Using China’s Central Government Balance Sheet to “Clean up” Local Government Debt Is a Bad Idea

By the end of June 2025, China’s total social financing—the most widely-used measure of total credit in the Chinese economy—stood at RMB 430.2 trillion, up 8.9% year‑on‑year. In comparison, nominal GDP grew just 4.1% over the same period. As a result, China’s total social financing climbed from 303% of GDP at the end of 2024 to 309% six months later. According to the IMF, debt has grown so rapidly in China that the nation accounts for more than half of the increase in the global economy’s debt-to-GDP ratio since 2008. Among major economies, only Japan’s debt-to-GDP ratio is higher, and given that debt-to-GDP ratios tend to rise with income levels, this makes China, as a middle-income economy, even more of an outlier than Japan.

 As high as those figures are, however, the debt in itself is not the problem. Financial crises occur mainly when a contraction on the liability side of a mismatched balance sheet forces a disruptive contraction on the asset side. Because Beijing largely controls the country’s financial system, and because Chinese regulators can restructure liabilities at will, the likelihood of China’s suffering a financial crisis is very low.

 But while Beijing’s ability to prevent banking crises is fairly well-known, the implications are often poorly understood. Many analysts and policymakers believe that the main danger posed by having excessive debt is the risk of financial crisis, and so they believe that the extent to which Beijing’s financial regulators can control this risk means that China’s heavy debt burden is unlikely to become a problem for the economy.

 Why is excessive debt a problem?

 This belief—however widespread—is mistaken.  While the debt itself may not be the fundamental issue, it reflects a deeper problem: the buildup of extensive unrecognized losses across China’s economy, hidden in collective balance sheets. These losses, albeit unrecognized, are real. While they can be overlooked temporarily as long as banks are able to refinance the associated principal and interest without difficulty, they cannot be ignored forever.

 This means that they must be allocated at some future point, usually when the regulators are unwilling, or the system unable, to roll over ever-larger amounts of principal and interest. Once it no longer can, the gap between the debt-servicing costs and the debt-servicing capacity of the assets must be bridged in the form of implicit or explicit transfers from one or another sector of the economy. These transfers can disrupt the economy in at least two important ways. The first, and most obvious, is the direct cost to whichever sector of the economy is forced to absorb the losses—households, businesses, or the government.¹ The second is the economically-disruptive changes in the behavior of households and businesses caused by uncertainty in the allocation of those costs—a process referred to by finance specialists as “financial distress” costs.

 The problem with China’s rising debt burden, in other words, is not the debt itself, but rather the large-scale misallocation of investment funded by the debt. For more than a decade, Beijing has imposed excessively high GDP growth targets on the economy, forcing local governments and state-connected entities to pour money into infrastructure, property, and manufacturing capacity, regardless of the investment’s productivity. Had these been productive investments, the debt funding these investments would have been sustainable, generating sufficient returns to service the debts.

 Boosting GDP growth through bad investment

 But since the late 2000s, a rising share of China’s investment has been non-productive, leading to one of the fastest-growing debt burdens in modern history.² While there are a few economists in China who still dispute that China has engaged in substantial non-productive investments, most now agree that this issue has existed for many years. The surge in the country’s debt-to-GDP ratio itself is very strong evidence of the non-productive nature of a rising share of domestic investment. That is because in a financial system like China’s, where almost all credit expansion is directed into investment and very little into consumption, if investment is productive—i.e. if the resulting increases in productivity exceed the cost of investment—any increase in debt should be more than matched over the medium term by an increase in GDP. Rising debt, in other words, would not result in rising debt-to-GDP ratios.

 This was the case in China until about 2006-08. Prior to then, China’s rapid debt increase was offset by an equally rapid rise in GDP. That’s why even as China’s total debt increased in the 1990s and early 2000s, its debt ratios remained low and stable. It was only after this period that debt growth began to deviate from GDP growth, with the former accelerating even as the latter decelerated. As a result, China’s debt-to-GDP ratio has risen significantly.

 Even China’s this understates the extent of the rise. Typically, losses from non-productive investment should be written down and recorded as an economic expense, which would reduce reported earnings and wealth, and in turn would lower the posted GDP growth rate.³ In fact it was precisely by artificially boosting GDP growth in this way—by investing large amount non-productively and not recognizing the resulting losses—that China was able to meet excessively high GDP growth targets.

 The investments were carried—mainly on the balance sheets of local governments and banks—at cost rather than written down to reflect economic losses, resulting in higher profits (i.e. lower losses) than were otherwise justified, along with assets whose recorded value was higher than their real economic value. In effect, as any accountant would have pointed out, China had “capitalized” an expense: it converted what should have been recognized as losses into fictitious assets. The result was, and is, overstated income, overstated asset values, overstated GDP growth, and, crucially, a mounting stock of debt that cannot be serviced from the returns on these assets.

 The problem, of course, is not unique to China. We have seen this process many times before, most famously in Brazil in the 1970s, and Japan in the 1980s. As the Hungarian economist János Kornai explained decades ago, in economies in which a substantial share of economic activity occurs in sectors that operate under soft-budget constraints—where the state or the financial system ultimately guarantees a politically important entity’s survival through preferential access to credit—losses from poor investments can remain hidden for years. This is unlike in a hard-budget economy, where such losses lead to insolvency and asset write-downs, cleansing the system and reallocating resources to more productive uses.⁴

 However, there are limits as to how far an economy is able to postpone any reckoning while rolling over debt and perpetuating the illusion of solvency. Once Beijing imposed debt constraints on local governments, as it has in the past few years, it has become impossible to sustain this game. Local governments must now find other ways to roll over and manage the debt. For that reason some analysts have proposed that China should federalize its local-government debt by issuing trillions of renminbi in central government bonds to swap for local governments’ hidden liabilities. They argue that such a swap would relieve local governments of debt pressures, lower interest costs, and buy fiscal space to continue to shore up growth with more investment.

 Should Beijing take on the debt?

 But there are at least three problems with this approach. First, transferring local-government debt to the central government in order to open up fiscal space for local governments to continue meeting GDP growth targets simply kicks the can further down the road. It preserves the fiction that the underlying assets remain sound. This means that, rather than resolve the problem of non-productive investment, China chooses to extend it further.

 Second, while lowering the interest rate on local-government debt may create debt relief for local governments, it does nothing to relieve the government’s overall debt burden from a systemic point of view. It “works” only by transferring the cost of servicing the debt from one part of the government to another. That’s because lower interest payments reduce revenues for the banks that made the original loans, and with these banks already under financial strain and in need of recapitalization, the savings on the interest bill represent one-for-one increases in the recapitalization bill. This financial sleight-of-hand is akin to moving money from one pocket to another while declaring yourself richer.

 Third, and perhaps most urgently, it threatens to undermine the central government’s own balance sheet—the only important remaining “clean” balance sheet in China. Once that too is burdened with high levels of debt that are not offset by rising revenues associated with the investment projects, China will find itself with far fewer tools to manage future economic disruptions.

 Japan may be a useful example here. After many years of similar overinvestment and a rapid buildup in debt, Japan started its adjustment process in the early 1990s with a relatively clean central government balance sheet—government debt was roughly 40% of GDP in the early 1990s, according to the World Bank. Government debt grew rapidly thereafter, and while Japan suffered a difficult economic adjustment— its share of global GDP dropped from 17-18% at its peak to 9-10% two decades later—this did not translate into social or political disruption. Japanese household income, in particular, continued to grow during this period, perhaps because of the cushioning effects of the surge in central government debt.

 Overvalued assets must be written down

 The key point here is that the longer China delays in recognizing losses on China’s collective balance sheets, the higher the ultimate cost will become. As losses remain unrecognized, financial distress costs accumulate and the government must increasingly rely on implicit and expect transfers from other parts of the economy to service the debt. The uncertainty of who will ultimately bear the burden will drive households, businesses, and local governments to behave warily, cutting spending and investment and disguising income in ways that further suppress growth. This uncertainty compounds itself, leading to lower confidence and higher economic volatility. History offers numerous examples, from Brazil in the 1970s, to Japan in the 1990s, to the US in the late 2000s, that show that postponing the resolution of bad debts is likely to magnify the problem and increase financial distress costs.

 To resolve China’s debt problem sustainably, Beijing must focus not just on restructuring the liability side of balance sheets but, more importantly, on revaluing the asset side. It must recognize that a significant portion of recorded assets on the balance sheets of local governments, property developers, and parts of the manufacturing sector, are worth much less than their book values implies, with all the consequences that this entails.

 Unresolved losses cannot grow forever. One way or another, they will eventually be written down and allocated to some sector or other of the economy. This will happen either as a matter of policy, with Beijing deciding how different sectors of the economy will absorb the losses, or—more disruptively—as the result of China’s inability to continue rolling over its growing debt burden. If it occurs in the former way, it will be painful, as it will force income revisions, depress reported GDP, and trigger a reversal of the wealth effects that previously fueled excess spending by households and local governments. However, this method of resolving losses will also be controlled, so it will occur in ways that are least likely to be politically or socially disruptive.

 If it occurs in the latter way, however, the costs are likely to be magnified by financial distress effects as each sector of the economy tries to protect itself from absorbing a disproportionate share of the pain. More importantly, it might be much harder for Beijing to control the unravelling.

 The choice is between recognizing losses now as they are allocated deliberately across the economy and letting them metastasize into even larger losses later and perhaps losing control over how these are absorbed by the economy. Given the structure of China’s economy and its capacity for managing social and economic transitions, the central government is the only entity large and resilient enough to absorb these losses without precipitating systemic instability. That is why the central government’s balance sheet should be as clean as possible until Beijing decides—or is forced to by debt pressures—to begin the adjustment process.

 China still has policy space to act, mainly in the form of a strong central-government balance sheet. But that window will not remain open forever. The real challenge for policymakers is not how to hide or refinance debt, but how to acknowledge the scale of fictitious assets and allocate the resulting losses in ways that minimize economic, social and political disruption.