Economy

Mapping the contours of Chinese policy transmission at home and abroad

China’s place within international trade networks and global supply chains makes the propagation of Chinese shocks a global phenomenon. This column discusses the channels of domestic and international transmission and shows that vulnerabilities extend also to large and seemingly closed economies like the US. Demand for industrial commodities plays a key role, which also translates into financial vulnerabilities for exposed emerging markets. But overall, amplification through the global financial cycle remains limited.

Debates about the global effects of monetary policy have long been dominated by the US. We know global spillovers from Federal Reserve shocks to be powerful (Acharya et al. 2025), operating not only through trade but also via investors’ risk-taking (Rey 2013), and affecting countries differently depending on their exposure to global financial cycles (Faia et al. 2024).

Yet the world economy has changed dramatically over the past two decades. China is now the world’s second largest economy and has become a central node in global goods trade and production networks (Figure 1). Despite this transformation, our understanding of how Chinese monetary policy transmits both domestically and internationally remains limited. This gap is increasingly difficult to justify, but not surprising. Over time, the monetary policy framework and toolkit of the People’s Bank of China (PBoC) have evolved considerably to meet the central bank’s multiple policy objectives. This makes the PBoC’s stance difficult to frame in a unified and coherent analytical framework.

Figure 1 China’s position in the global trade network

Notes: Directed network of bilateral goods exports, as reported in the IMF Direction of Trade Statistics (DOTS). Mirror data is used to fill missing bilateral trade whenever available. European Monetary Union (EMU) trade with non-EMU countries only consolidated into a single node. The node size is scaled to reflect the sum of each country’s gross external trade as a share of world total trade (gross exports and imports). We focus only on nodes that have either total exports or total imports of at least USD 10 billion. The edge width also captures the value of gross trade between each pair of countries (nodes), and is normalised between 0 and 20. For comparability, the USA node is positioned in the centre of the graph. Position of the remaining nodes is determined using the Fruchterman-Reingold force-directed algorithm, implemented through the Python NetworkX package. The colour scale reflects the eigenvector centrality of each country in the network, ranging from magenta for the most central node (USA), through purple for other nodes with high centrality (e.g. EMU, CHN in 2019) and dark blue for less central nodes (e.g. TWN), to cyan for the many peripheral nodes.

In a recent paper (Miranda-Agrippino et al. 2025), we resolve this impasse by proposing a single indicator to consistently measure the PBoC’s monetary policy stance over time, and identify plausibly exogenous monetary policy shocks using a non-linear rule that incorporate its multiple objectives. Our results point to a clear conclusion: the spillovers of Chinese monetary policy are powerful, and the global economy – including the US – is significantly exposed to them. Moreover, it is the dominant role that China plays in global trade and commodity markets rather than its influence on global financial risk-taking that brings them home. The impact of Chinese monetary policy on industrial metals is particularly strong.

The Chinese monetary policy framework: Many targets, evolving instruments

Studying Chinese monetary policy poses unique challenges. The PBoC has operated as China’s central bank since 1983 under a broad mandate to “maintain the stability of the value of the currency, and thereby promote economic growth”. As former governor Yi Gang emphasised, beyond the growth objective, price stability has two dimensions in China: domestic price stability and external stability of the renminbi.

Over time, the instruments used to pursue these objectives have changed profoundly. The PBoC has gradually transitioned away from quantity-based tools and intermediate targets towards a framework that relies more heavily on market interest rates (Chow and Perkins 2014, Fernald et al. 2014, Kim and Chen 2022). As a result, no single observable variable can consistently capture the stance of Chinese monetary policy across periods.

To address this challenge, we construct a composite monetary policy indicator (CMPI) that summarises information from a wide range of price- and quantity-based instruments, building on Xu and Jia (2019). This indicator allows us to track the evolving policy stance of the PBoC in a unified way.

We then embed the CMPI in a non-linear policy rule that reflects China’s institutional setting, extending Chen et al. (2018). The rule explicitly accounts for the government’s annual growth and inflation targets, and incorporates an exchange rate stabilisation objective, consistent with the dual nature of the PBoC’s price mandate.

Deviations from this rule provide a measure of plausibly exogenous Chinese monetary policy shocks, which we use to study their domestic (‘Ins’) and international (‘Outs’) transmission.

The Ins: Domestic transmission and role of the exchange rate

Domestically, Chinese monetary policy operates in ways that will look familiar for the most part (Figure 2). A monetary tightening raises short-term interest rates, compresses credit and money growth, and leads to a persistent contraction in economic activity. Trade responds even more strongly than output: imports fall sharply, and exports decline with a delay, consistent with policy efforts to cushion external demand. Inflation eventually declines as well, albeit with a lag. In the short run, consumer prices may even rise slightly, reflecting the importance of housing-related costs in China’s CPI basket.

Figure 2 Domestic transmission of Chinese monetary policy

Notes: Median IRFs to a contractionary monetary policy shock normalised to yield a contraction of Chinese industrial production of 1% yoy at peak. Shaded areas correspond to 68% and 90% posterior credible sets. Monthly BVAR(6). The green dashed line is obtained by replacing the CFETS EER with the BIS NEER. 2003:01-2019:12.

Where Chinese domestic monetary transmission differs most from advanced economies with flexible exchange rates is in the behaviour of the renminbi. Despite tighter monetary conditions, the trade-weighted renminbi barely moves on impact. Capital does not flow in, as standard open-economy models might predict. Instead, risk-sensitive private capital – particularly banking (other) and portfolio flows – leaves the country as the domestic outlook deteriorates (Figure 3).

Figure 3 Cross-border capital flight

Notes: Median IRFs to a contractionary monetary policy shock normalised to yield a contraction of Chinese industrial production of 1% yoy at peak. Each colour corresponds to a different VAR. Common variables as in Figure 2 excluding CFETS EER. VARs include Global FCI. Shaded areas correspond to 68% and 90% posterior credible sets. Monthly BVAR(6). 2006:01-2019:12.

This private capital outflow is partly offset by official intervention (Figure 4). Foreign exchange reserves decline, as the Chinese authorities move to counter the depreciation pressures on the renminbi due to the marked decline in all private inflows at medium horizons. At the same time, reserve requirement ratios are eased to sterilise interventions and prevent further tightening of domestic monetary conditions.

Figure 4 Asymmetric exchange rate adjustment

Notes: Median IRFs to a contractionary monetary policy shock normalised to yield a contraction of Chinese industrial production of 1% yoy at peak. Each colour corresponds to a different VAR. Common variables as in Figure 2 excluding CFETS EER. Blue IRFs also exclude M2 and TSF. Red IRFs replaces CNY NEER ex USD with USD NEER ex CNY. Shaded areas correspond to 68% and 90% posterior credible sets. Monthly BVAR(6). 2006:01-2019:12.

The combined effect of changes in private and official capital flows is a starkly asymmetric exchange rate adjustment (Figure 3b). The renminbi remains broadly stable and even depreciates slightly against the US dollar, while appreciating gradually against non-dollar currencies in the CFETS basket, against which the PBoC’s currency stability objective is defined. A desire to keep the US dollar value of the renminbi stable while at the same time facilitating the return to the CFETS central parity rate could explain the tendency of the Chinese authorities to let the renminbi appreciate against non-dollar currencies.

The Outs: Global spillovers through real channels

Chinese monetary policy shocks generate large international spillovers (Figure 5). A Chinese monetary tightening triggers a large and persistent contraction in global output excluding China, comparable in magnitude – though delayed – to the domestic effect. Global trade falls sharply, and commodity prices decline significantly and persistently.

Figure 5 Powerful global effects via trade and commodity prices

Notes: Median IRFs to a contractionary monetary policy shock normalised to yield a contraction of Chinese industrial production of 1% yoy at peak. Orange dash-dotted line denote responses of additional variables included in the same VAR. Shaded areas correspond to 68% and 90% posterior credible sets. Monthly BVAR(6). 2002:02-2019:12.

These effects are best understood through China’s central position in global production networks. A slowdown in Chinese activity reduces demand for raw materials and intermediate inputs, disrupting supply chains worldwide. Commodity prices – especially industrial metals – fall with a delay, consistent with real demand effects rather than immediate financial re-pricing (see also Gazzani and Ferriani 2025).

By contrast, global financial spillovers are modest. Global financial conditions tighten only mildly, the US dollar appreciates with a lag, and international capital flows respond slowly. This is consistent with global financial markets responding to the deteriorating global outlook rather than providing an additional amplification mechanism for the international policy spillovers. And it stands in sharp contrast to the global transmission of US monetary policy, which operates also through global financial markets and global risk appetite, as documented in earlier work (Rey 2013, Miranda-Agrippino and Rey 2020, 2022).

Spillovers to the United States

As China’s largest trading partner during our sample, the US is directly exposed to these spillovers (Figure 6). US industrial production contracts following a Chinese monetary tightening, driven by a sharp decline in exports, and to a similar degree as Chinese domestic production. Price spillovers are particularly striking: despite the dollar appreciation, falling commodity prices pass through to US import, producer, and consumer prices.

Figure 6 US trade and price spillovers

Notes: Median IRFs to a contractionary monetary policy shock normalised to yield a contraction of Chinese industrial production of 1% at peak. Orange dash-dotted line denote responses of additional variables included in the same VAR. Shaded areas correspond to 68% and 90% posterior credible sets. Monthly BVAR(6). 2002:02-2019:12.

Conclusions

Chinese monetary policy shocks generate powerful global effects. These spillovers operate primarily through trade, commodities, and production networks rather than through global financial risk-taking. But this balance may change as China continues its integration into the international financial landscape. As China’s largest trading partner during our sample, the US economy does not escape the consequences: its external trade and domestic prices all react strongly to Chinese monetary policy spillovers. As policymakers reassess the structure of global trade and supply chains, our results serve as an essential backdrop to the current political push to reshape the global trade landscape.

Source : VOXeu

GLOBAL BUSINESS AND FINANCE MAGAZINE

Recent Posts

What to do about energy sector reforms when governance incentives are the problem?

In countries where electricity outages and “load shedding” are a regular feature of life and…

1 hour ago

Sector-targeted Skills Development as Industrial Policy

Today’s blog is a background note I prepared for a forthcoming Policy Research Report on…

2 hours ago

How AI and machine learning can predict and explain social risks for more effective development operations

At the time when the Government of South Africa approached the World Bank’s Disaster Risk Financing…

2 hours ago

Middle East institutional investors to increase allocations in private markets

US asset manager Nuveen indicates investors are looking for diversification outside of developed markets. Middle…

2 hours ago

From AI investment to GDP growth: An ecosystem view

Forecasts on the economic impacts of artificial intelligence diverge sharply. This column assesses how the…

2 hours ago

The future is under the glass

Digital design increasingly confers a competitive edge in global tech markets. This column examines how…

2 days ago