This publication is a product of Carnegie China. For more work by Carnegie China, click here.
Last month, the People’s Bank of China (PBoC) released the aggregate finance data for December 2024. Also known as total social financing (TSF), this is a broad measure of credit and liquidity in the economy that includes off-balance sheet forms of financing such as initial public offerings, loans from trust companies, and bond sales. It is the PBoC’s preferred measure of total non-financial debt in the Chinese economy. Although a growing amount of debt is not captured in the TSF data (for example, late payments to businesses and workers by local governments caught in a cash squeeze), most analysts use TSF as a reasonable proxy for the evolution of Chinese debt.
At the end of 2024, outstanding TSF was 408.3 trillion RMB, up 8.3 percent from the end of 2023. With China’s GDP in 2024 closing at 134.9 trillion RMB, up 4.2 percent nominally in 2024, outstanding TSF was equal to 303 percent of China’s 2024 GDP, versus 292 percent of GDP at the end of 2023. This 10-11 percentage-point increase in China’s debt-to-GDP ratio compares to the 8.5 average percentage-point increase over the past eight years.[1]
As the chart above shows, except for a 6-percentage-point decline in 2021 (to make up for the 16-percentage-point COVID-related surge in 2020), China’s debt-to-GDP ratio has risen steadily and is up by 48 percentage points over the past five years. Even this is understated because of a decision in late December 2024 to revise China’s 2023 GDP upwards by 3.37 trillion RMB. Without this revision, China’s debt-to-GDP ratio in 2023 and 2024 would have been 299 percent and 310 percent, respectively.
It’s worth considering the relationship between the increase in Chinese debt and the nominal increase in China’s GDP, for which rising debt has been a major driver. In 2024, it took 30.2 trillion RMB in TSF to generate 5.8 trillion RMB in additional nominal GDP, or 5.52 units of debt for every unit of GDP. The table below shows how many additional units of TSF it took to generate each additional unit of GDP since 2017 (in 2016, the PBoC began adding government bonds to TSF, and so data before this date are not comparable).[2]
As the table indicates, since the pandemic, it has taken roughly twice as much TSF to generate a unit increase in GDP as it did before the pandemic. This gives an idea of just how debt-intensive China’s GDP growth has become. To understand why this is the case, it is worth remembering that GDP growth is equal to the growth in consumption, the growth in investment, and the growth in net exports.
Compared to other countries, China’s distribution of GDP is an extreme outlier. Whereas consumption globally represents 74-76 percent of GDP, and investment 24-26 percent of GDP, China has an extremely unbalanced economy in which consumption accounts for roughly 53 percent of GDP, investment 43 percent of GDP, and net exports are roughly 4 percent of GDP.
The problem lies in the astonishingly high investment share of GDP, towards which most of the Chinese debt expansion is directed. Normally, debt used to fund investment doesn’t result in a rise in the debt to GDP ratio because, over the medium term, the boost to GDP growth from investment should more than make up for the increase in debt.
But this assumes, of course, that the investment is productive. In China’s case over the past 10-15 years, investment in the property sector and in infrastructure seems to have become increasingly unproductive, and while soft-budget constraints have prevented any disciplinary correction in malinvestment, it has inevitably resulted in a surge in debt relative to GDP, hence one of the fastest rising debt-to-GDP ratios in global history.
Since the collapse of the housing sector in 2022, investment in property development has declined rapidly, but rather than bring down total investment growth, Beijing merely replaced it with a surge in non-productive manufacturing capacity, resulting in no discernible improvement in the quality of investment. That’s why the huge shift in the distribution of investment in China has been barely noticeable in the debt figures—this could only be the case if the new areas of investment (manufacturing) were no more productive than the old areas (property development).
What is different, however, is the role of the trade surplus in Chinese GDP growth. In 2023, China ran a trade surplus of $823 billion. This was by far the biggest annual trade surplus for any country in history and amounted in U.S. dollar terms to a hefty 4.6 percent of China’s GDP, or roughly 0.9 percent of the GDP of the rest of the world.
It was surpassed, however, in 2024, when China’s trade surplus grew by an astonishing 20 percent to $992 billion. This amounted to 5.6 percent of China’s GDP, or roughly 1.2 percent of the rest of the world’s GDP. What is more, according to the Mercator Institute for China Studies, “Net exports contributed 30.3 percent to GDP growth in 2024, their highest share since 1997.”
We don’t usually think of the trade surplus when we think about a country’s domestic debt, but in China it plays an important role in determining the total growth in debt. This is because Beijing sets annual GDP growth targets that must be met, and consumption growth and the growth in the trade surplus are forms of growth that are driven mostly by those parts of the economy that operate under hard budget constraints (that is, the private sector).
This makes them very hard to control because they depend on the overall health of domestic and foreign demand. This also implies that it is basically the role of investment under soft budget constraints to fill the gap. Put another way, the growth in China’s debt burden is largely a function of the amount of growth that must be delivered by those sectors of the economy that operate under soft budget constraints, and this amount of growth can be thought of as the gap between the amount of “healthy” growth generated by the hard-budget-constrained parts of the economy and the GDP growth target. Because this gap is mostly filled by non-productive investment, this is the aspect of Chinese growth that mostly drives the surge in China’s debt burden.
Most analysts expect China to set a GDP growth target for 2025 of 5 percent—the same as in 2024. How much of an increase in debt will this require? Beijing has said that this year it is determined that domestic consumption will play a greater role in driving this growth, but there remains skepticism about its ability to pull this off. Without a politically challenging transformation of the distribution of domestic income—away from local governments and businesses and towards the household sector—it is hard to see how Beijing can drive a higher consumption share of GDP without raising either household or government debt.[3]
Unless it can get a significant rise in domestic consumption, in other words, the main change in the share of growth driven by the part of the economy that operates under hard budget constraints is likely to be in the trade surplus. A further rise in China’s 2025 trade surplus will allow China to reduce the amount of non-productive investment it must unleash to achieve the GDP growth target. Conversely, if China’s trade surplus were to contract in 2025, it means that a larger share of China’s 2025 GDP growth must come from non-productive investment and, with that, China’s debt ratios should rise more quickly.
It may seem strange to consider China’s debt burden as partly a function of the trade surplus, but it makes sense once we recognize that Chinese GDP growth can be divided into two components—“healthy” growth under hard budget constraints, versus “unhealthy” growth under soft budget constraints—and that the extent of the latter every year is itself determined by the gap between the former and the politically determined GDP growth target.
To the extent that Beijing sets GDP growth targets that are much higher than the ability of the private sector, or other hard-budget-constrained sectors, to deliver, the only way to achieve these targets is to increase the role of local governments or parts of the private sector that can operate without credit constraints (that is, new growth sectors, along with the property sector before 2021). If these latter sectors were economically productive, this wouldn’t matter for China’s debt burden, but the evidence over the past 10-15 years is that they are not nearly productive enough to justify their cost in debt and resources.
That is why global trade conflict matters so much to China. Its growth model requires a rapid expansion in its share of global manufacturing relative to its share of global GDP, and because domestic consumption growth simply cannot keep pace with this supply-side growth, in order to keep production growing, China must either increase investment, and with it its debt burden, or increase its trade surplus.
If the rest of the world accommodates the latter, Beijing will have more time in which to restructure its economy and rebalance domestic demand. If not, an even faster rise in debt will make the need to transform the Chinese economy all the more urgent for Beijing.
[1] The average increase might seem like 7.6 percentage points from the data shown but, as I explain later, this is distorted by a revision to 2023 GDP that caused a 7-percentage point decline in the debt-to-GDP ratio.
[2] I have ignored the upward revision of GDP in 2023 because it distorts the relationship between 2022’s GDP and 2023’s GDP, and significantly inflates nominal GDP growth in 2023.
[3] Technically, a higher consumption share of GDP can also be driven by a reduction in household savings (the reverse of an increase in household debt), but this requires an exogenous boost in “confidence” that will be hard to manage without a substantial rise and redirection of fiscal spending.
