China's Consumption Challenge
With consumer confidence collapsing and retail sales barely growing, it has become common to think that consumption in China is suffering from a cyclical crisis that is intensifying the economic difficulties caused by the property slump. The reality is different, however, with consumption in China regaining its pre-Covid trend, a recovery that is rare among the major economies. This bounce back has been made possible by the normalization of the spike in savings seen during the pandemic. With the savings rate now back to the 2019 level, a further rise in consumption is dependent on growth in household income, but that is slowing. The recent burst of cyclical stimulus, by boosting demand for labor and thus stabilizing wages, should provide some respite, but given the bust in the real estate market, what is especially needed are structural policies that raise the non-wage component of household incomes and permanently reduce the savings ratio. Such policies have yet to be seen, even though there are proposals by onshore economists that have the potential to help households while avoiding the sort of welfarism that Xi Jinping has so publicly criticized.
With private consumption in the U.S. accounting for almost 70 percent of GDP and households saving less than 10 cents out of every dollar earned, it is easy to understand why consumer spending is considered a key force in the world’s biggest economy. In China, by contrast, the consumption share is under 40 percent of GDP, and households save fully one-third of their income.[1] As a result, before the Covid lockdowns and the property market slump of recent years, consumer spending was often thought of something of a sideshow in China’s economic story, with much more of the focus instead on investment. However, consumption is currently attracting more attention. This is not because it is fulfilling the hope of many onshore and external economists and filling the hole in aggregate demand left by the deflating real estate market. Rather, with consumer confidence failing to recover from its 2022 collapse and retail sales barely growing, the weakness of consumption is commonly thought to be dragging down the overall economy even further.
So far, this talk of both a cyclical crisis and structural softness in Chinese consumption is only partially correct. Even though the Chinese government has continued to abstain from making the cash payments to the household sector that were so common elsewhere during the pandemic, consumption in China in real terms has actually recovered more strongly than it has in many other economies, and it has now returned to pre-2020 trends. Such a lift is partly because consumption was slowing even before Covid hit. The recovery has also been made possible by a moderation in the surge of the savings ratio that occurred during the pandemic.
However, now that the savings ratio has normalized back to its 2019 level, consumption will be more dependent on income and hence on the labor market. This is a problem because even without a further slowing of the economy, growth in household earnings has been slipping. The government’s macro policy push since September, involving interest rate cuts and local government debt relief, is a reaction to the cyclical economic weaknesses. If such an effort strengthens job creation, a further cyclical downside in consumption might be forestalled. However, without structural policies that permanently reduce the household savings ratio and raise the household share of income, growth in consumption and services is unlikely to strengthen. Such structural reforms are thus far largely missing from the recent policy push. That is despite proposals from some economists that have the potential to help the household sector while avoiding the sort of “welfarism” that Xi Jinping has so openly criticized.
Consumption Cycle
The starting point for much of the discussion about China’s economy is the collapse in the measure of consumer sentiment published by the National Bureau of Statistics (NBS). But other measures of sentiment are not quite so dire. The People’s Bank of China (PBC) publishes a quarterly survey of urban depositors.[2] It shows that consumers do not feel nearly as confident about future income as they once did. But this deterioration is not new; rather, it is a trend that has been in play since the end of the boom that followed the global financial crisis. The measure of employment sentiment in the PBC surveys is weaker, but the Thomson Reuters Ipsos monthly consumer confidence survey for China is not so bad.
Before the pandemic, headline growth in retail sales, one of the main macroeconomic indicators published by the NBS each month, was running at around 8 percent. That has now slowed to zero. However, this headline series is not adjusted for inflation. Removing the effect of prices shows that the growth in volume of retail sales had already slowed to the low single-digits in 2019 – a time when the NBS measure of consumer confidence was suggesting consumer confidence had never been higher. In fact, it was just as consumer confidence began to surge beginning in late 2017 that retail sales started to slow. The specific cause was the fall in car sales. During the period just before the pandemic, it was this weakness in the auto industry, more than the strength of overall confidence, that set the tone for discussions about consumption.[3]
The poor relationship with retail sales suggests consumer sentiment is not a useful gauge of the strength of consumption. Equally, if the aim is to get a sense of the strength of consumer spending, then it probably is also not wise to depend on the retail sales data.[4] The headline retail sales number is basically a measure of spending on goods.[5] Even so, the total value of retail sales is equivalent to 96 percent of the national accounts measure of private consumption. That should be a red flag, given consumption in GDP includes services, and so it should be a much higher figure than retail sales alone.[6]
Importantly, there are other measures of consumption that can be used, although all have shortcomings. The aggregate consumption data in the annual national accounts show that spending rose to 39.2 percent of GDP in 2023, the highest ratio recorded since 2005. There is also a quarterly household survey which, unlike retail sales, is a per capita measure, and it also more fully captures spending on services. As a result, it is a useful and timely proxy for consumption in the national accounts. If we use this as the measure, it suggests consumption in real terms has largely recovered to its pre-Covid trends.[7]
Clearly, this does not look nearly as bleak as the picture painted by the monthly retail sales data.[8] One reason for the gap may be geographical coverage[9], and another is slowing population growth, a phenomenon that slows an aggregate measure of consumption like retail sales relative to the per capita household survey. However, another driver is spending on services, which in the household survey rose from 40 percent of the total in 2013 to 46 percent in 2019.[10] Unsurprisingly, that trend was interrupted by the pandemic, but it has since resumed. The PBC depositor survey points to the same trend. The proportion of respondents intending to spend more on “big ticket items” has now fallen to under 16 percent, down from almost 25 percent in 2016. By contrast, in June 2024 almost half of the depositors surveyed said they planned to increase spending on recreation and tourism, similar to the proportion in the last survey before the pandemic in 2019 and up from less than one-third in mid-2022.
Consumption Structure
This recovery of consumer spending to pre-2019 trends is less impressive when it is remembered that growth in consumption was already slowing before the start of the pandemic. Moreover, it would usually be expected that consumption, which tends to be less volatile than investment, would naturally rise as a proportion of GDP when the overall economic cycle slows. However, despite a truly epic housing bust, the share of consumption in China is still below 40 percent.
This very low consumption share makes China an oddity in a global context. According to the World Bank, the average ratio for upper middle-income economies was 48 percent in 2015–19, while for high income economies it was almost 60 percent. China is such an extreme outlier that some analysts argue that the data cannot be right. Some of their reasons are technical.[11] It is also becoming more common to argue that consumption is underestimated because the usual headline data do not include official spending, when “the Chinese government provides many services that households elsewhere might have to purchase themselves.”[12] The accounting rationale is that in addition to “final consumption expenditures” – akin to the “C” in economics textbooks – a full presentation of a country’s national accounts would include a measure of “actual individual expenditures.” The difference is the in-kind benefits provided by the government, such as education and health – spending that in the headline version of the national accounts would be counted as part of “G.”
It is only since 2018 that the NBS has published details of such in-kind benefits.[13] However, the resultant change is not big enough to end the debate about consumption being too low. The broader measure of “actual individual expenditures” in China in 2022 was still only 44 percent of GDP. Although it might be expected that socialist China would spend a lot more on such in-kind benefits than other governments, in reality the reverse is true. In the EU, for example, the standard measure of consumption is equivalent to 53 percent of GDP. If one were to include in-kind spending, the ratio would rise to 68 percent.
Given the remarkable growth of the Chinese economy during the last forty years, the role that the structural deficiency of consumption has played in constraining economic development has not been obvious. However, economic growth has now slowed, led by the dramatic fall in property market activity, and, as a result, the weakness of consumption is starting to matter more. It is now a common view – even among economists who argue that data problems exaggerate China’s consumption weakness – that the changing economic backdrop means “more must be done through policy changes to support consumer spending.”[14] That, in turn, is because the engines of investment growth are shifting away from manufacturing, infrastructure, and property, with the new big three being manufacturing, infrastructure, and services.[15]
Savings
So, while consumption in China is currently not as cyclically weak as is often portrayed, it is structurally depressed. At the same time, trends in both savings and income give reason to doubt that the relative stability of consumption spending since 2023 can be sustained. Taking savings first, the NBS household survey shows that the propensity to save, after shooting up during the pandemic to 40 percent, has now fallen back to close to the 30 percent that was reported in 2019.[16] In the PBC survey, the proportion of respondents reporting a stronger preference to consume fell by around 5ppts during the first half of 2020 to under 23 percent, but it has since crept back up to a level similar to that prevailing in 2017. These shifts in the savings ratio have supported consumption in two ways. The pandemic spike resulted in households accumulating a stock of “excess” savings, which has since been modestly drawn down. At the same time, the proportion of current income taken for savings has dropped back to the pre-Covid levels, freeing up more earnings for spending.
This moderation in savings is not apparent in the monthly PBC banking sector statistics. Indeed, those statistics show a huge and ongoing rise in household deposits, particularly in the form of longer-term time deposits. But rather than a rise in the savings ratio, this surge reflects a change in the type of savings that households are making. In the PBC urban depositor survey, there has been a big rise in the proportion of people saying they favor savings in the form of bank deposits, offsetting an almost equally big fall in the attractiveness of investments like equities and wealth management products. The same survey shows a fall in the willingness to buy property, a change that can also be seen in the flow of funds accounts. If less money is “saved” in the form of property and riskier investments but the overall savings ratio does not change, then deposits in the banking sector will rise.
While not signaling less consumption, the change in savings behavior has contributed to China’s current economic challenges. Most obviously, by putting savings in banks rather than in property consumers have contributed to the collapse in real estate construction. With the current account balance conceptually equal to the difference between savings and investment, the shift in household behavior has placed upward pressures on China’s external surplus.[17] Finally, the locking-up of savings in longer-term time deposits can be likened to households reducing liquidity in the economy, thus contributing to a decline in inflation and therefore a decline in nominal GDP growth.
Potentially, the negative interaction between real estate and consumption can be further amplified by declining housing prices causing a negative wealth effect that weighs on spending. While the potential for negative wealth effects cannot be dismissed, they are less of a risk in China than elsewhere. Extra spending resulting from the rise in property prices has likely been restricted by the limited availability of mortgage equity withdrawals. In addition, falling house prices do not affect the household sector uniformly; while such a development might dampen the sentiment of existing homeowners, it can also be expected to reduce the need of potential buyers to save so much to buy a property in the first place.[18] A 2018 study by the IMF concludes that while the wealth effect was beginning to become more evident, “elevated housing prices may continue to contain household consumption growth via substantial mortgage or down payment needs.”[19]
That finding was part of a broader study of China’s world-beating household savings ratio, with the IMF describing it as a multi-faceted phenomenon, the product of “demographic changes induced by the one-child policy and transformation of the social safety net and job security that occurred during the transition from a planned to a market economy. Housing reform and rising income inequality also contribute to higher savings.” Clearly, things like the social safety net, let alone the demographics, do not change overnight, so it is not surprising that the overall savings rate since 2023 has only normalized from the spike recorded during Covid-19 instead of continuing to fall further.
Incomes
Without a continuing decline in willingness to save, consumption growth is likely to slow. That is because, in contrast to spending, the household survey shows incomes since 2023 failed to recover to the pre-Covid trends. It is usual to break down household income into four broad “buckets” – wages, earnings from self-employment, “property” income (interest and other proceeds from assets held by households), and transfer payments, including benefits paid by the government. Wages are the biggest component of household income, and if earnings from self-employment are included, the wage-to-GDP ratio in China is not so unlike that in other major economies. Unsurprisingly, given the slowdown in the overall economy, the household survey shows wage growth has slowed since the pandemic. The decline so far has not been as big as the drop in overall economic growth. However, based on the deterioration in employment indicators in business sentiment surveys during 2024, a further decrease seems to be increasingly likely.
Until now, the bigger slowdown has been in growth in other sources of income. With Beijing not repeating the household support policies used elsewhere, transfer payments did not receive the same boost during the pandemic, and growth on a trend basis has been slowing for most of the last ten years. Similarly, while prior to 2019 property income was growing at double-digit rates, it has now dropped to just 2 percent. That is not particularly surprising, given the move in savings away from riskier – and so higher-yielding – financial investments, at the same time as lower inflation is contributing to a decline in interest rates on safer bank deposits. But this drop does matter. In other economies, property income is a much larger share of the total, and so from a relative perspective, the income and spending power of the household sector in China will remain constrained if this source of earnings does not grow.[20]
Policy Hangups
The normalization of the savings ratio and the rising risks of a slowdown in wage growth warn that the cyclical recovery in consumption since 2023 is unlikely to be sustainable. That it is in the services sector where signs of a softening of the labor market are most pronounced is of particular concern; in terms of employment, modern capital-intensive electric vehicle plants cannot replace the shoe factories of old. Indeed, China’s official data show that manufacturing employment in China peaked in 2014, and it has been falling ever since, even though during the same ten years the value of exports has grown by more than 50%.
As a result, it is unfortunate for wage growth that the 14th five-year plan, covering the 2021–25 period, includes an explicit call to keep the manufacturing share of GDP stable.[21] That fits with the existing mindset that manufacturing is the “real economy,” and that investment, because it involves the “accumulation of capabilities,” is superior to consumption that involves an “exhaustion of resources.”[22] It is this kind of thinking about the economy, as well as an incentive structure that encourages local officials to favor investments, that has in turn contributed to the structural weakness in household income and the structural strength of savings.
It is widely believed that Xi Jinping, given his now well-known criticism of the “idleness-breeding trap of welfarism,” has doubled down on this sort of thinking. In reality, the administration’s stance on issues like transfer payments is not nearly so absolute, but the nuance does not make it any more likely that the structural barriers to consumption will be broken down. One broad illustration is China’s spending on welfare, including cash payments to households as well as the in-kind education and health services provided by the government. As a proportion of GDP, it roughly doubled between the global financial crisis and the pandemic. However, at 12 percent of national income, the high of recent years is still modest compared with the 20 percent that is usual in other big emerging market economies and the 30 percent that is normal in the EU and U.S. Moreover, the increases in welfare spending during the years prior to 2020 seem to have since faded.
This September’s decision to raise the pension age offers a more specific example. Requiring people to work longer would seem to be the very antipathy of welfarism, while raising the age at which pension entitlement kicks in is a necessary change to tackle the high savings ratio.[23] If years in retirement are expected to be shorter, it can be hoped that people will be more persuaded that resources – either their own or those from the government – are sufficient to enable them to live comfortably without having to save more. However, the actual change announced in September was timid, with the increase only 3–5 years and phased in over a 15-year period. Even when this is fully implemented, female blue-collar workers will still be retiring at the enviable age of 55. Moreover, based on current trends, by 2035 the longevity of retirees is likely to have risen by around two years. This will mean that even after working five years longer, the youngest pensioners in 2035 will still be thinking that they need sufficient savings to last them through 28 years of retirement, not very different from the situation now.
Further discussion of the pension reform quickly faded in late September, drowned out by the barrage of macroeconomic policies launched by the government aimed at stabilizing GDP growth. These measures included interest rate cuts, support for the equity market, changes in property market policy, bank recapitalization, and CNY6trn in bond issuance to fund the restructuring of local government debt. If these bolster employment and wage growth, they will help prevent consumption growth from slowing further. And if the policy package is powerful enough to raise inflation, there could also be an increase in property income as interest rates rise.[24]
However, while there was some direct help for consumers, with the PBC cutting the cost of mortgages, the government’s approach does not yet seem to be departing from “investment first.” Earlier in the year the government rolled-out CNY300bn in funding for a consumer appliance trade-in scheme, and one-off payments have been made to lower-income households. These measures can help boost spending, but the scale is too small to think they can signal a structural shift in direction. When discussing the trade-in program in relation to the role of fiscal policy in promoting consumption, Minister of Finance Lan Fo’an said it would be effective in “driving investment growth, releasing consumption potential, and promoting industrial development.”[25] Given the existing institutional mindset, it is difficult to dismiss that order as being merely inadvertent.
There are voices calling for the government to adopt policies that directly benefit the household sector. A recent World Bank report investigating the redistributive role of fiscal policy in China finds that “spending on direct transfers amounts to 0.5 percent of GDP, which is the lowest share among several upper-middle-income countries and below the average of some lower-middle-income countries.”[26] As China’s own economic growth has faltered in recent months, prominent domestic economists have called for Beijing to offer broader handouts. Li Daokui, from Tsinghua University, has called for the central government to issue consumption vouchers “in the order of 1 trillion yuan during the National Day Golden Week,” arguing that “every yuan offered in the form of a voucher would generate roughly four yuan in consumption.”[27]
Politically Palatable Consumerism?
It would be optimistic to think that by belatedly adopting the sort of large-scale cash handouts that became common throughout the world in 2020–21, Beijing can in one fell swoop change the structural trajectory of consumption in China. The experiences of Korea and Japan are instructive, where households received big cash handouts in 2020–21, but consumption since then still appears to be weaker than that in China. To shift the trend in spending, deeper structural reforms will be needed.
The data suggest that the two big drivers behind the structural weakness of consumption are low non-wage incomes and the unusual unwillingness of consumers to spend out of income received. If these are the problems, then two sets of policy proposals stand out. One, suggested by Xu Gao, chief economist of Bank of China International, is a “Universal Shareholding Scheme for State-owned Enterprise Stocks.”[28] This proposal aims at resolving a basic contradiction in China’s economic structure – while “China’s massive state capital belongs to all citizens, very little of the returns are transferred to households.”
The idea is to set up a number of investment funds that will become vehicles for SOE ownership, with shares in these funds then distributed evenly across the population. By owning shares in the investment funds, individuals will then have access to the dividends paid by the state sector. This proposal is thus directly aimed at addressing the structural deficiency in non-wage income that the household sector currently receives. While net earnings from deposits are under pressure as interest rates fall, dividends from corporates in China currently generate income for the household sector of just 0.3 percent of GDP. That is tiny in both an absolute sense but also relative to the 6 percent in the UK and U.S., or the 8 percent in Germany.
This proposal is tailored to China’s political circumstances, given that the privatization of state assets is a non-starter. However, giving households access to dividends will only be helpful to household incomes if corporates actually distribute their profits. That, in turn, involves two issues. One is willingness, which in the state sector has been an obstacle going back to the early 2000s. The second is ability, and that is becoming more of a concern as the financial performance of the corporate sector deteriorates.[29] By promoting competition among the investment funds for ownership of the SOEs, Xu’s scheme is designed to reward better corporate performance. However, it is still worth remembering that issues surrounding consumption and income distribution cannot be disentangled from the performance of the wider economy.
The other big area for policy discussion is welfare spending, with calls for the government to spend more in this area now quite common. Together with his call for the government to fund vouchers to give a cyclical boost to consumption, Li Daokui advocates a policy direction that “focuses on the provision of basic social welfare.” In its most recent quarterly update, the Institute of Finance and Banking of the Chinese Academy of Social Sciences proposed that the government issue “special bonds to increase fiscal investment in social welfare sectors, such as pensions, childcare, education, healthcare, and affordable housing.”[30]
Among all the proposals for more generous welfare spending, Liu Shijin’s ideas are perhaps most interesting.[31] He focuses on several inter-related issues. The first is the 300m or so migrant workers who no longer reside in the rural areas but who, while living in the cities, have also not become fully-fledged urban residents. The obstacle in that transition is less in terms of formal hukou restrictions and more due to the continuing inadequacy of welfare provision. The second part of his proposal is based on a distinction between food and other daily necessities on the one hand versus health, education, and other services on the other. Liu argues, not unreasonably, that demand in China for the first group, so-called “survival consumption,” has largely been saturated, but the potential for “developmental consumption” is still rising. The final element in the framework is the inequality that he has been discussing for a number of years because “narrowing the income gap is essential for sustained economic growth.”
These issues come together in Liu’s proposal for a three-year effort “dedicated to equalizing basic services for the 300m farmers-turned-workers moving to the cities.” The goal is to boost developmental consumption and thereby allow services to replace property as an engine of economic growth. But this is not something that can happen without a big change in government fiscal policy. As Liu argues, by its very nature, developmental consumption involves mutual aid and collective efforts, meaning it requires “government support in building systems, funding, and providing basic public services.” Liu maintains that it is this – not more subways and other physical investments – that should be kept in mind when the government announces plans to issue more debt. “From the perspective of expanding domestic demand, if a city has ten subways, it is more beneficial to use that money to solve housing and public service issues for farmers-turned workers rather than to build more subways. Government-subsidized housing would prompt farmers-turned workers to decorate and buy furniture and to bring their families to the city, driving significantly more consumption than additional subways.”
Weaving housing into this proposal is important. Because it is real estate that has been at the epicenter of the weakness in aggregate demand since 2022, the economy is unlikely to find a floor if property remains in freefall. However, stemming the collapse has been difficult, partly because the continued fall in sales has resulted in a glut of unsold units. The authorities have announced some tentative meas ures to tackle this issue, but the Politburo’s call in April to “research policy measures to digest the existing housing stock”[32] is yet to bear any real fruit. A coordinated approach whereby the central government provides more funds to help local governments repurpose excess inventories to provide better quality housing for migrant workers would not only promote the new model of growth that economists like Liu are calling for. It would also help stabilize the old model, and in so doing it would likely go some way in addressing the weakness of consumer sentiment about which many China watchers have been so concerned.
For the government to arrange funding for the absorption of inventories is not beyond the realm of possibility, given that monetized resettlement played a role in the shanty town redevelopment scheme that started in 2014. Implementing the proposals to expand welfare spending would be a much bigger departure for the authorities. But just as Xu Gao’s proposals call for a sharing of the earnings on state assets in a way that does not amount to privatization, so too Liu’s framing of his ideas as collective consumption and investment in human capital rather than as cash handouts and welfarism at least hint at how consumption promotion might be politically palatable in Xi’s China. After all, as the experiences of almost every advanced economy confirm, more spending on health and education, in particular, is unlikely to entail smaller government. But Beijing would still be able to include these in-kind benefits when it calculates versions of the shares of household income and consumption in GDP. Given that the weakness of the social security net has been one factor driving up the household savings ratio, the provision of better health and education has the potential to help bring down precautionary savings.
Conclusion
While it was not such a hot topic when the economy was growing strongly, there has long been a compelling case for reforms in China that tackle the high household savings ratio and that boost the ongoing role of consumption in the economy. With the sudden stalling of the long-running real estate engine of growth, the need for consumption to play an enhanced structural role in the economy and to support aggregate demand is greater than ever. At the same time, the factors that have supported the cyclical recovery of household spending since the pandemic are running out of steam. All told, circumstances seem to be coalescing to challenge Beijing’s “investment first” approach in the direction of economic policy.
But despite the barrage of policy announcements in recent weeks, there is still little evidence of such a shift occurring. Large-scale cash handouts to households were probably always politically impossible in Xi’s China. However, there are not even hints of increased spending on health and education that would probably be the only politically palatable choice of policy support for the household sector. Perhaps it will still be forthcoming. However, it appears that Beijing is finding it tough to shed its “investment first” mindset, even as the collapse of the property market further increases pressures for meaningful structural reforms.
About the Contributor
Paul Cavey is a China-focused Asia economist with twenty-five years of experience covering the region. He now lives in Taipei and runs East Asia Econ, a firm specializing in macro research on China, Japan, Korea, and Taiwan.
Notes
[1] All the data and charts presented here can be viewed and downloaded at https://www.eastasiaecon.com/cn/#charts
[2] http://www.pbc.gov.cn/diaochatongjisi/116219/116227/index.html
[3] Speaking in 2019, Liu Shijin, who until 2015 was deputy head of the State Council’s Development Research Centre, commented that the auto market had “experienced a significant downturn for more than a year, exceeding expectations.” He went on to warn that this weakness could be the result of income inequality, with the earnings of low-income households insufficient to support the continued expansion of the auto market. 刘世锦, 宏观经济走势与新增长动能, November 2019.
[4] This is particularly the case since the pandemic. In addition to the YoY growth rate that serves as the headline measure, the NBS also publishes data for the MoM change, which are regularly revised. In principle, revisions are not unreasonable, but since 2020 they have been huge, and these changes to the MoM numbers are not reflected in corresponding changes in the published YoY rate of growth.
[5] The headline measure of retail sales does include spending on catering, but that is only 10 percent of the total. In recent months the NBS has started to publish a separate measure of retail spending on services, but (so far?) only as one data point showing the year-to-date percentage change, making it difficult to use it in any meaningful way.
[6] As an example, in Taiwan, where the services sector is not nearly as developed as it is in the U.S., retail sales are under 50 percent of total private consumption. Taiwan, like most economies, splits sales of goods into retail and wholesale. Such a breakdown does not exist in China. According to the NBS, retail sales only capture sales “for non-production and non-business purposes,” but it is widely assumed that retail sales include purchases by government departments and perhaps by companies as well. All told, retail sales series are an imperfect measure of consumption in China.
[7] In nominal terms, consumption is trailing the 2013–19 trend. This reflects the slowdown in inflation since the pandemic. Excluding core energy and food, CPI inflation dropped from a run-rate of near 2 percent before 2020 to under 1 percent in 2021–23, and this year it has fallen to below zero.
[8] It also contrasts with the sluggish recovery in consumption experienced in many other economies since 2022. https://www.eastasiaecon.com/china-consumer-nov-24/
[9] The household survey shows the recovery in spending has been driven by consumption in rural areas rather than by consumption in the cities.
[10] The official breakdown of household consumption into goods and services is only available in the annual release of the household survey.
[11] One reason is the measurement of housing consumption. J. Zhang and T. Zhu, “Reestimating China’s Underestimated Consumption,” Comparative Economic Studies 57(2015): 55–74, https://doi.org/10.1057/ces.2014.34 Another is the calculation of services prices. 庄巨忠：如何看待中国的居民最终消费不足, https://mp.weixin.qq.com/s?__biz=MzU1NzcyMjg5OQ==&mid=2247553262&idx=1&sn=901c2902440375c716732c03a53b06d1&chksm=fc332c92cb44a58447cad6373528a84768f6457dca6bab05f6ae11a0314fc53875fd91e953fb&scene=27
[12] Zhang Jun, “Why Is China’s Consumption Rate So Low?” https://www.project-syndicate.org/commentary/china-household-consumption-expenditure-is-probably-higher-than-official-figures-by-zhang-jun-2024-07; The Economist, “How Economists Have Underestimated Chinese Consumption,” October 10, 2023.
[13] The NBS presentation of consumption in the flow of funds isn’t particularly clear. There are two measures, which in English are labelled Household actual final consumption and “Final consumption expenditure”, but both are the same value, differing from “Household Consumption Expenditure” in the usual GDP accounts by the amount of in-kind benefits. The labelling in the Chinese version is a little more distinct, with “Final consumption expenditure” becoming “Actual final consumption expenditure (住户部门非金融交易资金运用实际最终消). https://data.stats.gov.cn/
[14] Zhang Jun, “Why Is China’s Consumption Rate So Low?”
[15] 刘世锦：投资的“三驾马车”格局发生变化，服务业投资的比重已经超过房地产投资,
https://finance.sina.cn/hy/2022-12-14/detail-imxwrcuv7758578.d.html
[16] Using national accounts to calculate the savings ratio on a more internationally comparable basis, it is now around 36 percent, which is again lower than the peak during Covid but still much higher than that in any other major economy.
[17] That might be offset by changes in the savings-investment balance in other sectors, though there is also considerable controversy about just how big China’s overall current account surplus is. Brad Setser, “China’s Imaginary Trade Data,” https://www.cfr.org/blog/chinas-imaginary-trade-data; Liu Yitong and Guo Kai, “How to Understand China’s Trade Surplus Data” blog, October 21st 2024. https://www.cf40.com/en
[18] The drop in property market activity that started in 2021 was triggered by the “three red lines,” which was a policy aimed at controlling the debt of property developers rather than a direct attempt to reduce demand and property prices. However, controlling property prices has been the goal of policymakers for years.
[19] IMF, “China’s High Savings: Drivers, Prospects, and Policies,” https://www.imf.org/en/Publications/WP/Issues/2018/12/11/Chinas-High-Savings-Drivers-Prospects-and-Policies-46437
[20] Given their sensitivity to comparisons with Japan’s economic malaise, policymakers in Beijing should be aware that property incomes are unlikely to be strengthened if bank deposits remain the favored destination for savings while interest rates continually fall. Japanese households benefited from net interest earnings of JPY15tn in 1991, but since 1996, net interest payments have gone to other sectors of the economy. 年次経済財政報告， https://www5.cao.go.jp/j-j/wp/wp-je23/23.html
[21] “中华人民共和国国民经济和社会发展第十四个五年规划和2035年远景目标纲要,”
https://www.gov.cn/xinwen/2021-03/13/content_5592681.htm
[22] Bai Chong-En, “Structural Adjustments in the Chinese Economy,” NUS Goh Keng Swee Lecture on Modern China, https://www.youtube.com/watch?v=uctz6ZDgEnQ
[23] Necessary, but not sufficient. In Korea, where aging is occurring even more rapidly, the savings rate actually increased after 2012, led specifically by the rising savings of households headed by individuals in their 50s and 60s. According to Korea’s central bank, the Bank of Korea, the driver behind this change was the dearth “of reliable sources of income for those in their 50s and over, causing them to become increasingly concerned about their future income.” The implication for China is that later retirement ages and higher pension benefits are needed. “Analysis of Factors Affecting Recent Changes in Propensity to Consume and Their Implications,” BOK Quarterly Bulletin, March 2020, https://www.bok.or.kr/eng/bbs/E0000829/view.do?nttId=10058394&menuNo=400063&pageIndex=1
[24] In theory, the rise in property income could be more than cyclical, with officials close to the PBC arguing that its measures are aimed at promoting “long-term market reform efforts” rather than merely a short-term rise in stock prices. However, during the last thirty years the government has consistently failed to find policies that successfully hitch equity market performance to China’s impressive economic development. “
“交易商协会徐忠谈央行两项支持资本市场发展工具：不是“央行下场入市” 投资者应有风险意识” https://wallstreetcn.com/articles/3729595
[25] “加大财政政策逆周期调节力度、推动经济高质量发展”新闻发布会图文实录,” https://mp.weixin.qq.com/s/ccKOdpw68cHMjRCfHHuAuw
[26] Maria Ana Lugo, et al., “How Redistributive Is Fiscal Policy in China? New Evidence on the Distributional Impacts of Taxes and Spending,” World Bank Policy Research Working Paper, no. 10887, September 2024, http://hdl.handle.net/10986/42116
[27] David Daokui Li, “Comprehensive Deepening of Reforms to Keep China on Track Towards High-quality Economic Development,"
http://www.accept.tsinghua.edu.cn/accepten/2024/0711/c94a6221/page.htm
[28] “推进消费转型需要‘全民国企持股计划,’” https://mp.weixin.qq.com/s/Z6FJwnmY-MCbcRBUY8315Q, with a useful translation at: https://open.substack.com/pub/eastisread/p/xu-gao-why-soe-stocks-should-be-distributed?r=sl6aj&utm_medium=ios
[29] “2024年1–9月份全国规模以上工业企业利润下降3.5%,” https://www.stats.gov.cn/sj/zxfb/202410/t20241027_1957183.html
[30] “2024年第三季度中国宏观金融分析报告,” https://mp.weixin.qq.com/s/pbRbPzmwP7PiNCvmH8Lmmw
[31] “重视和抓好 ‘2+1’关键环节，为稳增长促转型打开新局面,” at: https://mp.weixin.qq.com/s/76_Sip3-NP7eGFYe_MROpQ. There is a full translation here, https://open.substack.com/pub/eastisread/p/part-i-of-liu-shijin-search-for-growth?r=sl6aj&utm_medium=ios
[32] 中共中央政治局召开会议 分析研究当前经济形势和经济工作 中共中央总书记习近平主持会议, https://www.gov.cn/yaowen/2023-04/28/content_5753652.htm
Photo credit: Shwangtianyuan, CC BY-SA 4.0 <https://creativecommons.org/licenses/by-sa/4.0>, via Wikimedia Commons
