After discussing the troubled rural banks in Henan in my previous post, it is important to note that what happened in Henan was not the first adverse credit event to hit the Chinese financial system. It was just the most recent in the country’s latest string of notable financial events, which can be said to have started back in May 2019 with the intervention in Baoshang Bank. A little over a year later, Baoshang became the first Chinese bank to be shut down since Shantou Commercial Bank closed shop in 2001.
The Baoshang case was followed by interventions or investigations involving several other institutions, including banks, shadow banks, property developers, local governments, homebuyers, and other overextended borrowers. Because these events seem to be happening regularly, and with rising breadth and magnitude, it should be clear that they are not isolated events that can be blamed on the various triggers that set them off. They are far more likely to be part of a systemic problem that has been brewing in China’s economy for more than a decade.
This means, among other things, that even if the property market recovers next year as a consequence of the end of pandemic lockdowns, the recovery can only be partial and temporary. In the medium term, property prices will continue to decline, and insolvencies will keep on emerging. Until the systemic problem is addressed and resolved, there can be no permanent stabilization of China’s property market or of its economy more generally.
At the heart of this credit-deterioration process is the way in which the form and structure of economy activity in China has evolved over the past ten to twenty years. In most countries, GDP is a measure of the output delivered by economic actors over a specified period, whereas in China GDP is an input determined politically at the beginning of a time period. Once China sets its GDP target, local governments (and, until recently, the property sector) have had the responsibility of delivering enough economic activity to bridge the gap between the GDP growth target and what Beijing usually calls “high-quality growth”—that is, the underlying growth rate delivered by the private economy, consisting mainly of consumption, exports, and business investment.
Bridging the gap between the two was not a problem for the Chinese economy during the first thirty years of the period known as reform and opening up (the late 1970s until the late 2000s), mainly because China was seriously underinvested in property, infrastructure, and manufacturing capacity, so the investment that the GDP growth target required was, for the most part, productive. It is probably not a coincidence that during these years China’s GDP growth almost always exceeded the growth target, sometimes by a few percentage points.
This began to change ten to fifteen years ago, by which time China had largely closed the gap between the investment it had and the investment that the economy could productively absorb. When that happened, China should have dramatically lowered the share of production it reinvested, but to do so without causing a sharp drop in the growth of economic activity required rebalancing the economy toward greater consumption, which in turn meant transferring income from previously successful parts of the economy to the household sector.
This has always been the hardest part of rebalancing (as I discuss here). China, like other counties that have followed this model, found itself politically and institutionally unable to manage the transfers. It did, however, keep investment growth rates high and—again, like nearly every other country that has followed this model—China began to overinvest systematically in projects that contributed less to the economy than they cost. The result was a sharp increase in the country’s debt burden: it is only when debt is used to fund nonproductive investment that debt rises faster than a country’s debt-servicing capacity, for which GDP is a proxy.
The Chinese financial system is wholly dominated by banks, and the credit allocation decisions among the banks are effectively determined administratively, largely through various forms of “window guidance” by the People’s Bank of China, the State Council, and/or local-governments. Because of that, the rise in debt was most likely to occur either on the balance sheets of the banks or on those of entities supported by the banks.
This state of affairs also meant that the country’s financial system had to be underpinned by implicit moral hazard. This is because while the rise in debt on the banks’ balance sheets was naturally matched by a rise in the book value of assets, the real economic value of these assets was often far less than the book value, so what looked like healthy balance sheets were in fact often seriously strained and had embedded hidden losses. Banks and bank investors would only accept this condition if they believed that the government would make these losses whole.
By combining excessively high GDP growth targets with an administratively determined credit allocation process (rather than a market-determined one), China consequently ended up with a financial system underpinned by moral hazard. As a result, once the country ran out of easy investment opportunities, it was inevitable that there would be rising stress within the banking system and a rising risk of insolvency.
All of these issues were exacerbated by the way conditions developed in China’s real estate sector, which in many ways was a mirror of infrastructure spending. Like infrastructure spending, the property sector absorbed a sizable share (roughly 25 to 30 percent) of China’s extraordinarily high investment rate (with investment accounting for roughly 40 to 45 percent of Chinese GDP). As such, the property sector is one of the two main engines that has allowed Beijing to achieve its growth targets, but it has required surging amounts of debt, which rapidly have exceeded the value of the associated projects, and this arrangement has depended for many years on an implicit guarantee (in this case, through ever-rising property prices) that allowed developers a free hand to exploit leverage.
Another process is also important for understanding the stresses on the Chinese banking system, namely the way households, businesses, and financial institutions began systematically taking too much financial risk onto their balance sheets. This is a point whose extent and importance seems to elude most economists, although it is one that Hyman Minsky often made.
Any economy usually has a normal distribution of risk-taking behavior among businesses, households, and financial institutions. Normally entities that take too little risk underperform and eventually get squeezed out of the market, while those that take too much risk overperform at first, but then eventually run into trouble during corrections. This is how a well-functioning market effectively polices risk taking.
But economic actors will learn to game anything that an economic system rewards. The problem occurs when an economy undergoes many years—even decades, in China’s case—of rapid growth, expanding liquidity, and soaring real estate and asset prices. Under such circumstances, entities that tend to take on too much risk are never disciplined, and year after year they systematically outperform those that take on more prudent levels of risk.
In such a system, as Minsky explained, over time households, businesses, and banks are all forced to take on the same risky structures if they are not to fall behind their competitors. This creates two types of risks.
When the forms of excessive risk taking are highly correlated across a large number and wide variety of entities, this causes systemic risk, or the risk of a breakdown in the overall financial system, to surge. This is why Minsky considered the correlation among balance sheets to be more important to the overall economy and financial system than the riskiness of individual balance sheets.
Broadly speaking, there are three interrelated components that drive a country’s financial and economic system toward the kind of systematic instability that Minsky warned about. These are financial distortions in economic actors’ operations, warped balance sheets, and an exacerbated wealth effect.
Notably, while some of these distortions are consequences of policy decisions, others are natural and automatic consequences of many years of a particular set of growth conditions. Households and businesses adapt their behavior to their operating environments, so conditions that have been maintained over many years become embedded in their operations and balance sheets.
It may help to take in turn each of the three components (with concrete examples) that drive a country’s financial and economic system toward the kind of systematic instability China is experiencing. To take the first, the real estate sector offers the most easy-to-see signs of how several years of economic and financial expansion distorts the operations and balance sheets of households and businesses.
When property prices seem only to rise year after year, businesses that overinvest in real estate relative to their operational needs outperform and eventually displace those that don’t, while the banks that lend directly or indirectly against real estate tolerate excessively risky loans on the assumption that their risks will be mitigated by continually rising prices. This happened in China. In the 1980s, under Deng Xiaoping’s leadership, China underwent a successful series of liberalizing reforms that unleashed the country’s productive capacity. As capital poured into investment projects, Chinese growth levels soared almost immediately and stayed high throughout the 1980s and 1990s.
As China became more urbanized and as its businesses grew more productive, real estate prices began to rise. This was a natural and necessary development. As long as real estate prices reflect fundamental demand for current use, rising real estate prices perform an important economic function in a market economy. They calibrate demand and supply, thereby allocating land to the businesses that will use it most productively.
But, in China’s case, as real estate prices rose year after year, it changed businesses’ behavior in ways that in turn distorted prices. In any given economy, competing businesses normally include a range of risk appetites from the very prudent to the very adventurous. After a period of rising prices, at some point businesses with higher risk appetites and a more optimistic outlook will begin to anticipate their future real estate needs. On the assumption that property prices will continue to rise indefinitely, they buy more land than they operationally require, effectively speculating on future price increases. (Notably, in an environment of ever-rising prices, this form of speculation is often praised as foresight.)
In such circumstances, the most successful businesses are often not the ones that are most successful at producing and marketing products, but rather the ones that were most aggressive in speculating on real estate. As real estate prices continue to rise, businesses that are willing to take excessive property risk consistently outperform their more prudent rivals, and over time the former begin to displace the latter. Over time, real estate purchasing switches from being largely driven by fundamentals, based on the operational needs of the business at the time of purchase, to at least partly speculative and eventually largely speculative.
This gradual and pro-cyclical transformation of the market probably began to occur in China in the 1990s, a shift that had two important and seemingly contrary effects. First, the boost in demand generated by speculative purchasing caused prices to rise faster than they otherwise would have, thus undermining the economic function of prices in the efficient allocation of land. The pricing mechanism, rather than selecting for the most productive users of land, began to favor those willing to speculate.
At first glance, one might expect the consequences of this trend to be self-correcting. Property in China soared, becoming significantly more expensive relative to GDP compared to property in the United States and Europe. These extreme price distortions should lead to a suboptimal allocation of resources, which in turn should lead to weaker growth, and this in turn should depress expectations and reverse the price distortions.
Over the medium term, however, the tendency to self-correction can be overwhelmed by the second of the two effects. Real estate prices can rise rapidly for years, and as they do, because the mingling of speculative profits and operating profits cannot be easily disentangled, speculative profits inevitably begin to show up as a higher value-added component of GDP growth, as wealth effects boost spending even further. Over time, in other words, GDP growth is artificially boosted by soaring real estate prices and speculative real estate development, which in turn are boosted by high GDP growth expectations, with each element reinforcing the other.
Although I have used the example of real estate here to illustrate this Minskyite process in which long-term market conditions can become embedded in business operations, this self-reinforcing intermingling of operational and speculative activity occurs not just with real estate. If GDP growth rates are high for many years, to take another example, and businesses expect that policymakers will stimulate aggressively to prevent growth from declining, those that use leverage to accumulate market share tend to outperform and eventually displace those that don’t. In such cases, overall business leverage must rise.
If the central bank guarantees liquidity and represses interest rates, to take yet another example, the cost of maintaining high inventory levels declines and businesses adapt by maintaining production levels regardless of changes in demand. If businesses come to expect a stable and undervalued currency, to take still another example, the export sectors effectively bet on currency stability and exports expand relative to imports and domestic demand.
The point is that businesses adapt to the conditions around them, and when a set of extreme conditions remain in place for long periods of time, they begin to incorporate these conditions into their expectations and thus into their operations. They do so not as a policy but because of a sorting mechanism that automatically rewards certain types of risky behavior. In such circumstances, businesses are rewarded for individual behavior that collectively makes the system riskier, and they are penalized with consistent underperformance relative to their peers if they do not participate in this behavior. As a result, businesses collectively build up overreliance on a certain set of conditions in such a way that, when these conditions change, there is much less resilience.
The second of the three interrelated components listed above relates to how households, businesses, and financial institutions design their balance sheets. As they respond to a particular set of assumptions, these assumptions become embedded in how liabilities are structured on their balance sheets.
As more and more entities begin to align their liabilities in the same way, this increases balance-sheet susceptibility to the risk that the assumption will eventually be reversed. Over a long period of expanding liquidity, for example, households, businesses, and financial institutions learn to undervalue liquidity, in which case there is a tendency to reduce borrowing costs by shortening maturities on the liability side of the balance sheet relative to the asset side, or otherwise mismatching the two sides of the entity’s balance sheet. Businesses that fund long-term investment with shorter-term borrowing consistently outperform their more prudent rivals, especially if many years of rapid growth also cause credit spreads to drop. In such cases, riskier borrowers benefit from short-term borrowing not just from lower liquidity premia but also from declining credit spreads.
The problem that especially concerned Minsky—besides the obvious problem that lenders and borrowers are willing to engage in increasingly risky behavior on the assumption that they can ignore certain types of risk—is that as a growing share of the economy’s balance sheets become aligned in this way, the implicit value of liquidity actually rises, in such a way that each degree of mismatch becomes increasingly risky even as the amount of mismatching also rises. In other words, as more and more businesses effectively short liquidity by taking on illiquid balance sheet positions, they reduce the amount of liquidity in the operating part of the economy.
That is why when a liquidity shock does occur, as it eventually must, the consequences can be so painful and unexpected. As households, businesses, and financial institutions rush to convert their risky assets into cash and other liquid instruments, the price of these risky assets tends to fall and the value of liquidity tends to rise. As that happens, the entities can be caught in a squeeze in which asset values and revenues fall even as financing costs rise—until the central bank is forced to step in to absorb the liquidity mismatches of the economy.
There are many other kinds of financial conditions that become incorporated into balance sheets. With the perception of ever-rising prices for real estate or other asset classes, to take another example, banks and other lenders tend to become overly aggressive in lending against collateral. As was the case in China, banks may collude with borrowers to overstate land values in order to expand (seemingly risk-free) collateral-based lending as rapidly as possible.
The point, again, is that this is an automatic process, and it leads to balance sheets that are not only increasingly mismatched but, more importantly, mismatched in the same way across the economy. Households, businesses, and financial institutions don’t simply decide individually at some point to take on excessive levels of risk. They do so gradually and collectively because of operating and financial conditions that encourage them to adopt such behavior and that penalize (with consistent underperformance) those that do not do so.
The third of the three interrelated components that drive a country’s financial and economic system toward the kind of systematic instability that Minsky warned about is the way these conditions create a pro-cyclical wealth effect. I discuss how this works in much greater detail in an August 2021 blog entry, but the key point can be summarized briefly. Long periods of rapid growth are associated with a kind of financial exuberance that almost always leads to overvalued asset prices, usually including real estate prices and often including infrastructure, excess inventory, and other assets.
By making businesses and homeowners feel richer, surging real estate and asset prices can reinforce this rapid growth and financial exuberance by encouraging them to spend more money than they otherwise would have. This tendency is called the wealth effect. As perceptions of rising wealth justify decisions by homeowners and businesses to spend more money than they otherwise would have, the additional spending boosts income elsewhere in the economy, thus reinforcing the rapid growth and financial exuberance.
Unfortunately, the wealth effect works in both directions, although it usually works much more brutally on the way down than on the way up. When financial exuberance is reversed and overvalued asset prices begin to correct, businesses and homeowners begin to see a decline in their recorded wealth, and the negative wealth effect causes them to cut back sharply on spending just as the economy is already slowing. This, needless to say, exacerbates the slowdown, which in turn can cause a wide variety of asset prices to fall even more quickly. Signs of this have already started to appear in China beginning with last year’s clampdown on the property sector.
Four important points stand out. First, to paraphrase Minsky’s mantra that stability is destabilizing, when households, businesses, and financial institutions increasingly embed assumptions about the stability of various parts of the economy in their operations and balance sheets, the very act of doing so increases the riskiness of the financial system and eventually begins to undermine those assumptions.
Second, when instability occurs, people naturally tend to assume that the problem was one of stupidity or fraudulent behavior or even a product of recent policy decisions. In fact, while there is often plenty of unethical behavior—good times, as Walter Bagehot pointed out 150 years ago, breed fraud—stupidity, fraud, and policy changes are not needed to explain the emerging instability. The problem is that over many years most rational players have been incentivized to shift their behavior in response to a set of stabilizing distortions in such a way that any reversal of these distortions can become very painful and costly. The most extreme cases of this behavior are usually the first to fall, thus reinforcing the idea that the instability is caused by individual cases of fraud and stupidity, but in fact the problem is general.
Third, this is almost by definition a systemic problem. In China’s case, three decades of rising property prices, expanding liquidity, moral hazard, and high levels of property and infrastructure investment were economy-wide conditions. It would have been truly surprising if Chinese households, businesses, and financial institutions did not respond to decades of these extreme conditions without incorporating them into their operational and financial assumptions. As Minsky would have explained, it’s not just that lots of individual balance sheets have incorporated too much risk. The underlying issue is that they have incorporated too much risk in a similar manner, so any adjustment or shock affects much of the economy at the same time and in the same way.
Fourth, these are obviously not just China-related problems but rather problems that have affected and will continue to affect many financial systems around the world and throughout history. These risks have been especially prevalent in China over the past few years because of the highly unusual and highly coordinated historical circumstances of Chinese growth, but they are risks that should be understood in principle in a wide variety of contexts.
Among the key lessons from the history of these Minskyite processes is that analysts, even those who have correctly identified the relevant balance sheet distortions, almost always underestimate the adjustment costs when conditions correct. This is probably because they fail to consider the many self-reinforcing processes embedded in the balance sheets and operations of economies that have gone through this process.
This is especially likely to be the case with China. One of the most powerful of these feedback loops, for example, has involved the behavior of local governments, and it has become, not surprisingly, among the most disruptive processes in the current economic malaise. In China’s case, the relevant feedback loop went from rising property prices, to rising property development and rising government revenues, to rising government expenditures on infrastructure and services, to rising growth expectations generated by property development and government spending, and (by virtue of these rising expectations) back to further rising property prices. Reversing any part of this loop risked setting the whole process in reverse in a way that resulted in an opposite self-reinforcing process of declining property prices, slower growth, and reduced government expenditures.
Even the way in which Chinese mortgages evolved represents an example of this feedback process. As I discussed in part one of this two-part blog post, in China, to a much greater extent than in most other economies, it is possible to take out mortgages against property that hasn’t been built yet. This is good for the developers in a rising market. Wang Yongli, a former deputy governor of the Bank of China, explained this well in a recent article translated by Wang Zichen:
Under this arrangement, buyers actually provide a large amount of interest-free funds for the developers (establishing a form of entrusted agent construction), which greatly reduced the capital cost and financial risk of the developers, and provided a source of income for banks (the interest of the mortgages) and government (real-estate related revenue).
As long as it is assumed that property prices can only rise, taking out mortgages against nonexistent homes makes sense for everyone. The problem, Wang explains, is that this arrangement creates incentives for property builders to engage in behavior that is highly pro-cyclical, and these incentives become stronger just as conditions change:
As a result, developers will take all possible measures to expand the pre-sale of apartments. They then use the money to expand their business, such as purchasing more land, slowing down the construction of the apartments that have been paid for to take more advantage of the interest-free funds, or even cutting corners on the building quality and their supporting facilities. Once the developers cannot deliver the apartments on time or with good quality according to the contract, the scattered and (structurally) weak buyers can hardly protect their own rights and interests.
What was good for developers in a rising market turned out to be (not surprisingly) bad for developers in a falling market. As homebuyers became increasingly worried about the risks associated with pre-sales, this triggered behavior that cut off funding for property developers at the worst possible time, and this behavior in turn exacerbated declining prices as well as liquidity and solvency problems among property developers.
The pre-sales crisis may have been an idiosyncratic Chinese type of trigger, but again it is important to stress that, while the specific forms in which the various feedback loops are triggered may differ in unpredictable ways, they nonetheless must emerge for systemic reasons during a long period of surging property pieces and expanding liquidity. It is easy to confuse triggers with causes, but if these feedback loops don’t emerge in one form, they will emerge in another. Analysts will spend a lot of time and effort discussing the sources and consequences of specific triggers and what might have been done to divert them, but these triggers are simply symptoms of an underlying systemic problem. They are not the underlying problem.
There are many other similar feedback processes affecting the Chinese property sector and, more generally, the economy, including the high share of household savings held in property, the pro-cyclical nature of sectors that account for disproportionately large shares of the economy (like the property and infrastructure sectors), the systemic impact of fictitious wealth, extensive moral hazard, and so on. Because these feedback loops emerge in ways that reinforce growth during periods of expansion, this often makes it hard to tell the difference between the underlying growth processes driving the economy and the additional growth created mainly by self-reinforcing processes embedded in balance sheets and business operations.
One result of this confusion is that before the reversal is triggered, analysts and policymakers often assume the underlying growth dynamics of the economy are greater than they really are. This often leads them to overestimate the sustainable growth rate of the economy during the expansion period.
Chinese authorities have fortunately racked up a great deal of experience in managing the spread of financial contagion. They understood the contagion risks and reacted quickly to create the necessary liquidity to prevent things from spiraling out of control. A July 2021 Bloomberg article listed some of their responses:
The China Banking and Insurance Regulatory Commission . . . issued guidance in response to the boycotts aiming to expedite the delivery of homes to buyers, a newspaper published by the commission reported Sunday, citing an unidentified senior official at the agency. China is responding to protests that flared up at 100 housing projects across 50 cities, threatening to spread the real estate crisis to the banking system. Regulators met with banks last week to discuss the boycotts, while state media cited analysts warning that the stability of the financial system could be hurt if more homebuyers follow suit.
The city of Zhengzhou responded by announcing the establishment of a bailout fund to take over and complete these unfinished projects. According to a recent South China Morning Post article:
Henan’s local authorities assigned a bad-loans manager and a state-owned real estate developer to clean up the province’s property mess, taking drastic action to contain a crisis ahead of China’s twice-a-decade leadership conclave. A working team set up by Henan Asset Management Company and Zhengzhou Real Estate Group will help cash-starved developers to work out their funding woes, according to a report posted on the asset management firm’s website. The team will also aim to revive stalled projects, sell assets and restructure businesses to ensure the completion and smooth delivery of homes to contracted buyers, the report added.
In addition, the authorities forced some of the weaker banks to shore up their capital bases. To quote another Bloomberg article:
The central government will allow 320 billion yuan generated from the sale of special local bonds to be used to top up the capital of medium- and small-sized banks, the Financial News reported last week, citing an unnamed official with the China Banking and Insurance Regulatory Commission. The amount, including 120 billion yuan in unused funds from last year, is 60% higher than in 2020 when money from the sale of these bonds was first allowed to be used for that purpose.
On July 28, 2021, the Politburo, China’s top policy-making body, “pledged to stabilize the property market,” according to the Wall Street Journal. The Politburo “said it would work to resolve problems in the rural banking system but it said local governments should take direct responsibility for delivering unfinished homes and supporting demand for housing.” The article continues:
Local governments have rolled out a flurry of incentives in recent weeks to boost their property markets, including tax rebates, cash rewards and lower down payments. Zhangshu, a city of roughly 500,000 people in the eastern Chinese province of Jiangxi, is offering the equivalent of about $150 to property brokers who can make a sale. Guyuan, a city of 1.5 million in China’s northwest, has offered to subsidize 1% of purchases for first-time home buyers. In Chizhou, in China’s rural interior, potential home buyers can attend a government-run real estate fair, buy a home, receive 3,000 yuan of subsidy worth about $450, have their down payment reduced to 20% from 30% and have some of their property management fees waived.
Meanwhile, the People’s Bank of China announced that it would issue 200 billion renminbi of low-interest loans for banks to supplement with their own funding to help “fill the funding gap needed to complete unfinished projects.” Other steps have been taken, too, and there are likely to be more moves announced over the next few days and weeks.
These rapid reactions to stem the spread of financial contagion show that Beijing clearly recognizes that the problem is systemic, not specific to a few badly managed banks. This view was reinforced when, soon thereafter, Liang Tao, vice chairman of the China Banking and Insurance Regulatory Commission, reportedly warned of “high hidden risks” in the shadow finance sector, “as some products have complex structure and high leverage levels.” According to Reuters, he went on to insist on vigilance toward “a rebound of shadow banking risks as some institutions may use improper financial innovations to create new variants of shadow banking.”
But while Beijing reacted quickly to these events, it is nonetheless important to recognize that the various solutions proposed mainly involve resolving liquidity concerns. They do not address the more fundamental solvency concerns. Extending or guaranteeing liabilities simply extends the period during which insolvency can be resolved, whereas resolving insolvency, or other forms of bad debt, means above all allocating the losses to one or another specific sector of the economy.
That is not what has happened in China. The various solutions and proposed solutions have merely transferred the problem from the local banks onto either local governments’ balance sheets or—to the extent that smaller bad banks are to be merged with larger, healthier banks—onto the balance sheets of larger banks. The regulators, in other words, are addressing what is basically an insolvency problem by extending or insuring liabilities as if it were a liquidity problem.
That doesn’t mean that Beijing’s efforts will have no impact. Because financial breakdowns and contagion are caused by mismatches in balance sheets, rather than by negative equity, treating the events as if they mainly reflect a liquidity problem can be an effective way of preventing or limiting financial contagion. By restructuring liabilities—namely, by forcing state-affiliated lenders to replace disappearing deposits and customer liabilities—and by providing as much liquidity as necessary, the regulators can effectively slow down the liquidation process and prevent banking panics.
But these measures do not and will not solve the underlying problems. These problems include the following:
Under these conditions, it is hard to predict what will happen over the short term. My best guess is that because of the Chinese Communist Party’s upcoming National Congress, Chinese regulators will prioritize stability over all other things. Any additional outbreaks within the financial system will quickly be absorbed by local government borrowing and the bigger banks. Once the pandemic lockdowns end, the regulators may even unleash a series of “bazooka-like” policies to try to reverse [investor?] sentiments, kick-start rising property prices, and boost confidence among creditors and depositors.
But even if this works temporarily, it cannot go on forever. At some point, Beijing must take concrete steps to allocate the costs of rising insolvency, most likely to local governments, although this will not be easy. If it doesn’t do so, eventually the debt levels will become unsustainable and, perhaps with an evaporation of credibility, the system will force its own allocation of costs.
The latter scenario is usually referred to as a debt crisis, and while I think a debt crisis in China is still very unlikely, this is because I think Beijing still has the ability to restructure liabilities. But one way or another, losses must (and will) be forced onto some sector of the economy, either explicitly as a political decision or implicitly as the economy adjusts to allocate the losses to those sectors least able to protect themselves.
This leads to the last part of this blog post, which is tangential to the discussion above but nonetheless relevant. Resolving China’s debt burden and rebalancing the sources of demand in the Chinese economy may have an important impact on the distribution of power within the Chinese government. More specifically, this process may exacerbate and intensify a conflict between Beijing and provincial and municipal government officials.
We must start by recognizing that the economic costs that must eventually be allocated and absorbed are quite substantial. The costs of resolving China’s bad debt are large enough, as I discuss above, but there is also the cost of rebalancing demand within the economy. China invests roughly 20 to 30 percentage points of GDP annually in the property and infrastructure sectors of the economy (while total investment is roughly 40 to 45 percent of GDP and property and infrastructure each account for roughly one-third of total investment).
This is far too high a share of GDP to be sustainable and must (and eventually will) come down sharply. But supply and demand must balance, and for large economies that cannot count on ever-growing trade surpluses, a reduction in the investment share of GDP is just the obverse of an increase in the consumption share.
This has an important implication for the distribution of income. If the domestic share of Chinese consumption is to become an important enough driver of growth to accommodate a sharp reduction in the investment share, Chinese households will directly or indirectly have to retain a share of GDP that is at least 10 to 15 percentage points greater than their current share and, conversely, some other sector or sectors must suffer a 10 to 15 percentage point reduction.
There are various ways in which this could happen, many of them very painful and value-destroying, but the way that would be least damaging to the economy involves implicit or explicit transfers from one of the nonhousehold sectors of the economy to the household sector. Beijing is unlikely to force the bulk of the rebalancing costs onto the business sector (through measures like higher taxes, higher wages, or lower subsidies, for example) because doing so would destroy the main engine of healthy and sustainable economic growth. Given its desire to reduce its dependence on foreigners for agricultural and industrial commodities, it is also unlikely to force the bulk of rebalancing costs onto the agriculture and mining sectors.
That leaves the government as the only sector that can ultimately bear these costs. But which level of government—local governments or the central government? Given its centralizing tendencies, my assumption is that Beijing will try to force local governments to absorb the bulk of the adjustment costs, which means ultimately stripping them of a substantial share of their revenue sources and, perhaps more importantly, their assets.
The required transfer is large enough that it will almost certainly result in a major redistribution of political power, making this decision among the most important and contentious political decisions China is likely to face for many years. While this process hasn’t started in earnest yet, it seems to be the logical culmination of the way Beijing has handled the limited adjustments China has already made toward a new growth model. National policymakers have already come down hard on the property sector, which was a major source of revenue for local governments, and while it now seems that it underestimated the economic impact of its attempts to slow down the out-of-control property sector, it is pretty clear that it had long wanted to do this anyway.
Beijing’s next step is to control the ability of local governments to raise large and risky amounts of debt. In that light, it is helpful to refer to what Adam Y. Liu, Jean C. Oi, and Yi Zhang referred to as the “grand bargain” of the mid 1990s. They describe it like this:
While much scholarly attention has been paid to the consequences of the 1994 reform that left localities with a tremendous fiscal gap, our findings show that Beijing in fact gave localities the green light to create new backdoor financing institutions that counteracted the impact of fiscal recentralization. In essence, these institutions were the quid pro quo offered to localities to sustain their incentive for local state-led growth after 1994. The bargain worked, and growth continued. The drawback, however, was that China’s economic growth has been accompanied by the accumulation of local government debt with little transparency and central control.
It seems that the conflict in the mid-1990s between Beijing and local government officials was postponed, not resolved, by this “grand bargain” and that this conflict may be reemerging as Beijing tries to bring local government debt under control. The spending of local governments seems to be rising, especially as local governments are still responsible for generating the additional growth needed to bridge the large gap between what Beijing calls high-quality growth (the sustainable, healthy growth generated primarily by consumption, exports, and business investment) and the country’s politically determined GDP growth targets.
If, at the same time, the revenues of local governments are permanently reduced, local governments will need to avail themselves of even more debt financing to bridge the gap between rising spending and declining revenues. But this won’t be easy. Because Beijing is determined to regain control of such borrowing, it is limiting the borrowing ability of local governments.
The only way to square that circle, I would argue, is ultimately for local governments to liquidate or otherwise deploy their extensive ownership of real estate and state-owned enterprises to fund what in effect must be a major transfer of income and wealth, both to resolve bad debt and to increase the household sector’s share of GDP. This will inevitably be a difficult process and might prove politically disruptive, but one way or another I expect it to be at the heart of China’s adjustment process over the next several years.
Aside from this blog, I write a monthly newsletter that focuses on global imbalances and the Chinese economy. Those who would like a subscription to the newsletter should write to me at ch***********@ya***.com, stating their affiliations. My Twitter handle is @michaelxpettis.
Carnegie does not take institutional positions on public policy issues; the views represented herein are those of the author(s) and do not necessarily reflect the views of Carnegie, its staff, or its trustees.
Sign up to receive China Financial Markets in your inbox!
Check your email for details on your request.
1779 Massachusetts Avenue NW
Washington, DC 20036-2103
Phone: 202 483 7600
Fax: 202 483 1840
In a complex, changing, and increasingly contested world, the Carnegie Endowment generates strategic ideas and independent analysis, supports diplomacy, and trains the next generation of international scholar-practitioners to help countries and institutions take on the most difficult global problems and safeguard peace.
© 2022 Carnegie Endowment for International Peace. All rights reserved.
By using this website, you agree to our cookie policy.
You are leaving the website for the Carnegie-Tsinghua Center for Global Policy and entering a website for another of Carnegie’s global centers.
你将离开清华—卡内基中心网站，进入卡内基其他全球中心的网站。
About Author
