CICC estimates Chinese cloud vendors' ROIC at about 11.9%, judging that AI's crowding out of domestic demand may be faster than in the US

Introduction

As an "enemy," AI's crowding-out effect manifests in the appropriation of resources and financial resources, policy space (in the US, crowding out occurs through interest rates, because it relies mainly on private financing; in China, crowding out occurs through the direction of fiscal and credit allocation), and employment. China's crowding out may be faster, such as this year's domestic demand.

So when does it change from "friend" to "enemy"? Our estimates show that the US as a whole can withstand 6% interest rates and a 50% price increase, with some room to spare. China crowds out domestic demand earlier, but non-quantitative and long-term objective functions also make its tolerance higher.

Since July, the global AI chain has continued to adjust. Beyond rising costs, the more important factor is market concern that the returns AI can generate cannot sustain large-scale investment. For example, the market appears to be worried about CoreWeave's credit risk, but behind that may be concern about a slowdown in Anthropic's ARR, so the market needs to calculate the ROIC (return) versus WACC (cost) equation, which is exactly the same "script" as the market's concern about private credit in the fourth quarter of last year.

Chart: CoreWeave's CDS spikes (default concerns) trigger a selloff

Source: Bloomberg, CICC Research Department

Chart: As of Q2, US cloud providers' 19.9% ROIC still covers 9.6% WACC

Source: FactSet, Bloomberg, CICC Research Department

In addition to having to "work hard" to prove that their returns are sufficient to cover ever-expanding capital expenditure, AI in both China and the US has also reached another critical juncture: their own capabilities (the money they earn themselves, operating cash flow) can no longer support their ambitions (the investment demand of capital expenditure), and they need to obtain support through external financing (free cash flow turning negative), rather than being able to invest only with the money they earn themselves as before.

Of course, borrowing is not the problem; the problem is borrowing without earning matching returns.

But regardless, as AI in both China and the US begins to draw resources from the whole of society, another increasingly important question is gradually surfacing: will AI gradually move from being a "friend" that drives other economic sectors to its opposite, becoming an "enemy" that crowds out traditional demand? In other words, can the upper half of the K-shaped divergence continue to offset the lower half, so that overall growth still trends upward?

English text is being prepared.

However, China and the US differ in their transmission and crowding-out paths, and their tolerance levels and thresholds also differ. Clarifying this "threshold" also clarifies the boundaries that the economy and policy can withstand, which is of great benefit to us in judging the policy paths of China and the US and choosing investment directions.

In the second quarter of this year, US AI reached a critical juncture: it can no longer use the money it earns itself (operating cash flow) to support ever-expanding capital expenditure, and therefore needs external financing (equity or debt) to fill the funding gap. The landmark event is that in the second quarter of this year, the free cash flow of major cloud providers (Hyperscalers) other than Microsoft and Meta all turned negative (operating cash flow minus capital expenditure).

Chart: US cloud providers' capital expenditure approaches operating cash flow

Source: FactSet, CICC Research Department

Of course, this is not to say that borrowing is bad. On the contrary, borrowing can leverage more funds and can also raise ROE (ROE = profit margin * turnover * leverage). Only borrowing without earning back the corresponding money is bad. For cloud providers, as of Q2, a 19.9% ROIC can fully cover a 9.6% WACC, not to mention corporate bond financing at less than 5-7%.

This is true of the US, and similarly of China. Moreover, because China's AI investment ROIC is lower (as of Q2, the weighted ROIC of China's leading cloud providers such as Tencent, Alibaba and Baidu was about 11.9%, lower than the 19.9% of US cloud providers), and free cash flow in Q2 has all turned negative, more external funding support is needed. In an environment where factor resource allocation is not fully determined by market pricing, the crowding out of traditional demand may be faster.

Chart: As of Q2, the weighted ROIC of China's leading cloud providers is about 11.9%, lower than that of US cloud providers

The economic impact of a technological revolution cannot be observed only in the technology production sector. The economic impact of a technological revolution may come from investment and production in the new technology itself, from productivity improvements and new supply creation brought about by the application of the new technology, and may also affect sectors that have not yet effectively adopted the new technology through factor prices.

It is not difficult to understand that the impact of new technology on different sectors differs markedly. Referring to the approach of Stiroh (2002), we divide the economy into four sectors to better reflect the impact of the technological revolution. Specifically:

1) AI production sector, mainly involving AI large model production links, such as computers, software, data centers/cloud services, and IT services;

3) AI application sector, including finance, professional services, R&D, pharmaceuticals, film and television media, etc.;

4) Other sectors, sectors not directly related to AI investment and consumption, such as food and beverages, real estate, transportation and warehousing, and entertainment, etc.

For the US, as of the first half of 2026, compared with 2022, the share of the AI production sector rose 2.4ppt to 10.3%; the AI application and infrastructure sectors were basically flat, together accounting for about 40%. Other sectors still exceed 50%, but are down 1.4ppt compared with the end of 2022. In short, AI-related and non-related sectors are roughly split fifty-fifty.

For China, AI-related and non-related sectors are also roughly split fifty-fifty, but the structure of AI-related sectors differs markedly from that of the US. As of the end of 2025, the AI production sector accounted for less than 5%, only half that of the US; the share of the AI application sector was about the same as in the US; however, the share of the AI infrastructure sector was close to 14%, about twice that of the US; other sectors accounted for about 50%, close to the US.

Source: BEA, CICC Research Department

Chart: China is also roughly split fifty-fifty, but the structure differs considerably from the US, with a higher infrastructure share

Source: Wind, CICC Research Department

► Directly driving growth through investment and exports, among other channels. This has been particularly evident over the past few quarters, in both China and the US. In the second quarter, US real GDP grew 2.2% quarter-on-quarter at an annualized rate; excluding the import factor, broad AI demand (AI production investment and infrastructure) contributed 28% to US growth, adding about 0.6ppt. Over the same period, China's real GDP grew 4.77% quarter-on-quarter at a seasonally adjusted annualized rate, with AI contributing 47%. For China, the boost was mainly reflected in exports: AI-related products contributed more than 60% to the year-on-year increase in net exports, and net exports contributed 21% to GDP growth, adding 0.6ppt to GDP growth. The same is true of earnings: AI-related sectors contributed 53% of US equity earnings growth, and 19% and 2.4% of A-share and Hong Kong equity earnings growth, respectively. In other words, strip out AI from the US, and half the earnings are gone; strip out AI from China, and half the growth is gone.

Chart: In the second quarter, US real GDP grew 2.2% quarter-on-quarter at an annualized rate, with AI-related demand contributing 28% of growth

Source: Haver, CICC Research Department

Chart: In the second quarter, China's real GDP grew 4.77% quarter-on-quarter at a seasonally adjusted annualized rate, with AI-related demand contributing 47% of growth

Source: Wind, CICC Research Department

► The wealth effect from market gains. Stock market gains under the AI rally generate a pronounced wealth effect, which in turn boosts households' consumption expectations. Chodorow-Reich et al. (2020), using empirical tests on the total wage bill of local non-tradable sectors (such as retail, catering and other service industries directly driven by local demand), show that for every $1 increase in stock wealth, households increase real consumption spending by about 3.2 cents per year. This is more evident in the US, thanks to the stronger performance of US equities and the larger stock market exposure of the household sector (401K, etc.). According to Federal Reserve statistics, in 2Q26, individual and other non-institutional investors held more than 42% of total US stock market capitalization, up 1.4ppt from the start of the year; equity-type assets accounted for 47% of households' balance sheets, up 2ppt from the start of the year; and so far this year, dividends and other such income have accounted for more than 9% of disposable income. By contrast, China's wealth effect still has room to improve.

Chart: After a stock market wealth shock, the total wage bill of non-tradable sectors rises

Source: Chodorow-Reich et al., (2020), CICC Research Department

Chart: In 2Q26, equity-type assets accounted for close to 47% of US household assets

Source: Bloomberg, CICC Research Department

Chart: So far this year, the median return of the 5,500 A-share companies is -17%

Source: Wind, CICC Research Department

Chart: As of 2025, the share of equity-type assets in household allocations remains very low

Source: CHFS, CICC Research Department

► Higher labor productivity and the positive externalities from knowledge spillovers. The first-principles channel through which AI drives the economy is the improvement in labor productivity from completing tasks at relatively low cost and high efficiency. Anthropic estimates that AI assistance can shorten the completion time of some tasks by an average of 80%, at a lower relative labor cost. As large-model application scenarios become increasingly rich, overall labor productivity in the US nonfarm sector has improved. Among industries with heavier AI investment, such as information technology and finance and insurance, labor productivity improvements have been particularly pronounced. Improved production efficiency, on the one hand, raises total factor productivity and, on the other hand, drives continued rapid growth in token usage, prompting cloud providers to keep scaling up computing investment and rapidly expanding demand for AI computing power, infrastructure and related segments.

Knowledge spillovers bring positive externalities: the spread of models, software, data and organizational experience to more firms can lower subsequent costs, gradually turning other sectors that previously had lower AI usage intensity into application sectors. Knowledge diffusion may also generate positive externalities that are not fully priced, making the social returns of AI higher than the private returns. In 2025, Brynjolfsson et al. (2026) estimate that US users derived annual consumer surplus of about $116.2 billion from generative AI chat tools. Over the same period, Gartner forecasts that global end buyers' spending on generative AI models was only $14.2 billion, about 8.2 times less than the former. Compared with other industries, every $1 of base fare paid by UberX users generates about $1.57 of consumer surplus, mobile communications generates 1 times consumer surplus, while generative AI reaches as high as 8.2 times. The results show that the social returns of AI are significantly higher than the private returns, so measuring its economic contribution solely by model market transaction value would systematically omit the largest piece of social welfare value.

However, once the returns AI generates for society (its own ROIC plus positive externalities) fall short of the resources AI crowds out from other sectors, AI may turn from a "friend" of traditional demand into an "enemy." The point at which AI begins to require external financing and to push up raw material prices is a typical example of factor resources crowding out traditional demand, reflected in three aspects:

► Crowding out physical resources. AI consumes key resources such as electricity and pushes up the marginal cost and capacity utilization of key factors in other sectors. AI capital expenditure is not merely a monetary outlay; it also requires physical inputs such as chips, servers, electricity, grid connections, and land. When the relevant supply is inelastic in the short term, new AI projects may crowd out physical resources from other sectors by pushing up electricity prices, equipment delivery prices, construction costs, and high-skilled wages. Since the end of 2022, U.S. private-sector spending on data center construction has grown nearly fivefold. Global data center electricity consumption as a share of electricity generation rose from 1.0% in 2020 to 1.5% in 2025; in the United States in particular, this share rose rapidly from 2.6% in 2020 to 4.9% in 2025. In addition, in recent years, with the development of AI, capacity utilization in U.S. industries such as oil and gas, electrical equipment, mining, machinery, and electronics has risen markedly. Since the start of this year, memory prices have risen 110%, pushing the year-on-year growth rate of the electronic components and accessories sub-index in U.S. PPI from 4% at the end of last year to as high as 29% in June, making it one of the main contributors to the rise in U.S. PPI year-on-year.

Chart: Global data center electricity consumption as a share of electricity generation keeps rising, especially in the United States

Source: IEA, Ember (2026), CICC Research Department

► Crowding out financial resources. After AI crosses the threshold at which it begins to require external financing, its crowding out of other sectors is reflected in both price and quantity dimensions: 1) On the price dimension, it pushes up overall financing costs. Strong AI demand, combined with external supply disturbances such as oil prices, jointly pushed up costs. For example, as U.S. Treasury yields rose, the 30-year mortgage rate has risen to a high of 7.5%, and the short-term loan rate for small businesses has also risen markedly to 8.2%. Looking at credit spreads, the credit spread on U.S. investment-grade corporate bonds has not risen significantly relative to the technology sector, but high-yield bond spreads have widened markedly.

Chart: The credit spread on overall U.S. investment-grade corporate bonds has not risen significantly relative to the technology sector

2) On the quantity dimension, massive AI bond issuance will also crowd out financing demand, including for U.S. Treasuries and other needs, but this is not yet significant. On the one hand, additional long-term investment-grade bond issuance by AI giants will crowd out long-only funds' allocation demand for U.S. Treasuries, and combined with the "small peak" in Treasury supply itself in the third quarter, this jointly led to the August "U.S. Treasury storm" ("Where Is the 'Problem' with U.S. Treasuries?"). On the other hand, AI bond issuance will also crowd out financing for non-AI companies; in particular, the market expects U.S. technology leaders' credit bond issuance to increase by another $200-300 billion in 2027, though the crowding-out effect is not yet significant. The technology sector (information technology + communications) has risen to 40-50% of U.S. nonfinancial corporate bond supply, but nonfinancial non-technology issuance has not declined significantly. Even after deducting maturities, net investment-grade supply from nonfinancial non-technology sectors has continued to grow.

Chart: The technology sector (information technology + communications) has risen to 40-50% of U.S. nonfinancial corporate bond supply

Source: Bloomberg, CICC Research Department

Chart: Net investment-grade corporate bond supply from U.S. nonfinancial non-technology sectors continues to grow

Source: Bloomberg, CICC Research Department

► Constraining policy space. Strong AI demand also constrains the Federal Reserve's policy space. The U.S. economy is also K-shaped, with traditional demand represented by real estate under pressure from high interest rates, and existing home sales weakening again after May. By rights, the Fed should cut rates to boost traditional demand, but now it has to raise rates to cope with stronger investment and higher inflation. According to Fed estimates, the U.S. natural rate of interest has risen from 0.99% at the end of 2022 to 1.44%. Year to date, most of the 79bp rise in the 10-year U.S. Treasury yield has been contributed by the real interest rate (71bp) rather than inflation expectations (9bp).

Chart: Year to date, the rise in the 10-year U.S. Treasury yield has mostly been contributed by the real interest rate (71bp) rather than inflation expectations (9bp)

Source: Bloomberg, CICC Research Department

Compared with the United States, China's crowding out is mainly reflected in fiscal and credit resources. On the one hand, fiscal resources are tilted more toward new quality productive forces such as the construction of the "six networks"; on the other hand, the allocation of credit resources also reflects a clear policy orientation, with China's new loans concentrated in technology and other fields, while the share of loans related to real estate and consumption has fallen sharply. Since the start of this year, consumption has remained weak and household balance sheets have contracted.

► Substituting for some jobs. While AI raises production efficiency, it also substitutes for employment, especially in industries with high AI exposure. Since the launch of ChatGPT at the end of 2022, U.S. nonfarm payrolls have increased by 4.7 million, but the gap across industries is striking: employment rose in education and health (+2.78 million), government (+610,000) and leisure and hospitality (+440,000), while industries more clearly affected by AI, such as business services (-370,000), information (-300,000) and financial activities (-90,000), saw larger declines. In addition, according to Dallas Fed research, wage growth is higher in industries with high AI exposure, but the wage gap between younger and experienced groups is larger, which may leave younger job seekers facing greater challenges.

Source: Haver, CICC Research Department

Chart: U.S. tech industry and financial employment have fallen markedly since the end of 2022

Source: Haver, CICC Research Department

V. When does AI turn from "friend" to "enemy"? The paths and tipping points differ between China and the U.S.; U.S. rates at 6%, China's policy tolerance is high

Now that AI has begun to draw resources from society, the next question is: when will AI turn from a "friend" into an "enemy"? This is essentially about measuring the cost that the economy and policy can bear—for example, when resources are all channeled toward technology, how long can the upper end of the K-shape offset the lower end? When will the Fed stop raising rates, or even pivot to cutting? When will Chinese policy pay more attention to weaker domestic demand?

In the specific measurement, it should be noted that because the sources of AI funding and policy objectives differ between China and the U.S., the crowding-out paths and tipping points also differ: 1) Most U.S. AI financing comes from the private sector, so crowding out occurs mainly through interest rates and Fed policy; 2) In China, at least 30% comes from policy-related funds, and bank credit and other channels also have policy guidance, so crowding out occurs more through the tilt of fiscal and credit resources. In addition, China focuses more on long-term strategy, so its tolerance for short-term crowding out is also higher.

► How high a rate can the U.S. bear? About 6% overall, above 7% for AI, about 4% for traditional sectors. The difference between investment returns and financing costs is significant between U.S. AI and traditional demand sectors. The ROIC of the five major cloud vendors (20%) remains far above their WACC (9%), while for traditional demand such as rate-sensitive real estate, the mortgage rate (7.5%) has long exceeded the rental yield (6.86%). Within U.S. equities, the divergence between AI and non-AI is similarly large. We take the information technology and communication services sectors in the S&P 500 plus Amazon in the consumer discretionary sector as the representative of U.S. AI stocks, and the remaining 406 constituents as the representative of non-AI traditional demand. As of the third quarter of 2026, the gap between ROIC (about 35%) and WACC (10-15%) for the AI segment was as high as 21%, while for non-AI it was only 3%.

Chart: As of Q3 2026, the ROIC-WACC gap for the U.S. AI segment was as high as 21%, while for non-AI it was only 3%

Source: Bloomberg, FactSet, CICC Research Department

AI can bear a rate of about 7-7.5%. Assuming ROIC is unchanged, if the gap between ROIC and WACC for the five major cloud vendors narrows to the average of 8ppt before the AI trend took off in 2015-2023, the corresponding WACC would be about 11.9%. Taking into account the impact of rising U.S. Treasury yields on bond costs and equity costs, and assuming β is unchanged, we back out a risk-free rate of about 7-7.5%.

Traditional sectors can bear a rate of about 4%. 1) Taking real estate as an example, assuming the 30-year mortgage rate equals the rental yield in recent years (6.2%), and based on the averages since 2015 of the MBS spread and the primary-secondary spread among the components of the mortgage rate (split into the 10-year Treasury yield, the MBS spread and the primary-secondary spread), we infer a corresponding 10-year Treasury yield of about 4-4.2%. 2) For the "non-AI" segment of the S&P 500 (excluding information technology, communication services and Amazon), assuming the ROIC-WACC gap returns to the average of 4.5% since 2015, the corresponding WACC is 9.7%, which regresses to a corresponding 10-year Treasury yield of 4%.

The overall U.S. economy can bear a rate of about 5.5-6%, considering: 1) AI contributes about 30-50% to U.S. GDP, and 2) the market cap weights of AI and non-AI in the S&P are roughly "fifty-fifty," giving a weighted rate of about 5.5-6% that the overall U.S. economy can bear.

Chart: Assuming ROIC is unchanged, if the cloud vendors' ROIC-WACC gap returns to the average, the corresponding Treasury yield is 7-7.5%

Source: Bloomberg, FactSet, CICC Research Department

Chart: If the mortgage rate equals the rental yield in recent years (6.2%), the corresponding 10-year Treasury yield is about 4%

Source: Bloomberg, Haver, CICC Research Department

► How much cost increase can the U.S. bear? 50% overall, 4-5% for AI application sectors, 8-10% for other sectors. AI production investment on the one hand boosts demand in the upstream supply chain, but it also drives up prices of raw materials and related goods, squeezing profits in other sectors. Our estimates show that assuming cloud vendors' capital expenditure reaches $900 billion, after excluding import leakage effects, if AI-related costs rise 50%, then the $161.7 billion in gross operating surplus (GOS) generated by capital expenditure would be offset by cost increases and import leakage; if costs rise 3.6%, the gains that application sectors obtain from capital expenditure would be crowded out by higher costs; if costs rise 8.6%, the gains of other sectors would also be wiped out. The AI production and infrastructure sectors overall suffer no obvious losses during cost increases, and some even benefit.

Source: BEA, CICC Research Department

► How much crowding out can Chinese policy bear? Because policy focuses more on long-term strategic goals and cannot be simply quantified, the policy tipping point needs to be "measured" from three dimensions: 1) Growth. The "Tutorial 100 Questions" points out that to achieve the long-term growth goal, per capita GDP should reach the level of moderately developed countries by 2035; 2) Employment, which concerns social stability, has not yet fallen to the low point before the "924" policy push; 3) Leverage and balance sheets. Any low-return investment can ultimately only turn into debt and excess capacity. Whether it is bank or local government debt, the credit behind it is ultimately that of the central government. But in a narrow sense, fiscal space is relatively limited (the broad deficit ratio is already above 8%, and the overall leverage ratio of the non-financial sector is 308%, even higher than the U.S.'s 251%).

The "15th Five-Year Plan" devotes a dedicated chapter to "accelerating high-level scientific and technological self-reliance and self-strengthening, and leading the development of new quality productive forces," with independent controllability and technological innovation as the two key medium- and long-term levers. With clear strategic goals, controllable short-term pressure and not-so-large fiscal space, a major shift in fiscal intensity and policy direction may still require some time or greater pressure.

Source: FactSet, CICC Research Department

Chart: China's capacity utilization in computers, communications and electronics, and general equipment manufacturing is at a relatively high percentile

Source: Wind, CICC Research Department

Source of this article: Kevin Strategy Research. Risk disclosure and disclaimer

Markets carry risks, and investment requires caution. This article does not constitute personal investment advice and does not take into account the specific investment objectives, financial conditions or needs of individual users. Users should consider whether any opinions, views or conclusions in this article are suitable for their particular circumstances. Investment based on this is at the user's own risk.