Published on: 2026-04-09
This first article in the China Demand Dial series examines credit as a key lever in China’s economy. Changes by the People’s Bank of China (PBOC) in monetary flow impact not only China but also global commodity markets, influencing refinery activity, metal inventories, and pricing. For commodity traders, monitoring China’s credit trends is essential for distinguishing meaningful signals from background noise.
In this context, 'cooling' means that while headline financing appears stable, private credit demand and the loan mix are weakening.
A key caveat is that January data is typically volatile, and this year timing effects were significant. February data provides a clearer picture, showing private credit demand remained weak, especially in household borrowing.
In January 2026, new bank loans totalled 4.71 trillion yuan, below the forecast of 5.0 trillion yuan and last year’s 5.13 trillion yuan.
Total social financing (TSF) increased by approximately 7.22 trillion yuan, exceeding expectations, largely due to government borrowing and bond issuance.
Infrastructure project approvals reached about 295 billion yuan in early 2026, indicating policy front-loading, though local government balance sheets remain a constraint.
From January to February, fixed-asset investment grew 1.8% year on year, and infrastructure investment jumped 11.4%. This suggests that government-driven projects are doing better than household credit demand.
As the second largest economy in the world, China’s credit cycle matters for commodities because it funds activities that consume raw materials. The catch is timing. Credit changes tend to hit the real economy with a lag of a few quarters, and global spillovers can take longer.
February’s financing data shows that the slowdown in private credit demand is ongoing, not just a one-off issue in January.
The pace of local government special bond issuance provides real-time insight into the level of infrastructure financing.
Monitor property sector loan extensions under the “whitelist” programme to assess whether they stabilise projects without increasing buyer demand.
China’s credit system differs from Western rate-led cycles. While policy rates are important, the system relies more on quantitative tools and directed flows. Markets therefore monitorTSF, in addition to interest rates, to interpret policy intentions.
Simply put, new loans represent bank lending, while TSF measures total credit flow, including loans, bonds, and other channels. TSF may appear strong even if bank lending is weak, which is significant for commodities.
Recent data reveals underlying strain. Banks issued 4.71 trillion yuan in new loans in January, a large figure but one that fell short of expectations. The breakdown is more telling.
Household loans rose about 456.5 billion yuan in January after contracting about 91.6 billion yuan in December.
Corporate loans jumped to about 4.45 trillion yuan in January from about 1.07 trillion yuan in December.
This mix is consistent with directed credit. Banks have been nudged to support priority areas such as infrastructure and technology upgrading, which tends to show up first in corporate lending rather than household borrowing.
TSF data shows a similar trend, rising to about 7.22 trillion yuan in January. New loans, however, have recently contributed only half or less of total financing growth, with government borrowing playing a larger role. For commodities, this distinction is important: bank lending to builders and factories directly boosts metal and energy demand, while bond-driven support impacts demand more gradually.
In practice, bond issuance can support demand, but its impact on commodities typically shows only after funds are used for procurement, construction, and material orders.
The credit impulse is the year-on-year change in new credit growth, adjusted for GDP. This measure helps markets focus on the direction of credit creation rather than just its level.
China's credit cycles can quickly influence rates, FX, and risk sentiment, but their impact on commodities typically occurs through a slower, real-economy channel. Research shows that commodity prices are a key transmission channel, accounting for a significant share of producer price spillovers to advanced economies.
The sequence tends to look like this:
Funding loosens or is directed (credit growth picks up, TSF composition shifts, project approvals rise).
Project activity starts (infrastructure, manufacturing capex, grid build, local government projects).
Physical demand follows (steel, copper, aluminium, diesel, and downstream transport demand).
Prices to be confirmed later (inventory draws, tighter nearby markets, curve shifts).
This lag is important. A credit impulse today does not immediately translate into copper demand; it initiates a process that may drive demand over the following quarters.
Confidence is a critical factor in the current cycle. If households and private firms remain cautious, increased credit may not translate into physical demand, resulting in muted commodity responses even when TSF figures appear strong.
This is where property still matters. Mortgages remain the biggest swing factor in household lending, and housing construction is one of the most materials-heavy parts of the economy. If activity stays weak, credit can look healthy on paper while the parts that drive steel, copper, aluminium, and diesel demand remain soft.
To limit near-term damage, authorities have leaned on the “whitelist” programme, allowing banks to extend loans for up to five years for approved projects. That helps reduce the risk of unfinished developments and forced liquidations.
But extensions are not the same as new demand. Unless buyer confidence improves, funding support mainly buys time rather than restarting the cycle. That is why household lending trends remain the cleaner tell.
Shadow credit refers to non-bank or off-balance-sheet channels that sit inside China’s broader financing system, especially trust loans, entrusted loans, and undiscounted bankers’ acceptances. These channels matter because they can fund property-linked and local-government-linked activity that is highly commodity-intensive, and they can change the true credit impulse even when headline TSF looks steady. The sector is smaller than at its peak. PBOC-linked estimates show shadow banking assets fell to a single-digit share of bank lending by 2023, down sharply from the late-2010s, but the flows can still amplify turns at the margin.
Not all commodities respond on the same schedule. The lag depends on where the commodity sits in the chain, how much inventory can buffer demand, and how policy constraints shape imports and production. Metals tied to construction and grids usually react earlier than energy, while oil can lag because quotas, margins, and strategic buying often matter as much as the credit impulse.
Industrial metals, particularly copper and aluminium, tend to respond quickly due to their proximity to infrastructure, grid, manufacturing, and construction supply chains. China’s significant demand share means macroeconomic shifts can rapidly affect positioning and physical premia. An IMF working paper highlights China’s substantial influence on demand for major industrial commodities, especially metals.
Copper serves as a key barometer, providing a forward-looking indicator of construction and electrification demand. When a credit shift is credible, local markets often lead, with Shanghai pricing and inventory signals preceding broader global confirmation.
Iron ore exhibits more volatile, boom-bust behaviour due to its link to steel and, in turn, construction. The property downturn has kept steel demand subdued. Industry commentary for the 2026–2030 period anticipates limited growth in property, infrastructure, and manufacturing through at least 2026.
This context is important for interpreting credit data. Infrastructure investment may support steel demand to some extent, but a weak housing market can limit overall growth.
Oil does not follow a straightforward “credit in, crude up” pattern in China, as policy and incentives play a significant role. Quotas, margins, and strategic purchases affect imports and refinery activity. China maintained the 2026 non-state crude import quota for independent refiners at 257 million tonnes, unchanged from 2025, and continues to issue these quotas in batches. State-owned majors operate under a separate framework, further decoupling crude demand from credit trends.
Therefore, in the oil market, confirmation from refiners, margins, and import volumes is necessary, not just credit data.
Term structure serves as a valuable confirmation tool. When real demand increases, nearby contracts strengthen relative to deferred ones. Conversely, weak demand or ample inventory typically results in contango.
A key caveat: in metals, curve shape may also reflect inventory location, financing conditions, and warehouse dynamics. Use it as confirmation rather than an initial signal.
Here are the next checkpoints that decide whether “cooling” is a pause or a trend:
Monthly aggregate financing:
February’s data already shows the loan mix is still weak, with TSF growth steady at
8.2% year on year. The key question now is whether future data will show more private borrowing, or if growth will keep coming mostly from government and bond support.
Infrastructure funding follow-through:
Project approvals signal intent, but financing and execution are crucial. The early 295-billion-yuan batch is notable, but markets will focus on the pace of issuance and conversion of the project pipeline into actual orders.
Property and household credit:
Whitelisting loan extensions might help steady the system, but household lending is still the best sign of whether buyers are coming back. February already showed more weakness, with household loans shrinking after a rebound in January.
PBOC policy stance:
The seven-day reverse repo rate stands at 1.40%, but the more current signal is that the PBOC has reiterated an appropriately loose monetary stance and a commitment to ample liquidity. The question is less whether policy is supportive in principle, and more whether that support can lift private borrowing rather than just sustain government-led credit.
Metal inventories and premia:
SHFE (Shanghai Futures Exchange) and LME (London Metal Exchange) warehouse stocks are widely watched as near-real-time proxies for deliverable inventory. Falling stocks and firmer physical premia tend to confirm genuine consumption, while rising stocks can suggest positioning is running ahead of demand.
Customs import volumes:
For context, China imported 557.73m tonnes of crude oil and 1.26bn tonnes of iron ore in 2025. In January-February 2026, crude imports rose 15.8% year on year to 96.93m tonnes, while iron ore imports rose 10% to 210.02m tonnes. But higher import volumes alone do not confirm stronger end demand. Recent data suggests part of the increase reflected stockpiling and inventory building, especially in crude and iron ore, so traders still need confirmation from refinery runs, steel output, inventories, and physical premia.
Copper concentrates are dominated by Andean supply, led by Chile and Peru, while LNG supply is split among Australia, Qatar, the US , and other producers, with volumes sensitive to spot prices and seasonal demand.
March activity data was a bit stronger, with the official manufacturing PMI rising to 50.4 from 49.0 in February.
Disclaimer & Citation
This material is for information only and does not constitute a recommendation or advice from EBC Financial Group and all its entities (“EBC”). Trading Forex and Contracts for Difference (CFDs) on margin carries a high level of risk and may not be suitable for all investors. Losses can exceed your deposits. Before trading, you should carefully consider your trading objectives, level of experience, and risk appetite, and consult an independent financial advisor if necessary. Statistics or past investment performance are not a guarantee of future performance. EBC is not liable for any damages arising from reliance on this information.
