China's economy is sending two signals at once. The G7 should read both before it builds the next critical-minerals deal.

For the first time in more than three years, China's retail sales declined in May. The figure, flagged by Polymarket's data desk at 15:17 UTC on 16 June 2026, lands at an awkward moment. G7 leaders were already in their working sessions by 17 June, drafting language on critical-mineral supply chains — the rare policy file where Western governments, Beijing, Jakarta, and Santiago share an agenda. And on the same day, Reuters Breakingviews published a column arguing that Beijing's job-first industrial policy may be blunting the productivity gains the country is racing to capture from artificial intelligence.

The Western policy reflex is to treat China as a single, adversarial supply-chain problem. That framing is increasingly out of sync with the data coming out of Beijing. The economy is sending two signals at once: a consumer base that is exhausted, and a state that is still willing to spend. The G7, which is convening on the assumption that the binding constraint on critical-mineral supply is geopolitical, is debating the wrong lever.

The consumer is no longer the engine

Reuters's Polymarket feed is blunt about the headline: May retail sales fell for the first time in over three years. This is not a small data point. Retail sales are the cleanest read on the household side of the Chinese economy — what people are actually spending on goods, week to week, after the Lunar New Year. A three-year run of positive prints has been one of the standard pieces of evidence Western economists cite when arguing that Beijing's model is durable. That evidence has now thinned out in a single monthly release.

The mainstream wire read will be that this is a post-stimulus hangover — Beijing front-loaded demand last year and households are now paying off the credit cards. That is plausible. It is also incomplete. The structural read is that Chinese households entered 2026 carrying a balance sheet scarred by three years of property-sector deleveraging, weak wage growth outside the platform and EV sectors, and a youth unemployment rate that has only recently been restored to official publication. The retail figure is the surface. The underlying story is that the Chinese consumer, the great hope of every Western multinational earnings call since 2023, is not yet ready to be the marginal buyer of the world.

The state is still spending — but on what?

Reuters Breakingviews, published at 04:10 UTC on 17 June, makes the related point from the other direction. Beijing's instinct when consumer demand softens is to redirect credit and policy support toward employment, particularly in sectors that absorb large numbers of relatively low-skilled workers. That is a defensible political choice. It is, however, a poor match for an economy that is simultaneously trying to lead the world in frontier AI capability. The column's argument is that subsidising labour in legacy industries slows the labour-market churn that AI deployment requires. If the state props up the human call-centre worker, the AI call-centre rollout proceeds at the speed the bureaucracy tolerates, not at the speed the technology permits.

This is the dilemma that does not survive translation into G7 communique language. Western capitals want China to be a reliable supplier of lithium, cobalt, rare earths, and graphite. Beijing is signalling that it is willing to be a reliable supplier. The risk is that the same state that secures the supply is also the state whose industrial policy is now visibly torn between two of its own priorities: a consumer recovery it cannot engineer cheaply, and an AI buildout that requires the consumer recovery to fail.

The G7 lever is real, but smaller than it looks

Reuters's G7 thread, posted at 04:50 UTC on 17 June, frames the summit's critical-minerals track as a question of diversification away from Chinese processing capacity. The argument is straightforward: a small number of firms in a small number of Chinese provinces refine the majority of the world's battery-grade lithium and a high share of the rare earths that go into defence, EVs, and wind turbines. Concentration is a vulnerability. Diversification is the policy response.

The steelman of the Chinese position is worth giving. China's dominance of mid-stream processing is not a theft; it is the outcome of a fifteen-year industrial policy that subsidised the unglamorous, capital-intensive, environmentally costly middle of the value chain. The West declined to make that investment. Beijing is now harvesting the return. The Chinese counter-frame, carried in outlets from Xinhua to the South China Morning Post, is that "de-risking" is a euphemism for protectionism dressed up in security language, and that it will slow the global energy transition by raising the cost of the inputs both economies need. That argument is not without merit, and it deserves to be weighed against the concentration-risk argument rather than dismissed.

Two signals, one summit

What the G7 should be reading, then, is not a single China file but two. The first is the supply-side file it has already prepared: a slower, more diversified critical-minerals base, built with partners in Indonesia, Chile, the DRC, and Argentina, and a coordinated stockpiling regime. The second is the demand-side file it has barely begun to write: what happens to global commodity prices, and to the marginal pricing of lithium and rare earths, if China's households do not return to growth in the second half of 2026.

If retail sales stabilise in June, the supply-side file is the one that matters and the G7 agenda is correctly calibrated. If June prints another decline, the demand-side file becomes the binding constraint. A diversified critical-minerals base built for a Chinese consumer who is not buying does not de-risk anything; it just builds inventory at a higher cost. Beijing's own hand is forced here. A consumer recovery requires either a property-sector reflation, which the leadership has shown little appetite for, or a household transfer, which the fiscal system is poorly equipped to deliver. The state that built the world's processing capacity may not be able to build the consumer to match it.

The honest uncertainty in this picture is on the consumer side. The May retail figure is one data point. The Breakingviews column is an argument, not a verdict. The G7 communique will be drafted as if the supply-side file were the whole story. It is not. The Monexus view is that Western capitals are preparing for the China of 2024 — confident, manufacturing-led, structurally expansionary — while the data coming out of Beijing at 04:10 and 15:17 UTC this week describes a more complicated country, one that is simultaneously indispensable to the energy transition and visibly short of domestic demand to justify its own build-out. The summit's working language should reflect that, even if the communique's language does not.

Desk note: Monexus is steelmanning both the Chinese position on critical-minerals processing and the consumer-spending data that complicates it. Western wires have largely framed the G7 file as a clean diversification story; we are reading it against the same week's retail and industrial-policy signals.

Wire provenance

This editorial synthesis draws on the following public wire/social posts:

• http://reut.rs/4a1l1Ey
• http://reut.rs/4vN8eOe
• https://x.com/polymarket/status/2067107522616369152