Yves here. Time again for a speech I wind up giving much too often on the cognitive bias, halo effect, and how discussions of China’s economy are a case study. The halo effect is the propensity to see individuals or groups as all good or all bad. So China fans, who admittedly often face China critics with a dated or skewed view, are too eager to dismiss bona fide problems China faces, such as its still-unwinding monster real estate bubble.
For instance, we were often attacked for saying that China had an overinvestment/overcapacity problem that was bad for China as well as the global economy. Such a thing could not possibly happen, or if it did, it could not be bad! Then the Chinese government admitted, as it was also going into deflation (which is particularly stressful for country like China with a very high level of private debt) that it did have crippling overcapacity in several large manufacturing sectors (which it cutely branded as “involution”) and is trying to rationalize those fields (which is not easy for reasons explained long form elsewhere).
The US is clearly on a much worse footing than China. But that does not mean that China is not facing some strong headwinds and fundamental challenges.
By Kenneth Rogoff, Thomas D. Cabot Professor of Public Policy and Professor of Economics, Harvard University and Yuanchen Yang, Economist International Monetary Fund. Originally published at VoxEU
A growing debate has emerged over whether China risks repeating elements of Japan’s post-real estate bubble stagnation of the 1990s. This column compares China’s current real estate adjustment with Japan’s experience and uncovers striking parallels in investment dynamics and consumption responses. The key lesson from both episodes is that overinvestment during a housing boom simply cannot be unwound quickly. Excess supply hangs over the economy, discouraging new investment and weighing on activity long after volumes and prices peak. China still has room to shape the outcome of its adjustment, but the window narrows as overcapacity, weak consumption, and negative sentiment reinforce one another.
China’s prolonged real estate downturn has become one of the central policy concerns in global macroeconomics. (e.g. Copestake et al. 2026). With residential investment still contracting, house prices drifting down across much of the country (Figure 1), and consumer confidence stubbornly weak, a growing debate has emerged over whether China risks repeating elements of Japan’s post-bubble stagnation in the 1990s.
Figure 1 House price decline by city in China
Cumulative monthly decline from peak (peak = 1)
Sources: National Bureau of Statistics of China and authors’ calculationsNotes: The price index is constructed using the National Bureau of Statistics’ monthly residential property resale price index for 70 large and medium-sized Chinese cities. For each city, the index is normalized to its historical peak (the series begins in 2011, with peaks occurring between 2017 and 2023 depending on the city). All other monthly observations are then expressed relative to this benchmark.
This turn of events has come as quite a surprise to many observers. True, some had questioned whether the extraordinary rate of appreciation of Chinese real estate was sustainable and would lead to major adjustment problems (see Rogoff 2015 on how the debt super-cycle would inevitably come to China and be quite severe). However, for years the conventional belief was that any real estate adjustment in China was still far off in the future (e.g. Glaeser et al. 2017) and, even if came, would be much less crippling than Western-style real estate crashes. China’s homeowners are far less leveraged than in the US and, to the extent there were any problems, China’s powerful central government would be able to adjudicate and clean up any defaults far faster and more efficiently than the Western legal systems, eliminating the long-drawn-out uncertainty that makes debt crises so debilitating.
Now, however, as China’s crisis goes into its sixth year, the view that real estate crises will generally be muted and short-lived unless amplified by a banking crisis (Bernanke 1983, Schularick et al. 2014) appears overstated, with other amplification mechanisms also being important. Rognlie et al. (2018), for example, emphasised that housing overbuilding can generate long-lived demand shortfalls even without a classic financial crisis, while Rogoff and Yang (2020) showed that overbuilding in the case of China had reached breathtaking proportions nationwide.
In our latest research (Rogoff and Yang 2026), we take up this issue by comparing China’s current real estate adjustment with Japan’s experience in the 1990s, basing our analysis on exploring relative growth across different cities (almost 300 in the case of China) or prefectures (47 in the case of Japan). In both cases, regions with large overbuilding tended to suffer a severe downturn. The comparison is imperfect – China today is poorer, less financially liberalised, and operates under a very different political system. Nevertheless, despite profound institutional differences, we uncover striking parallels in investment dynamics and consumption responses in these two countries.
Real Estate as a Growth Engine – and Its Limits
For decades, real estate played a pivotal role in China’s growth miracle. Including upstream and downstream linkages and related infrastructure, real estate-related activity accounted for close to one-third of aggregate demand at its peak (Rogoff and Yang 2021). Massive investment helped support rapid urbanisation and boost local government revenues through land sales.
Yet size itself became the problem. By the late 2010s, China’s per capita housing stock had reached levels comparable to far richer economies. In many smaller and less diversified ‘tier 3’ cities, which are oftentimes losing population, construction continued even as effective demand weakened. Returns to new investment declined steadily (Rogoff and Yang 2024a, 2024b).
Japan experienced a remarkably similar dynamic in the 1980s. Residential and commercial construction surged alongside infrastructure spending, pushing land and house prices to extraordinary levels. When the bubble burst in the early 1990s, real estate investment remained subdued for an extended period, and in fact never returned to its pre-crisis level (Figure 2).
Figure 2 Real estate investment as a percentage of GDP
Sources: National Bureau of Statistics of China, Cabinet Office of Japan, and authors’ calculations.
The key lesson from both episodes is structural. Housing is highly durable, and overinvestment during the boom simply cannot be unwound quickly. Excess supply hangs over the economy, discouraging new investment and weighing on activity long after volumes and prices peak.
Declining Returns and the Investment Overhang
Using city-level data for China and prefecture-level data for Japan, we document a common pattern: regions that built more aggressively during the boom experienced deeper and more prolonged slowdowns afterward. In both countries, the contribution of real estate investment to growth deteriorated over time and eventually turned negative.
Importantly, this decline began before headline crises. In China, returns to real estate investment were already falling well before the massive 2021 Evergrande default and the tightening associated with the ‘three red lines’ which was intended to deleverage the sector. Again, the strength of this effect shows up significantly in inter-city comparisons. This suggests the downturn was not primarily a regulatory miscue, but the result of long-standing structural imbalances.
Japan’s experience provides perspective on duration. There, the negative growth impact of real estate investment persisted for more than a decade after the bubble burst. Even as GDP eventually stabilised, the real estate sector continued to weigh on recovery through the late 1990s and early 2000s.
This implies that simply stabilising prices or restoring credit flows is unlikely to revive growth quickly. When the problem is too much rather than too little capital, stimulus risks diminishing returns.
Wealth Effects: The Underappreciated Channel
Cumulative investment, often exceeding underlying demand, gradually builds up excess supply and places sustained downward pressure on prices, especially in cities characterised by significant investment overhang. In both countries, falling house prices have powerful wealth effects. In Japan, declining land prices devastated household balance sheets, consumption weakened sharply even in regions where credit did not collapse.
China faces an even stronger version of this mechanism. Roughly 70% of Chinese household wealth is held in housing, far higher than in advanced economies, while consumption accounts for only around 40% of GDP (Figure 3). When house prices fall, households respond by cutting spending, because they feel poorer and because housing serves as a form of precautionary saving in a system with limited social insurance.
Figure 3 Valuation of different asset classes, 2017 (trillions of dollars)
Sources: World Bank, BIS, National Bureau of Statistics of China, Bank of Japan, FRED, Zillow, and authors’ calculations
Our estimates suggest that a large nationwide house price correction could reduce aggregate consumption by 2-4 percentage points of GDP, much more than announced consumption-support policy measures are designed to offset.
Sentiment: Amplification Beyond Wealth Effects
Beyond wealth effects, we find that sentiment plays a crucial amplifying role. When households expect prices to continue falling, they delay purchases, raise precautionary savings, and exacerbate the slowdown. This mechanism, stressed in earlier work on expectations and ‘animal spirits’ (e.g. Soo 2018), was also visible in Japan, where pessimism about land prices endured long after the initial crash.
In China’s case, sentiment appears particularly important. Using news-based measures of housing market tone, we find that pessimistic sentiment substantially magnifies the impact of price declines on consumption. Once households internalise the belief that housing is no longer a safe store of value, the feedback loop becomes self-reinforcing.
Why China Is Similar, and Why It Is Different
Despite these parallels, China is not Japan. The structure of leverage differs: Japan’s crisis centred on private banks and corporations, whereas China’s vulnerabilities are more concentrated in local governments and state-linked entities. China also retains greater administrative capacity to postpone loss recognition and prevent outright financial sector collapse.
China also has other advantages that Japan lacked. Productivity growth remains stronger, and China sits near the global frontier in several fast-growing sectors, including electric vehicles, renewable energy, and more recently, artificial intelligence. These factors may help prevent a full-scale replication of Japan’s stagnation.
But the differences cut both ways. China is ageing faster than Japan did in the 1990s, and as a still-developing country, it lacks Japan’s extensive social safety net. Moreover, newer growth engines, no matter how dynamic, are still small relative to real estate and infrastructure. Rapidly shifting from one export-led boom to another is unlikely to substitute fully for a domestic demand shortfall in an economy of China’s size.
Looking Beyond Short-Term Stabilisation
The central lesson from Japan, and now China, is that avoiding a banking crisis is not enough. When a growth model built around investment, whether in real estate or infrastructure or emerging industries, runs into diminishing returns, the adjustment could be long and difficult – even absent a full-blown banking crisis – unless demand can be effectively rebalanced towards consumption.
Several key dynamics emerge from this analysis. Accelerating the unwinding of excess housing supply, even at the cost of recognising losses, may shorten the adjustment. Prolonged forbearance risks zombifying local governments and developers, much as zombie banks constrained Japan (IMF 2026). At the same time, restoring household confidence requires more than stabilizing prices. Stronger social insurance, clearer income prospects, and credible policy commitment to rebalancing could help reduce precautionary savings and revive consumption (Katz 2026).
Japan’s lost decade(s) offer a cautionary tale. China still has room to shape the outcome of its adjustment, but the window narrows as overcapacity, weak consumption, and negative sentiment reinforce one another. The real estate sector may no longer drive growth, but how China manages its retreat from this once-dominant sector will shape its macroeconomic trajectory for years to come.
See original post for references
