10% below their 2015 level. U.S. real prices rose 35% over the same period, peaked in 2024-Q4, and are down 4.1% since.

Why housing reaches the rest of the economy

Housing is the main store of household wealth in China, collateral for lending, and, through land sales, a revenue source for local governments. A sustained price decline therefore affects consumption, bank asset quality and the LGFVs in the debt sections above at the same time.

What the IMF proposes

The IMF recommends accelerating the exit of unviable developers and providing targeted central-government support of about 5% of GDP to protect homebuyers. The recommendation is policy advice, not a measure of the cost already incurred.

10A | CONSTRUCTION

China's building sector has roughly halved since 2021

New starts are down 70%; the decline continues in 2026.

The size of the adjustment

China's real estate development investment fell from 14.8 trillion yuan in 2021 to 8.3 trillion in 2025, from 12.6% to 5.9% of GDP. Sales of new buildings fell 54% by value, and new starts fell 71% by floor space. Investment was down a further 19.9% year on year in January-August 2026.

A gross measure, handled accordingly

The NBS series includes land purchase costs and is not a national-accounts component, so it overstates construction's direct share of GDP. The U.S. comparable, residential fixed investment, was 3.6% of GDP in 2026-Q2 on a national-accounts basis and is not charted on the same axis.

The link to returns

An economy that invested heavily in housing now carries the supporting debt while the flow of new building declines. The next sections ask how much growth the remaining investment is producing.

11 | RETURNS ON INVESTMENT

China's investment yield has fallen to U.S. levels

China invests twice the U.S. share for the same growth yield.

A simple efficiency screen

Exhibit 15 divides the average gross investment share of GDP by average real growth over five-year periods, a measure known as the incremental capital-output ratio. A higher value means more investment per point of growth. China's ratio rose from 3.6 in 2005-09 to 8.6 in 2020-24, close to the U.S. ratio of 8.8.

What changed

China still invested 41.8% of GDP in 2020-24, about twice the U.S. share of 21.6%. Real growth averaged 4.9%, down from 11.5% in 2005-09. The same investment effort now produces roughly half the growth.

Limits of the screen

The ratio ignores depreciation, which is larger in a mature capital stock, and treats all growth as a return to investment. The 2005-09 U.S. value is inflated by the recession. It is an aggregate screen, not a firm-level return on capital.

11A | PRODUCTIVITY

China's output per worker is converging more slowly

China's worker output reached 31% of the U.S.; growth slowed.

Levels

At 2021 purchasing-power parity, U.S. GDP per person employed was about $157,000 in 2025 and China's about $48,100. China's level rose from 9% of the U.S. level in 2005 to 20% in 2015 and 31% in 2025.

Rates

China's output per worker grew 9.2% a year from 2005 to 2015 and 5.8% a year from 2015 to 2025. U.S. growth rose from 1.1% to 1.5% a year over the same periods. Convergence continues, but at a slower pace, alongside the weaker investment yield in the section above.

Still to add

Output per hour, labor participation and the UN population scenarios would separate hours, employment and ageing from efficiency. They are not estimated here.

12 | COMPANY CASE: SUPPLY NETWORK

Apple moved U.S. supply, not its China base

Tariffs moved Apple's U.S. supply; China still builds for other markets.

Decision

After new U.S. tariffs in April 2025, Apple said India would be the country of origin for the majority of iPhones sold in the U.S. in the June quarter, with most U.S.-bound Macs, iPads, Watches and AirPods coming from Vietnam. China would remain the source for most products sold elsewhere.

Economics and constraint

Apple estimated tariff costs of $0.9 billion for the June 2025 quarter, reported about $0.8 billion incurred, and projected $1.1 billion for September. Fiscal 2025 gross margin was 46.9%, compared with 46.2% in 2024, so margins rose despite the cost. Its 10-K still describes a significant majority of manufacturing performed by partners located primarily in China, India, Japan, South Korea, Taiwan and Vietnam.

Outcome

Greater China net sales fell 3.8% to $64.4 billion while total net sales rose 6.4%, reducing China's share of revenue from 17.1% to 15.5%. The case illustrates the trade section's point: the bilateral deficit can narrow while Chinese capabilities remain central to a supply chain.

13 | COMPANY CASE: CONSUMER MARKET

Starbucks sold control to keep growing in China

Starbucks kept the brand and 40%, and gave up control.

Decision

In November 2025 Starbucks agreed to sell 60% of its China retail business to Boyu Capital at about a $4 billion enterprise value, keeping 40% and ownership of the brand, which it licenses to the venture. The deal closed on 2 April 2026, covering about 8,000 stores.

Economics and constraint

In fiscal 2025 China comparable sales fell 1%: transactions rose 4% while the average ticket fell 5%. In the fourth quarter, transactions rose 9% and ticket fell 7%. Customers were visiting, but at lower prices, which is the household-demand constraint in the macro sections seen at store level.

Outcome

Starbucks reported $2.544 billion of net proceeds and a $536.3 million gain, used part of the proceeds to retire about $1.3 billion of debt, and saw its international operating margin rise from 13.6% to 19.1% as China moved to a licensed model. It values its total China interest above $13 billion. The venture aims for up to 20,000 stores; that is an aspiration, not a result.

14 | IMPLICATIONS

What the stress test changes

Scale is not the constraint; financing and returns are.

For the central argument

The earlier sections showed that faster real growth did not translate into dollar weight. Sections 08 to 13 add the cost of that growth. China's investment now yields roughly the U.S. growth per unit while its public, corporate and property balance sheets carry the legacy. The U.S. has a cleaner private sector and a heavier federal interest bill.

For a company choosing a demand market

Test household demand in local prices and ticket trends, not just traffic or population. Starbucks grew transactions while revenue per visit fell.

For a company choosing a supply base

Map origin by destination market. Apple's response changed where U.S.-bound products came from, not whether China remained central to its network.

For an investor

Treat U.S. sovereign risk as a rate question and Chinese risk as a perimeter and property question. Keep the matched IMF view and the augmented view side by side rather than choosing one.

What would change the conclusion

A recovery in China's real house prices, a fall in its private debt-service ratio, or a return to lower investment per point of growth would weaken the case above. A sustained fall in U.S. interest costs relative to GDP would weaken the U.S. financing concern.

Conditional implications drawn from Exhibits 9-18; not forecasts.