Vol.226 Macro Talk 108 | China's 2026 Midyear Economic Review and 2 Strange Chinese Economic Phenomena (Recorded 7.16)

Summary

• The main drags on the economy in H1 were government-related investment and property. Special-bond quotas were unchanged, but Q2 issuance and spending slowed sharply: “The money has arrived, but has not yet translated into physical work on the ground.” A stronger push may have to wait for the July Politburo meeting to set the direction. Foreign trade offset the drag—the analogy from 丰叔 was that among 3 or 4 chess pieces, “1 or 2 are useful every year,” while 李翔 described it as “rotating a few lids over the same pots.”
• The evidence that property is nearing a bottom is mounting. New-home inventories have fallen month on month for roughly 4 consecutive months, meaning supply is shrinking faster than demand and the market is moving closer to balance. Prices rose month on month in 20 of 70 cities; “tier-1 cities have at least stabilized at the bottom.” In aggregate, annual transaction volume has fallen from more than RMB10T at the peak toward RMB6-7T and “will be hard-pressed to go lower”—“it has already fallen quite far, perhaps even slightly overshot.”
• The biggest long-term shift is in the financial structure. Loans now account for only 60% of total social financing, and the structure “will move in this direction from now on and will not turn back.” Roughly RMB50T of time deposits maturing this year is prompting households to activate their savings: household deposits increased by RMB7T in H1, but RMB3T less than last year, with money flowing into insurance, wealth management and equities. Smaller insurers have reported premium growth “beyond their own expectations.” The June rebound in deposits was merely a midyear bank-gathering pause; net household deposit declines will probably return in July.
• The first strange phenomenon is the turn in state-capital investment from shunned to sought after in 3.5 years. During the “worst 3 years” from H2 2022 to H1 2025, state capital was almost the only investor backing companies in the growth phase. As a result, “the final rounds for all those companies now worth RMB100B or more, in some cases approaching RMB1T, were funded mainly by state capital.” Financial investors are now copying the state-capital playbook, while hot technology projects deliberately reserve allocations for state-backed investors. Document No. 54 is applying the lessons of rural banks and putting the brakes on county-level activity.
• The second strange phenomenon—and the most tradable segment of this episode—is the stabilization-fund trade. In May-June (5-6), the stabilization fund sold an “extremely large amount, extremely quickly—more than 90% of its position” to slow the index’s rapid rise driven by individual stocks. Insurers bought the blue-chip inventory at distressed prices, with P/Es below 10x and dividend yields around 4%. Household savings flowing into insurance could create a closed loop in which “retail investors’ short money becomes long money,” while the stabilization fund “got all its bullets back” in preparation for an AI-bubble bust: “The only long money is insurance.”
• The durability of the export bright spot rests on 3 conditions. The RMB must remain relatively strong—China’s surplus with Europe hit a record and the EU continues to complain that the currency is undervalued. Higher export value-add is absorbing the currency pressure: June chip export volumes edged down, but unit prices rose, benefiting from AI infrastructure spending. Structural adjustment has also passed its point of fundamental change. If an AI bubble bursts, exports could take a short-term hit, but “after the bubble bursts, people care more about buying cheaply… for China, that is not necessarily a bad thing.”
• The hidden logic behind Moutai’s channel reform is consumption tax. If collection at the retail end is implemented and the direct-sales plus agency-sales system reaches 90%, factory settlement would effectively be based on the RMB1,500 retail price, allowing Guizhou to retain close to 90% of the consumption tax, versus only 30-40% when products were sold at the RMB1,100 wholesale price. “We can try watching it for another year.”
• The primary-market temperature gauge is clear: after robotics cooled, quantum computing and nuclear fusion are the season’s most extreme trades. The tell for overheating is when investors rush into themes that “cannot be proven or disproven.” By contrast, new-drug licensing reached $110B in H1 and could approach $200B for the full year, giving China nearly 70% of the global market.

Deep dive

1. Midyear report’s main line: government spending underperformed, and money has not translated into physical work

• 李峰峰（丰叔）summed up the July 15 midyear data this way: government-related projects in H1—primarily infrastructure and investment—were weaker than expected. Special-bond quotas were unchanged, but the pace slowed: Q1 was in line with expectations, while Q2 was slower than expected. On a year-on-year basis, last year’s data imply that a substantial portion of the full-year quota was front-loaded. This year, special bonds for Q2 were not issued in volume until early June: “The money has arrived, but has not yet translated into physical work on the ground.”
• Why the slowdown? 丰叔’s guess is that officials may not have been sure where the money should go. Beyond the 6 major networks and urban renewal, the next set of priorities has not been finalized. “Perhaps they are waiting for the July Politburo meeting to set the direction, and then everyone can spend at a normal pace.”
• Government investment and property accounted for most of the negative pull from investment. Manufacturing “pulled itself back to flat”: expansion in new capacity in technology, electronics and other areas offset declining investment in traditional capacity such as light-industry products.

2. The PBOC’s message: social financing will no longer grow rapidly every year, and pure investment-led GDP growth is far less efficient

• At the Lujiazui Forum, the PBOC governor said China’s total social financing “is unlikely to keep growing rapidly year after year and quarter after quarter as it did before,” reflecting changes in the country’s development stage and economic structure. 丰叔’s translation: “The efficiency of relying purely on investment to drive GDP has fallen substantially.” Each yuan of investment now generates materially less GDP, while total social financing edged down year on year in H1.
• The structural shift matters more: loans now account for only 60% of total social financing, while direct financing after bond issuance accounted for nearly half last year. “This structure will move in this direction from now on and will not turn back”—the shares of bonds, direct financing and capital-market financing will gradually rise.

3. Property: unsold inventories have fallen for 4 months, and tier-1 cities are stabilizing at the bottom

• The logic chain is intact: new-home inventories have declined for roughly 4 consecutive months, while new-home investment and starts have fallen even faster. “Supply is shrinking faster than demand, so at the new-home level you are closer to supply-demand balance.” Prices rose month on month in 20 of 70 cities, and both official commentary and the data support the view that “tier-1 cities have at least stabilized at the bottom.” Shenzhen and Shanghai are doing somewhat better.
• This fulfills the third of the 4 forecasts the 2 speakers made at the start of the year. The first—consumption stabilizing around midyear—is showing early signs of coming through; the second—a move out of simultaneous CPI and PPI declines—has been visible continuously since March. The fourth, on the RMB and HKD, was not discussed in this episode.

4. The aggregate floor: from more than RMB10T to RMB6-7T, it will be hard to go lower

• 丰叔’s order-of-magnitude view is that annual property transactions could ultimately fall from more than RMB10T at the peak to half that, or RMB6-7T. “It will be hard to go below an annual total of RMB6-7T or RMB7-8T. It has already fallen quite far, perhaps even overshot slightly.” The anchor is the medium- to long-term increase in urban permanent residents, plus new white-collar demand created by economic transformation. One reason higher-end housing in Shenzhen has risen is that employees at newly listed technology and chip companies have used their incentive income to buy homes.
• Policy is moving in parallel: since May, governments in tier-1 cities have begun acquiring old, dilapidated housing. Once the rent-to-price ratio exceeds 2, “at least the rent can cover the potential interest cost.” The units can then be converted into housing for new urban residents and apartments for young people. The program has only just begun and remains limited in scale.

5. Consumption: May was mainly a high-base issue from subsidies; the clean read comes in July and August

• Last year’s consumer subsidies were concentrated in 4-5 (April-May), with the eligible categories later broadened to communications equipment, computers, home furnishings and construction materials. This year’s Q2 therefore faced a high base. Most cities had exhausted their subsidy budgets by mid-June, while only the wealthier cities lasted through the end of June. 丰叔’s rough calculation is that in June the base effect had been erased by 50-60%, and retail sales still grew by 1-point-something on top of that.
• The mix tells the story: rural and county-level areas were more resilient than cities, lower-tier markets outperformed higher-tier ones, and sales below the designated-size threshold were stronger than those above it. Services grew clearly faster than goods retail. The drags were precisely last year’s subsidy categories—autos, appliances, communications equipment, building materials and home renovation. “It is hard to say how much of the decline is demand and how much is the base.” The real test comes after the subsidy base effect fully disappears in July and August.

6. Household savings activation: RMB50T of time deposits are maturing, and June was only a bank-deposit-gathering pause

• The data 丰叔 cares about most: roughly RMB50T of time deposits mature this year, and rollover rates have fallen to just 1-2%, prompting households to activate their savings. Household deposits rose by RMB7T in H1, but the increase was RMB3T smaller than last year. Deposits fell on a net basis in April and May. Why did they rise in June? Banks were gathering deposits for the June 30 midyear assessment; some large banks relaunched 5-year large-denomination certificates of deposit and pulled some of the already activated money back into term deposits, widening the M1-M2 scissors spread slightly.
• Once the assessment period passes in July, “the free market starts doing its work again.” There is a high probability that net household deposits will fall again, while deposits at non-bank financial institutions continue to rise—corresponding to insurance, bank wealth-management products and brokerage equities. Low rates still have to serve 3 purposes simultaneously: internal debt restructuring by swapping high-cost debt for low-cost debt and short-term debt for long-term debt; stimulating the economy; and activating savings.

7. Insurance confirms the shift: smaller insurers are seeing premium growth “beyond their own expectations”

• Since Q4 last year, medical, life and health insurance have all maintained unexpectedly strong growth. The effect is especially pronounced at smaller insurers because of their low bases: a company with an annual premium target of RMB20B generated roughly RMB8B in Q4 alone last year, while its first 2 quarters this year already reached 70-80% of last year’s full-year total. This is the clearest channel among the destinations for household savings released as time deposits mature.

8. The 3 questions for exports: strong currency, high value-add and structural transformation

• The sustainability case rests on 3 pillars. First, the RMB must remain relatively strong to preserve purchasing power for imports and narrow the trade surplus moderately. China’s surplus with Europe has hit a record, and “the EU is still complaining that the RMB is undervalued”; that strength is likely to persist “over the visible horizon of the next 1, 2 or 3 years.” Second, growth is coming from higher-value-added products such as electromechanical goods. Responding to the observation that June integrated-circuit export volumes fell by a fraction of a percent year on year: “Then you should assume unit prices rose… this is clearly benefiting from AI infrastructure.” Third, “structural adjustment is basically complete”—China has moved beyond the era when light-industry products carried its exports, allowing it to withstand a strong currency.
• New quality productive forces, including internet services rather than only chips, new-energy vehicles and new-energy components, contributed 40% of the economy’s incremental growth.

9. If the AI bubble bursts: a short-term export shock, but “not necessarily a bad thing” over the medium to long term

• 丰叔’s dialectical framing is worth preserving in full. During a bubble, “people care about whether they can get the product; if I can grab it, the price does not matter yet.” Once the bubble bursts, “people care more about buying cheaply—whoever offers the best value gets my business.” That “is not necessarily a bad thing for China.” The process is the same one China has gone through in other export industries.

10. The K-shaped recovery’s inflows and outflows: keep the floor from falling and the ceiling rising

• What households care about is whether the lower diagonal line of the K can stop moving down and whether the upper line can keep moving up. “As long as the bottom stops falling and the top keeps rising, we have crossed the hardest phase—the one where water enters from one side and exits from the other while the level in the pool stays unchanged.” The H1 data suggest that “inflows and outflows are roughly balancing.”
• The policy pieces rotate. Last year, export challenges were offset by government investment and the early release of big-ticket consumption subsidies. This year, exports and foreign trade are offsetting slower government spending. “There are only 3 or 4 chess pieces; 1 or 2 are useful each year.” 李翔 called it “rotating a few lids over the same pots.” How should one explain 10%-plus growth in tobacco and alcohol? 丰叔’s honest non-answer: “I don’t know either… when you ask something that specific, I don’t know.”

11. Autos: policy support has moved on, leaving an industry with “3 meals and a late-night snack” of competition

• The signal is clear: apart from halving the purchase tax for new pure-electric passenger cars, extended-range hybrids will no longer receive exemptions. “There is no longer any need for major policy to stimulate this industry.” China’s auto competitiveness has turned the page. The country is now launching 3.6 new models every day, with more than 500 new models introduced by May. BYD joked that this has become “3 meals a day plus a late-night snack.”
• The result is exceptional supply-chain efficiency and an exceptionally high survival bar. Auto exports rose roughly 70% this year, in 丰叔’s recollection. 李翔 asked whether local-government backstops were preventing industry clearing. 丰叔’s response: the problem was more visible in the past; this year, every province and city ran a deficit in Q1. Under the unified national market, local governments “cannot subsidize one specific company,” and “there are more constraints today than before.”

12. Biopharma: $110B in licensing in H1, with China’s global share approaching 70%

• New-drug licensing reached $110B in H1. Last year’s full-year total was more than $130B, already accounting for over half of the global market. At the current pace, China could reach $200B this year; if the global market is slightly above $200B but below $300B, China “will take close to 70% or more of this market.” The progression was “just over 30% the year before last, then just over 50%, and this year it will clearly be around 70%.” Like new-energy vehicles, these are “2 extremely striking industries,” with efficiency forged through intense competition.

13. The hidden logic of Moutai’s channel reform: keep 90% of the consumption tax in Guizhou

• 丰叔’s half-joking, half-serious hypothesis is that collection of the consumption tax at the retail end, with the central-local revenue split tilted toward local governments, is an expected direction after the Third Plenum of the 20th Central Committee. Local governments currently live on transfers and special-purpose bonds earmarked for specific uses. “Spending other people’s money is painful,” so the center will ultimately give local governments a new tax source, much as the 1994 tax reform left land-related taxes with local governments.
• Moutai’s channel reform, launched in January, shifted distributors to agency sales—they hold no inventory, earn a 5% commission and cannot profit from price spreads—alongside i Moutai e-commerce direct sales. If the direct-sales system reaches 90%, factory settlement would effectively be based on the RMB1,500 retail price, allowing Guizhou to retain close to 90% of the consumption tax, versus only 30-40% when products were previously sold at the RMB1,100 wholesale price. Why is the reform possible now? 丰叔 offered 3 guesses: several changes of party secretary have reduced resistance; short tenures have diluted the old distribution of interests; and the Guizhou government has a direct incentive to support the reform through the consumption tax. “We can try watching it for another year.”

14. Strange phenomenon one: state-capital investment went from shunned to sought after in 3.5 years

• The shift had 3 stages. From 2017 to 2020, direct investment by state capital within the fiscal system was “not welcome.” Entrepreneurs feared China-US uncertainty and disliked slow decisions and heavy procedures; there were also disputes over equity in form but debt in substance and joint liability for founders. Starting in 2022-23, State Council executive meetings and the Politburo repeatedly called for investment in early-stage, small and technology companies. That was followed by “the coldest, coldest winter of my investment career”: foreign capital could not invest, RMB fundraising was difficult and confidence was weak. During the 3 worst years from H2 2022 to H1 2025, “only state capital was investing” in growth-stage companies valued at more than RMB1B. The result: “The final rounds for all those companies now worth RMB100B or more, in some cases approaching RMB1T, were funded mainly by state capital.” State capital made money in a very short capital-market cycle.
• The reversal is now visible. Financial investors have begun “copying the state-capital playbook,” subconsciously assuming that state capital has better knowledge of which companies will receive policy support and reach an IPO window. When competition for frontier technology projects is intense, projects will “deliberately reserve some allocation for state capital.” Backing from the state capital of an important region can even serve as proof that “the project is positioned at the right frontier of technology.” 李翔 compared it with receiving an investment from Tencent during the mobile-internet era.
• 丰叔 left 2 questions open: “Is it reasonable for state capital to cover the entire industry chain and full life cycle? Is it reasonable for financial investors to copy the state-capital playbook?” These “are probably 2 questions for time to answer.” He was not dismissing the model wholesale: Shenzhen Capital, Suzhou, Beijing and Shanghai “certainly have people who can do it.” But venture capital needs a high tolerance for failure and a 10-year horizon, neither of which fits comfortably within government evaluation systems. Early-stage investment by government “is a process of accumulating seeds, definitely not a process of accumulating tax revenue.”

15. Document No. 54 is a risk-control measure: applying 20 years of banking lessons to hit the brakes

• 丰叔 compares Document No. 54, which prohibits counties and districts from setting up new funds, with the previous round of financial restructuring. In 1999, 4 asset-management companies were created to strip bad loans from the Big 4 banks. From 2003 to 2005, city commercial banks were consolidated into provincial- and city-level banks. But problems emerged at the final tier: of more than 1,000 rural banks, nearly 300 have been absorbed or merged in the past 2.5 years, and eventually “perhaps 40-50% or more” could be consolidated. This time, direct investment is opened to provinces and sub-provincial cities, while counties and districts are coordinated by the province: “Jiangsu can set up a bank, Nanjing can set up a bank; whether Changshu can do so independently is for Jiangsu to decide.”
• The next step could resemble the consolidation of local credit cooperatives into provincial commercial banks: local state-capital investment platforms may be integrated into 1 or 2 unified platforms. The Shenzhen Capital model came first; Suzhou Venture Capital and Western Venture Capital have followed, while Shanghai has consolidated into 3 systems led by Guosheng, Guoji and Guotou.

16. Put the money the state values most to work: social-security fund-of-funds and full mobilization of participants

• Early this year, the NDRC used ultra-long special government bonds to set up 3 fund-of-funds vehicles of RMB50B each. The social-security fund is also committing RMB10B-RMB20B to technology-innovation funds of funds in multiple provinces and cities this year, typically contributing 40%. 丰叔’s interpretation: the amount is not large for the social-security fund, “but putting social-security money in means the state has put the money it values most into these funds of funds,” mobilizing the willingness and capital of every participating party.
• The full picture includes government, banks—financial holding companies are providing leverage to local state capital for equity investments and have also made money in this wave, while bank-affiliated investment has become highly active—and CVC industrial capital, with CATL participating in numerous frontier-technology investments. Awareness, willingness and incentives have shifted across all 3 economic groups. “China’s most important objective has been achieved.” The underlying logic is that the upper line of the K needs long-duration R&D investment, not loans: “Different economic structures require different financial structures.”

17. Strange phenomenon two: the stabilization-fund trade—national-team selling, insurers absorbing, households entering

• The first leg: China Securities Finance and Central Huijin entered the market in 2023 after volatility triggered by snowball knock-ins. In February 2024, when the index fell to 2,600-2,700 and was deeply oversold, they bought blue chips and broad-market ETFs at scale. In Q4 last year, and especially in May-June (5-6) this year, they sold “an extremely large amount, extremely quickly—more than 90% of the position” to dampen the index’s rapid rise driven by individual stocks. The comparison was South Korea’s Samsung Electronics and SK Hynix, and Taiwan’s TSMC and UMC, each responsible for roughly 70% of index volatility. The national team effectively recovered the RMB1-2T invested since 2023. “The main purpose was stabilization, not getting the money back.”
• The second leg was the collapse in “old-guard stock” prices: dividend yields reached 3%-plus and 4%-plus, while P/Es fell below 10x and, in some cases, below 5x. Almost all of the insurers entering the top-10 shareholder lists of blue chips this year were insurance companies, because equity risk coefficients have been cut repeatedly over the past 1.5 years. Insurers want exactly these high-dividend, high-stability blue chips. The third leg is household money moving into insurance because time-deposit rollover rates are too low; as the income side expands, insurers invest even more. Under “an extremely unusual and extreme historical market,” the 3 legs could form a closed loop.

18. What the loop means: retail investors’ short money becomes long money, while the stabilization fund reloads completely

• “The only long money is insurance, because insurance is the only money with a duration of more than 20 years.” The extreme market transferred the basic blue-chip inventory to long-duration capital at exceptionally low prices and in substantial size. Households can capture part of the equity-market return through insurance; if that positive feedback loop works, it would address the third pillar of the social-security and medical-insurance system. Meanwhile, the stabilization fund “got all its bullets back.” If an AI-bubble collapse causes a rapid fall in the equity indexes, it would again have almost its full arsenal available for stabilization. Over the past 3 days, the national team has already bought broad-market ETFs again amid volatility in Japanese, Korean and US equities.
• 丰叔 stressed that this was not a fully designed plan: “I don’t believe these things can be designed 100% from the beginning. My guess is that only 40-50% can be designed… no one fully planned for most of the sold inventory to be taken up by insurers.” 李翔 called it “the reverse of the fallacy of composition”: each participant’s rational behavior created gains for everyone. The backdrop is RMB170T of household deposits, the largest pool in human history. “Once it starts moving, it can also be frightening… it is best if it moves in an orderly direction.” At the start of the last property upswing, household deposits totaled only RMB40-50T.

19. Methodology and the primary-market thermometer: “cannot be proven, cannot be disproven” means overheating

• The Buffett-style conclusion: “Your only lasting valuation standard is whether you bought cheaply.” The greatest cost of buying at the bottom is another 5% or 10% decline, but you know where the bottom is. The final leg of an uptrend is the hardest place to make money because “once you miss the top, the psychological factor is that I once made 100.” “Buying cheaply enough and having money that is long enough—these 2 things are the only truly effective strategy.” Yet only long-duration capital dares to buy deeply oversold assets; under short-term performance reviews, “even if you know it is cheap enough, you still cannot buy it.”
• The primary market is hot only in narrow themes, and the heat rotates in bursts. Robotics is no longer as hot as it was in Q1; the most extreme themes this quarter are quantum computing and nuclear fusion because they are “destined with certainty to be the technologies of the future,” while investors are also copying the last trade—“I invest in whatever the state encourages.” The overheating test is straightforward: “When most of what people prefer to invest in is something that cannot be proven or disproven… when everything is hot, you know the market is at least somewhat overheated.” 李翔’s closing question: “Isn’t this a scientific question? Can capital solve it?”