China has the highest investment share of GDP in the world. It also has among the fastest growing debt burdens in history. These are not unrelated. With growing amounts of investment directed into projects whose economic benefits are less than their economic costs, the surge in China’s debt burden is a direct consequence of this very high investment share.1
What’s more, even as nonproductive investment has led to a faster rise in debt than in GDP, it has also boosted current and past GDP growth to higher levels than China would have otherwise been able to report.2 In theory, Beijing could solve the debt problem associated with high levels of investment by implementing policies that result in a sharp increase in the productivity of investment. It has been proposing this for years, for example by refocusing investment on the technology sector, but it has been unsuccessful so far and, as I explain in a recent paper, it is unlikely ever to be successful. Because China’s imbalances are so deep, the limited number of ways in which they can be resolved all require a significant reduction in China’s reliance on investment to spur growth.
Given the disproportionately large investment share of economic activity, however, any reduction in its share of GDP must put significant downward pressure on growth in overall economic activity. In this essay I summarize a series of mathematical simulations to show how much GDP growth rates must decline as China attempts to rebalance its economy.
Beijing Says It Will Boost Consumption
GDP growth in most large economies is driven primarily by consumption and investment (with net exports usually too small to matter). For this reason, reducing the share of investment in the economy is largely the obverse of raising the share of consumption.
In recent years, an increasing number of policymakers in Beijing have called for such a sustained expansion in the role of consumption (even if they remain vague on how to do so). For example, China’s Central Economic Work Conference in December said policymakers would focus “on boosting domestic demand in 2023 by prioritizing the recovery and expansion of consumption.” In March, during the all-important “Two Sessions” meetings, policymakers called for a greater role for domestic consumption in the Chinese economy and stronger coordination between supply-side structural reform and domestic demand expansion.
But while nearly everyone agrees on the need for a greater role for consumption, there are significant political constraints to rebalancing that have always made it very difficult for countries that have followed an investment-driven development model similar to that of China’s. That’s because after decades during which investment grew faster than GDP, and GDP grew faster than consumption, the relationship between the three must be reversed. For at least the next decade or two, consumption must grow faster than GDP and GDP must grow faster than investment. That is the definition of rebalancing.
But while policymakers and their advisers discuss increasing the household income share of GDP, there is so far very little discussion (at least in public) about the key constraints on such policies. By definition, increasing the household income share of GDP involves reducing the corporate and government income shares. This means that after decades of doing the opposite, rebalancing policies must effectively transfer income, either explicitly or implicitly, from governments and businesses to households. As of yet, however, there have been few concrete proposals about who will pay for these transfers and how they are to be effected.
The Decline in Growth Will Be Unevenly Distributed
Before discussing the simulations and the implications for growth, it is worth noting the political implications. Although much slower GDP growth is probably inevitable for China, the same is not necessarily true for the growth of Chinese consumption or of Chinese household income. China’s rebalancing would require that investment growth declines by much more than GDP growth, which means both consumption growth and household income growth would decline by much less. What may seem like a brutal adjustment for the overall economy does not need to be a brutal adjustment for ordinary Chinese people.
But this also implies a much more rapid decline in the share of income retained by other sectors, mainly local governments. In a nondisruptive adjustment, the brunt of the pain will be borne by sectors of the economy that have benefited disproportionately from Chinese growth in the past three decades—mainly local governments. But as Albert Hirschman noted in his writings nearly five decades ago, these sectors have a disproportionate share of political power and tend to be the main constraints on adjustment. We already saw a preview of this during the presidency of Hu Jintao, with the emergence after 2007 of so-called “vested interests” who opposed policies aimed at economic rebalancing.
The way in which a rebalancing takes place also has implications for the global economy. As China’s GDP growth declines, most of this decline will be driven by an even more rapid decline—or even a contraction—in the growth of investment. The parts of the Chinese and global economies that depend heavily on Chinese investment growth, including metals and industrial commodities, will bear the brunt of a Chinese adjustment. The parts of the Chinese and global economies that depend heavily on Chinese consumption, including agricultural commodities, will be affected much less.
What Is a “Reasonable” Investment Share of GDP?
Some analysts, both inside and outside of China, have argued that because it is an economically and politically difficult process, rebalancing is not truly in the interests of Beijing’s policymakers, and that for institutional, ideological, or cultural reasons (or all three), they will resist rebalancing indefinitely. But to argue over whether or not Beijing intends to rebalance its economy is to misunderstand the Chinese economy. Beijing cannot choose whether or not it will rebalance. It can at best choose the way and pace in which it can do so.
To understand why, we start out with the simple observation that China currently invests around 42–44 percent of its GDP, and it has invested similar or larger shares for decades. This is extraordinary in the annals of economic history. According to the World Bank, investment comprises around 25 percent of global GDP. Not all countries invest the same amount, of course. For more mature, capital-intensive economies, investment can comprise roughly 15–20 percent of GDP. The same is true for many poor, stagnant economies, often those with weak protections for investors’ rights.
On the other hand, for developing economies in the midst of their high-growth stages, such as Mongolia, investment levels tend to be much higher, typically around 30–35 percent of GDP. Unbalanced economies that have traditionally relied on high investment to drive growth, such as South Korea and Taiwan, are also in this range.
China is different. Investment rose from 24 percent of GDP in 1990 to 40 percent by 2002 and to 47 percent by 2010. It then declined to 43 percent by 2019 and has been between 43 and 44 percent during the COVID-19 period.3
This is where simulating the rebalancing process can be so powerful in helping us understand the consequences of China’s much-needed rebalancing. With a few reasonable assumptions we can easily construct a robust model of the Chinese economy and run simulations that suggest what a rebalancing Chinese economy must look like.
The first necessary assumption is to determine what the “appropriate” investment share of GDP would be in China’s case. Should it be 25 percent, the global average, or 20 percent, like that of mature economies, or 30 percent, like that of other rapidly growing, underinvested economies, or some other level?
Some analysts might argue that middle-income countries like China, with per capital investment levels of one-fifth or one-sixth the U.S. level, will be able to maintain very high investment levels as they “catch up” to more advanced economies. They say this in spite of the lack of evidence that poorer economies do indeed regularly “catch up” to richer ones except in extraordinary circumstances.
Their argument assumes away the institutional differences that determine why some countries (or even regions within a large country) are more economically advanced than others. If developing economies have lower investment levels than advanced economies largely because of institutional constraints that limit their abilities to absorb higher levels productively, China is likely already to be excessively invested in property development and infrastructure relative to its ability to absorb more of either, so that further development is more likely to be the result of institutional reform than of further capital deepening.4
This might suggest that, to be sustainable, China should probably bring investment levels closer to the 20 percent of GDP typical of highly-capital-intensive economies. For the purposes of this exercise, however, I will assume a more favorable path for China in which the appropriate goal is to reduce the investment share of GDP to 30 percent—an investment share typical of rapidly developing, underinvested economies in the midst of their growth booms and one that would still make China among the highest-investing major economies in the world.5
I will also assume for the purpose of the simulations that China has ten years to drive down the investment share of GDP to a sustainable 30 percent. This too may be overly optimistic, given how high China’s debt burden already is, but setting these parameters helps define the limits for China’s rebalancing process.
Of course, the investment share of GDP can only decline from 42 percent to 30 percent as some other share of GDP rises and, in principle, any reduction in the investment share of GDP must be primarily matched by an increase in the consumption share. If investment growth declines sharply, either consumption growth must rise just as sharply or GDP growth must fall. Either way, as the investment share of GDP declines, the consumption share will rise.
The Arithmetic of Rebalancing
Once we have these assumptions in place, the arithmetic of rebalancing is pretty straightforward. As investment declines to 30 percent of GDP in ten years, China’s GDP growth rate depends mainly on the pace of consumption growth.
We can model a simple but robust description of the Chinese economy by setting investment at 42 percent of GDP (a little better than the current 43–44 percent); net exports at 4 percent of GDP; and consumption at 54 percent of GDP. In this case, growth in China’s GDP over the ten-year period is just the weighted average growth of investment and consumption (assuming net exports are flat) and so will depend primarily on the assumptions we make for investment growth and consumption growth.
I summarize below the five broad scenarios under which China can rebalance. (The actual numbers are included in Appendix 2.)
A. Rebalance With a Surge in Consumption
One scenario would involve a surge in the rate of consumption growth as it replaces investment growth as the driver of GDP. Given that this has never happened in any country before, this is likely to be a very optimistic scenario.
Consumption in China has recently been growing by 4–5 percent a year, excluding the COVID-19 years, which were terrible for consumption but are unlikely to be repeated. For consumption to surge even as investment and GDP growth drop sharply, an external factor would need to directly or indirectly drive up growth in household income. This might happen if China were, for example, to transfer roughly 1–2 percentage points of GDP every year from local governments and wealthy individuals to ordinary households.
If that happened, it would be possible in theory to grow both household income and consumption by 6–7 percent a year even as investment growth drops sharply. In that case, how quickly would investment and GDP grow as China rebalances its economy within ten years?
It turns out from the simulation that if investment over the ten-year period grows by 0–1 percent a year, China will have rebalanced in ten years. To give my model more flexibility, I have assumed that private sector investment grows as fast as GDP, in which case any contraction in investment must be fully absorbed by the property and real estate sectors.
This would require that investment in the property and real estate sectors declines annually by 2 percent a year. We can play around with the assumptions, but the main point is that if consumption grows annually by 6–7 percent a year and investment grows annually by 0–1 percent a year, China will have rebalanced its economy within ten years. During that time, China’s average annual GDP growth rate would be 4 percent.
B. Rebalance While Maintaining Current Consumption Growth Rates
The problem with the previous scenario is that such large income transfers from local governments to ordinary households are politically difficult to manage without major changes in the country’s economic, financial, and political institutions. This is probably why no high-investment country has ever managed such a transition.
This is also why a second, more realistic rebalancing scenario requires a much sharper slowdown in GDP growth. In this moderately optimistic case, Beijing would implement policies that maintain household income growth and consumption growth at close to pre-adjustment levels—we will assume this means roughly 3–4 percent annually—even as GDP growth drops during the adjustment period. The historical precedents suggest in fact that the required drop in investment and GDP growth would most likely put greater downward pressure on the growth in household income and consumption, but our assumption does provide an upper limit for a moderate adjustment.
Assuming Chinese consumption could grow 3–4 percent a year annually for ten years, how quickly must GDP and investment grow for China to rebalance over that period? It turns out that if investment growth over the ten-year period is a negative 1–2 percent a year, China will have rebalanced. Assuming that private sector investment continues to grow at least as fast as GDP, investment in the property and real estate sectors combined must contract by 3–4 percent annually.
Once again, the point is that if consumption grows annually by 3–4 percent a year and overall investment contracts annually by 1–2 percent a year, China would have rebalanced its economy within ten years. During that time, however, China’s average annual GDP growth rate would have dropped to roughly 1.5 percent.
C. Rebalance With a Sharp Decline in Consumption Growth
The scenarios so far have involved fairly benign assumptions about the evolution of consumption in China: that consumption growth remains broadly at current levels, or higher, while China is adjusting. It is worth noting that most historical precedents suggest that during the very difficult adjustment period, the growth rate in household income, and thus consumption, often declines substantially.
In this third scenario, consumption growth would decline to 1–2 percent. In this case, rebalancing would require an annual decline in investment of nearly 3 percent and GDP growth would be flat. Under these conditions, however, China’s net exports would likely surge, the effect of which would be slightly improved overall growth.
D. Rebalance With a Sharp Contraction in GDP
The fourth scenario involves a very sharp, short-term contraction in GDP in which investment would contract far more sharply than household income and consumption. This, for example, is how the United States rebalanced its economy in the early 1930s and Brazil in the 1980s. But while this form of adjustment can be more economically efficient over the longer term, it is usually a chaotic and politically disruptive process in the short term.
For that reason, I assume Beijing would do all it can to avoid this path. It is very difficult to model this scenario without making all sorts of assumptions about the pace and nature of the initial contraction, but, in any case, it means by definition at least a few years of negative GDP growth. Given China’s closed capital account and powerful regulators, the financial crisis that would lead to such a scenario is very unlikely, and so I don’t bother to model this scenario.
E. Rebalance Over a Much Longer Time Period
It is always possible that China could take much more than ten years to rebalance its economy, as was the case with Japan. The constraint with maintaining excessively high investment rates is that during this period, debt must continue to rise unsustainably. The higher it rises, the greater the risk of a disruptive rebalancing with a sharp, short-term contraction in GDP and the greater the costs associated with the debt.6 For the purposes of this paper, I have run the same simulations over fifteen- and twenty-year periods, assuming that consumption grows by 3–4 percent over these periods and that soaring debt creates no other costs to the economy. In the two time periods, GDP growth over drops to 2 and 2.5 percent, respectively.
China’s Limited Options
Table 1 summarizes the five scenarios. The summaries are not predictions, but they do indicate the kinds of constraints a rebalancing China must face if we assume—reasonably enough—that China cannot retain the highest investment share of GDP forever and must eventually adjust towards a more “normal” share.
| Table 1 | |||
| Consumption growth | Investment growth | GDP growth | |
| Optimistic scenario (A) | 6–7% | 0–1% | 4% |
| Moderate scenario (B) | 3–4% | -1–2% | 1.5% |
| Difficult scenario (C) | 1–2% | -2–3% | 0% |
| Crisis scenario (D) | NA | NA | NA |
| 15- and 20-year scenario (E) | 3–4% | 0–1% | 2.5% |
It’s worth summarizing a few key points:
- China’s only alternative to rebalancing is for it to maintain its extraordinarily high investment share of GDP. Given that this high share is precisely what has driven the very rapid surge in China’s debt burden, continuing this trend will mean an unsustainable increase in the country’s debt burden. My simulations all assume, by the way, that there is no limit to how much debt can grow and that there are no financial distress costs associated with rising debt burden, even though these assumptions are clearly very optimistic.
- There are arithmetically only a limited number of ways China can rebalance its economy. They all require by definition that after three decades in which consumption growth sharply lagged GDP growth, it must now drive GDP growth. A declining investment share of GDP means that GDP growth must outpace investment growth, which in turn means that consumption growth must outpace GDP growth.
- Unless Beijing can somehow manage to set off a surge in consumption growth, which is possible but highly unlikely, it will be very difficult for a rebalancing China to grow at rates much higher than 2 percent. This is because investment in China plays such a disproportionately large role in generating economic activity that any attempt to rein it in will cause significant reductions in the growth of economic activity. We have already seen how this works with Beijing’s belated attempt in 2021 to rein in the property sector.
- China can slow down the adjustment pace to one that is more politically acceptable, but this involves two costs. The first is that a longer period in which debt continues to rise faster than the country’s debt-servicing capacity increases the risk of a disruptive financial adjustment. The second is that, like with Japan, slower growth stretches out over a much longer time period. It is worth noting that in the early 1990s, Japan’s share of global GDP was close to 18 percent. Within twenty years, it was less than 8 percent. This is a nearly unprecedented peacetime decline.
- The slowdown in growth associated with China’s rebalancing will be very unevenly distributed across the various sectors of the economy. If done correctly, ordinary households will not be affected much because the growth in household income can be maintained at levels close to pre-adjustment growth levels. The brunt of the pain will be borne by sectors of the economy that have benefited disproportionately from Chinese growth in the past three decades—mainly local governments.
- China will probably see a surge in consumption in 2023, but this will be driven more by a partial reversal of the terrible contraction in 2022 than by any serious structural adjustment towards a more balanced economy. Until there has been a substantial implicit or explicit redistribution of income from various other sectors of the Chinese economy to the household sector, there can be no sustainable rebalancing of the Chinese economy.
- There is no way to measure how long China can continue along its current unbalanced path, but historical precedents suggest that once the country decides, or is forced, to get its surge in debt under control, the rebalancing process will have begun. Historical precedents also suggest that the longer China waits—and so the greater the consequent buildup in debt—the more difficult the adjustment is likely to be as the financial distress costs associated with the debt must rise.
A rebalancing of the Chinese economy inevitably involves a difficult transition to a very different set of business, financial, and political institutions. But China’s choice is not whether to rebalance from investment to consumption, but rather how to manage the rebalancing from investment to consumption so as to minimize disruption. This means that China’s overall GDP growth rate will depend crucially on the pace of future consumption growth. While it is possible in principle to engineer an acceleration in consumption growth, China’s experience over the past fifteen years (not to mention Japan’s over the past thirty years) shows just how difficult this is.
But it is impossible to keep such a high investment share indefinitely. China must rebalance to get debt under control. Without much, much faster consumption growth, driven by substantial and politically contentious income transfers, the arithmetic is unassailable: the only way to rebalance involves much slower GDP growth.
Appendix 1: China’s Investment-Driven Growth Model
When China began its reform and opening up in the 1980s, its economy—after five decades of anti-Japanese war, civil war, and Maoism—was severely underinvested for its level of social development. What its economy needed above all was significant investment in transportation and other infrastructure, urban property development, and manufacturing facilities, which is why it benefited enormously from the very high investment that propelled growth forward and accommodated the rising needs of Chinese businesses and workers.
But high domestic investment requires either high domestic savings or high imports of savings from abroad (such as through large current account deficits). China, like other countries that followed a similar “Asian” model, chose to fund its rapidly-growing investment—the most rapid in history—by forcing up domestic savings.
By definition, savings is the part of national income (GDP) that isn’t consumed, and so forcing up the share of GDP that is saved is just the obverse of forcing down the share of GDP that is consumed. Because most consumption is household consumption, and the constraint on household consumption is the direct and indirect income households receive, the most effective way to force down the share of GDP that is consumed is to constrain growth in household income relative to GDP. As households receive a declining share of what they produce (GDP), consumption will also become a declining share of GDP.7
As a result of consumption-constraining policies implemented in the past three to four decades, Chinese banks, where most of the savings accumulated, were able to pour funding at very low interest rates into a wide variety of projects. This at first benefited the Chinese economy as productive investment grew at the fastest pace in history.
But, by definition, a good development model is one that addresses and resolves the imbalances that had repressed development. A successful development model, in other words, is one that also makes itself obsolete. Precisely because China’s growth model was extremely successful, it rapidly closed the gap between China’s actual level of investment and the level of investment its businesses and workers could productively absorb given its particular set of institutions and constraints.
The same thing happened to every other country that followed this growth model. But when that happens, by definition the productivity benefits of additional investment must decline. As they decline, a growing share of additional investment begins to generate less economic value than the value of labor and resources employed.
We can see this in China’s debt numbers. When productive investment is funded by debt, the contribution to GDP exceeds the associated increase in debt, and so there is no increase in the country’s debt burden (typically measured as its debt-to-GDP ratio). But once these investments are no longer productive enough to justify the spending, the associated debt will automatically rise faster than the investments’ contribution to GDP.
This eventually happened to every country that has followed this growth model: a period of rapid, sustainable, investment-driven growth with stable debt levels was followed by a period of still rapid but unsustainable growth, driven by a surging debt burden. This started to happen to China about fifteen years ago, as the continuation of China’s high investment levels led necessarily to a rising debt burden.
That is why, one way or another, even if it is possible to switch some of its nonproductive investment into more productive sectors, the investment share of China’s GDP must decline sharply in the next few years. There is simply no way China can continue to maintain what is the highest investment share of GDP in history. The conditions that made this high level of investment sustainable have long ago disappeared, and historical precedents suggest that the sooner the investment share of GDP is reduced to a sustainable level, the better for the long-term health, growth, and stability of the economy.
Appendix 2: Simulations Associated With Each of the First Three Scenarios
The top row for each scenario shows the assumptions associated with that scenario. We show what a straight-line application of these assumptions would look like in years one through ten. The average ten-year GDP growth rate (in the right-most column) is calculated from the numbers generated in the GDP column.
| Scenario 1 | |||||||||
| Average 10-year growth rate | 6.5% | Flat | -2.0% | -2.0% | 4.0% | 0.5% | 4.1% | ||
| Year | Consumption | Trade surplus | Infrastructure investment | Real estate investment | Other investment | Total investment | Investment share of GDP | GDP | GDP growth |
| 0 | 54.0 | 4.0 | 14.0 | 13.0 | 15.0 | 42.0 | 42% | 100.0 | NA |
| 1 | 57.5 | 4.0 | 13.7 | 12.7 | 15.6 | 42.1 | 41% | 103.6 | 3.6% |
| 2 | 61.2 | 4.0 | 13.4 | 12.5 | 16.2 | 42.2 | 39% | 107.4 | 3.7% |
| 3 | 65.2 | 4.0 | 13.2 | 12.2 | 16.9 | 42.3 | 38% | 111.5 | 3.8% |
| 4 | 69.5 | 4.0 | 12.9 | 12.0 | 17.5 | 42.5 | 37% | 115.9 | 4.0% |
| 5 | 74.0 | 4.0 | 12.7 | 11.8 | 18.2 | 42.7 | 35% | 120.6 | 4.1% |
| 6 | 78.8 | 4.0 | 12.4 | 11.5 | 19.0 | 42.9 | 34% | 125.7 | 4.2% |
| 7 | 83.9 | 4.0 | 12.2 | 11.3 | 19.7 | 43.2 | 33% | 131.1 | 4.3% |
| 8 | 89.4 | 4.0 | 11.9 | 11.1 | 20.5 | 43.5 | 32% | 136.9 | 4.4% |
| 9 | 95.2 | 4.0 | 11.7 | 10.8 | 21.3 | 43.9 | 31% | 143.0 | 4.5% |
| 10 | 101.4 | 4.0 | 11.4 | 10.6 | 22.2 | 44.3 | 30% | 149.6 | 4.6% |
| Scenario 2 | |||||||||
| Average 10-year growth rate | 3.5% | Flat | -3.5% | -3.5% | 1.5% | -1.4% | 1.5% | ||
| Year | Consumption | Trade surplus | Infrastructure investment | Real estate investment | Other investment | Total investment | Investment share of GDP | GDP | GDP growth |
| 0 | 54.0 | 4.0 | 14.0 | 13.0 | 15.0 | 42.0 | 42% | 100.0 | NA |
| 1 | 57.5 | 4.0 | 13.7 | 12.7 | 15.6 | 42.1 | 41% | 103.6 | 3.6% |
| 2 | 61.2 | 4.0 | 13.4 | 12.5 | 16.2 | 42.2 | 39% | 107.4 | 3.7% |
| 3 | 65.2 | 4.0 | 13.2 | 12.2 | 16.9 | 42.3 | 38% | 111.5 | 3.8% |
| 4 | 69.5 | 4.0 | 12.9 | 12.0 | 17.5 | 42.5 | 37% | 115.9 | 4.0% |
| 5 | 74.0 | 4.0 | 12.7 | 11.8 | 18.2 | 42.7 | 35% | 120.6 | 4.1% |
| 6 | 78.8 | 4.0 | 12.4 | 11.5 | 19.0 | 42.9 | 34% | 125.7 | 4.2% |
| 7 | 83.9 | 4.0 | 12.2 | 11.3 | 19.7 | 43.2 | 33% | 131.1 | 4.3% |
| 8 | 89.4 | 4.0 | 11.9 | 11.1 | 20.5 | 43.5 | 32% | 136.9 | 4.4% |
| 9 | 95.2 | 4.0 | 11.7 | 10.8 | 21.3 | 43.9 | 31% | 143.0 | 4.5% |
| 10 | 101.4 | 4.0 | 11.4 | 10.6 | 22.2 | 44.3 | 30% | 149.6 | 4.6% |
| Scenario 3 | |||||||||
| Average 10-year growth rate | 1.5% | Flat | -4.5% | -4.5% | 0.0% | -2.7% | 0.1% | ||
| Year | Consumption | Trade surplus | Infrastructure investment | Real estate investment | Other investment | Total investment | Investment share of GDP | GDP | GDP growth |
| 0 | 54.0 | 4.0 | 14.0 | 13.0 | 15.0 | 42.0 | 42% | 100.0 | NA |
| 1 | 54.8 | 4.0 | 13.4 | 12.4 | 15.0 | 40.8 | 41% | 99.6 | -0.4% |
| 2 | 55.6 | 4.0 | 12.8 | 11.9 | 15.0 | 39.6 | 40% | 99.3 | -0.3% |
| 3 | 56.5 | 4.0 | 12.2 | 11.3 | 15.0 | 38.5 | 39% | 99.0 | -0.3% |
| 4 | 57.3 | 4.0 | 11.6 | 10.8 | 15.0 | 37.5 | 38% | 98.8 | -0.2% |
| 5 | 58.2 | 4.0 | 11.1 | 10.3 | 15.0 | 36.4 | 37% | 98.6 | -0.2% |
| 6 | 59.0 | 4.0 | 10.6 | 9.9 | 15.0 | 35.5 | 36% | 98.5 | -0.1% |
| 7 | 59.9 | 4.0 | 10.1 | 9.4 | 15.0 | 34.6 | 35% | 98.5 | 0.0% |
| 8 | 60.8 | 4.0 | 9.7 | 9.0 | 15.0 | 33.7 | 34% | 98.5 | 0.0% |
| 9 | 61.7 | 4.0 | 9.3 | 8.6 | 15.0 | 32.8 | 33% | 98.6 | 0.1% |
| 10 | 62.7 | 4.0 | 8.8 | 8.2 | 15.0 | 32.0 | 32% | 98.7 | 0.1% |
Notes
1 Chinese debt levels are often compared to those of the United States and the European Union (EU), but there are at least two reasons why these comparisons make little sense. First, and most obviously, even the most cursory glance at comparable debt levels globally makes clear that more advanced economies are able to sustain higher debt levels than less advanced economies. Second, and more importantly, most government debt (the largest category) in the United States and the EU is created to fund transfers among different sectors of the economy. A much smaller share of U.S. or EU debt, in other words, is associated with the systematic overvaluation of domestic assets than Chinese debt, and so does not require the writing down of fictitious wealth. In a February 8, 2022, paper, I discussed how excessively high debt levels can affect an economy. In an August 23, 2021, paper, I reintroduced and expanded John Kenneth Galbraith’s concept of the “bezzle,” or fictitious wealth, and explained how it matters to the Chinese economy.
2 Entities that operate under soft-budget constraints—the term used by Hungarian economist János Kornai to describe entities (usually government-related) that can ignore budget constraints—are able to incur losses as they engage in economic activity whose benefits are political and social rather than economic. Because they are able for very long periods of time to carry nonproductive investment on their books at cost, they are effectively capitalizing a portion of investment that, because it did not create value for the economy, should have been expensed. This, in turn, creates both fictitious wealth and overstates GDP that must eventually be reversed.
3 Whether this astonishingly high level of investment is good bad for overall growth depends on underlying circumstances. In Appendix 1, I describe China’s investment-driven growth model, investment and consumption levels during the past twenty years, and the conditions that made the investment-driven growth model so successful in its early decades but obsolete in the past ten to fifteen years.
4 China’s surging debt in the past fifteen years is very strong evidence that China is overinvested, as I explain in Appendix 1. A number of analysts have argued that as long as China’s investment per capita is much lower than that of countries at the capital frontier, such as the United States, it must by definition have room to increase investment, but this assumption is based on confusion over what is an appropriate investment level for each economy. The amount of investment an economy can productively absorb per capita depends on a number of factors, most importantly its set of institutions and constraints that structure political, economic, and social interactions (see Douglass North) and that determine the overall productivity of workers and businesses. In a 2020 paper, I referred to the maximum level of investment that an economy can productively absorb as its “Hirschman level.”
5 The GDP share of investment in Japan, whose experiences in the 1980s and subsequent adjustment are increasingly seen as a potential model for China, was around 37 percent in the 1970s and 33 percent in the 1980s. It declined to 31 percent in the 1990s and to 25 percent over the next two decades, with an average GDP growth rate declining from around 4 percent in the two decades before 1990 to less than 0.5 percent in the three decades since.
6 See note 1.
7 There are many ways to constrain the growth of household income relative to GDP, but they all involve implicit or explicit transfers from the household sector to businesses or government. Undervalued exchange rates, for example, effectively transfer income from households (who are net importers) to manufacturers (who are net exporters). Repressed interest rates and administered credits transfer income from households (who are net lenders into the financial system) to businesses and governments (who are net borrowers). Laws that discriminate against organizing labor, or against labor mobility, put downward pressure on wage growth relative to productivity growth. Consumer goods shortages (as in the Soviet Union) reduce the real value of household income. Excessive spending on transportation and logistical infrastructure that benefit businesses even when their cost to the overall economy exceeds their economic value is a major hidden transfer from households to businesses. Consumption and income taxes can repress the relative growth of household income. Even environmental degradation is a kind of transfer, in which businesses boost profits at the expense of higher healthcare costs for workers. There are many other such transfers.
