Miao Yanliang: how did China's low inflation come about?

Former SAFE Chief Economist and current CICC Chief Strategist urges fiscal action to complement monetary policies.

Miao Yanliang joined China International Capital Corporation Limited (CICC) in March 2023 as Chief Strategist and Executive Head of the Research Department. Before that, he served for 10 years at the China State Administration of Foreign Exchange (SAFE), part of the People’s Bank of China, including as its Chief Economist since May 2018. He joined SAFE in 2013 as Senior Advisor to the Administrator and then Head of Research. Before SAFE, he was an economist with the IMF for five years. He holds a Ph.D., an M.A., and an M.P.A. from Princeton University,

In a seminar hosted by the National School of Development (NSD), Peking University on December 7, 2024, Miao Yanliang challenges the application of the traditional notion that inflation is solely a money supply problem, arguing that structural bottlenecks and persistently low expectations are key factors in explaining China’s low inflationary environment.

He particularly points to China’s unexpected fiscal contraction, which defied widespread market expectations in 2024. Despite initial hopes for fiscal stimulus, the government’s fiscal tightening led to a sharp decline in fiscal revenue, particularly from land sales, forcing local governments to prioritise early debt repayment, further tightening fiscal conditions and suppressing overall economic activity—a phenomenon Miao names China’s unique “fiscal accelerator effect.”

This unexpected shift in fiscal policy also dampened market confidence, leading to sluggish credit expansion. With businesses and consumers reluctant to take on credit, much of the liquidity injected into the economy remained unused, contributing to an increase in idle funds.

The former central bank official advocates for a coordinated approach between monetary and fiscal policies, similar to the strategies implemented by the U.S., with a more active deployment of fiscal funds. In essence, he finished the speech by calling for China’s treasury to step up spending.

Miao’s speech remains available on the NSD’s official WeChat blog, published on March 3, 2025.

缪延亮：通胀形成机制与风险应对

Miao Yanliang: Inflation Determination and Risk Response

A substantial body of literature has examined the mechanisms underlying inflation, and I have also conducted in-depth research in this field, most recently in my latest book 信心的博弈：现代中央银行与宏观经济 The Game of Confidence: Modern Central Banks and Macroeconomics. Today, I would like to focus on three key themes: (1) what we thought we knew: for example, the determinants of inflation, particularly the quantity theory of money and its limitations; (2) what we know: building on current understanding of inflation, I will analyse how the United States transitioned from low inflation to high inflation and then back to moderate inflation. I believe that inflation is determined by the interplay of fiscal policy, monetary policy, and expectations; (3) exploring potential paths for China to escape its current low-inflation dilemma.

I. "What We Thought We Knew" — The Quantity Theory of Money

1. The Quantity Theory of Money and Its Limitations

A comparison of the year-on-year growth rates of broad money (M2) in China and the US reveals an interesting contrast. In China, the M2 growth rate currently stands at 7.5%, while the U.S. saw a year-on-year decline in M2 throughout 2023, followed by a modest rebound by the end of 2024—though it remains at a relatively low level.

If one accepts the premise that money supply determines inflation, China should be experiencing high inflation and the U.S. low inflation. In reality, the opposite is true: U.S. inflation peaked at 9% and has since declined to 3%, whereas China teeters on the edge of deflation, with a Consumer Price Index (CPI) of around 0.3% and the Producer Price Index (PPI) remaining in negative territory for over 20 consecutive months.

This suggests that slow or even negative growth in the money supply does not necessarily correspond with deflation; on the contrary, high inflation may still occur.

Milton Friedman once made a compelling claim: “Inflation is always and everywhere a monetary phenomenon.” I had heard of this statement even before I began studying economics, but I now believe it is flawed.

At the heart of this claim lies the equation MV = PY, where M represents the broad money supply, and V denotes the velocity of circulation. This equation, originating from the Cambridge School, which Alfred Marshall most famously represents, functions as an identity—valid so long as all transactions are properly accounted for.

However, in practice, money serves multiple roles: it can be used for purchasing goods or for conducting asset transactions. It becomes challenging to align the variables in the equation precisely with real-world metrics. Although the equation itself is not incorrect, accurately quantifying its components is problematic.

Furthermore, the definition of M has long been debated. Recently, the People’s Bank of China (PBOC) revised the definition of M1, which I regard as a step in the right direction.

Second, even assuming that V remains constant, an increase in M does not necessarily result in a rise in P. Moreover, the M referred to in this context is the broad money supply used in real economic transactions, whereas central banks control the monetary base rather than the broad money supply.

The relationship between the monetary base and the broad money supply is governed by the money multiplier, which differs markedly between China and the United States. In the U.S., the multiplier remains relatively low, while in China, it has been on the rise. This increase in China’s multiplier is attributable to a reduction in the reserve ratio, which leads to a higher money multiplier.

There is a common misconception that, because the PBOC has not expanded its balance sheet—and has even slightly reduced it—some conclude that it is pursuing monetary tightening rather than easing. In my view, this interpretation overlooks essential nuances.

There is no doubt that the PBOC is currently in an easing cycle, as evidenced by its reductions in the reserve ratio. Lowering the reserve ratio effectively boosts the money multiplier, thereby expanding the supply of broad money in the economy—a hallmark of monetary easing.

However, the impact of a reserve requirement reduction on the PBOC’s balance sheet must be assessed on three levels.

First, from a direct accounting perspective, lowering the reserve ratio reduces the PBOC’s balance sheet size.

Second, in terms of the money multiplier effect, although the reserve ratio declines, the resulting increase in the multiplier may keep the overall supply of broad money unchanged.

Third, from the standpoint of real-world economic frictions, not all commercial banks deploy the freed-up reserves for new lending. Some may instead use the additional liquidity to repay high-interest debt, particularly those institutions that rely heavily on such financing.

As a result, the PBOC’s balance sheet may appear to contract. However, a shrinking balance sheet does not necessarily signify monetary tightening, as reducing the reserve requirement is a component of an easing policy.

It is important to note that the current monetary policy remains in its conventional phase—short-term interest rates are still adjustable and not near the zero lower bound, and reserve requirement ratios remain relatively high, at around 7%. The stance of monetary policy should not be assessed solely based on the size of the People’s Bank of China’s balance sheet. Instead, greater attention should be paid to liquidity indicators such as interest rates, reserve ratios, and broad money supply growth.

Analyses of China’s monetary policy should not be confined to the framework of quantitative easing (QE), which is relevant only when short-term interest rates are near zero and when the scope for further rate adjustments is limited. Under those conditions, the central bank’s balance sheet becomes the sole indicator of policy direction.

Even in the case of the U.S. Federal Reserve, the purpose of QE—expanding the monetary base—was ultimately to stimulate growth in broad money supply and increase market liquidity.

It must be recognized that China has not implemented QE and remains in a phase of conventional monetary policy operations, such as lowering reserve requirement ratios. As such, a shrinking central bank balance sheet should not be misinterpreted as a sign of monetary tightening when evaluating policy direction.

Finally, and most importantly, the classic MV=PY formula of the quantity theory of money, despite its influence, has three major issues: first, the exact definition of M (money supply) remains contentious; second, M is not entirely controlled by the central bank; third, V (velocity of circulation) is an endogenous variable.

In China, V—or the turnover rate of money—has been declining due to the low opportunity cost of holding money. This encourages people to allocate funds toward low-risk investments. The root of this low opportunity cost lies in low interest rates, themselves a consequence of excessive money supply.

Therefore, as the money supply increases, interest rates drop, which suppresses velocity. This chain of effects may ultimately annul the traditional relationship between money supply, nominal GDP, and price level, rendering the quantity theory of money invalid.

2. Declining Velocity of Money Circulation in China

Clues pointing to the decline in China’s velocity of money circulation can also be found in the country’s M2 growth composition. In financial data, the widening “scissors gap” between M1 and M2 indicates this trend.

China’s M1 and M2 data are highly reliable, and M1, in particular, serves as a key leading indicator for the Chinese economy. Most macroeconomic indicators tend to lag behind real economic activity, with inflation and employment being among the most delayed. Indicators like industrial production and the Purchasing Managers’ Index (PMI) offer more real-time insights. Leading indicators, however, are scarce. Notably, when M1 experiences deep negative growth, it often signals economic bottlenecks.

At the same time, a pronounced shift toward term deposits has emerged in China’s money market. The share of term deposits in overall savings has been steadily rising. Of the 21 trillion yuan increase in M2, personal deposits accounted for 14 trillion yuan, and more than 70% of these personal deposits were term deposits. This growing preference among individuals and businesses for term deposits to secure higher interest income reflects subdued expectations for future investment opportunities and a lower tolerance for risk.

Regarding the widening “scissors gap” between M1 and M2, I believe it reflects the presence of idle funds within the broader M2 aggregate. Specifically, due to existing nominal money supply growth targets, banks may be compelled to release more funds. However, when genuine borrowing demand is weak, banks often respond by lowering lending rates to retain or attract deposits—leading to a buildup of idle funds that remain within the financial system without being effectively channeled into the real economy.

This phenomenon is particularly pronounced in the rapid increase of deposits held by large enterprises. When banks cannot deploy these funds through loans, they may turn to purchasing bonds and other financial instruments. Yet, with current government bond yields at low levels, returns on such investments are often unsatisfactory. In response, banks may resort to techniques such as financial engineering to boost yields, which, however, could heighten systemic financial risks.

The relationship between falling interest rates and capital gains is also noteworthy. A 10-basis-point drop in interest rates, compounded over ten years, can result in substantial capital appreciation—especially when leverage is involved. Therefore, it is foreseeable that a large number of small and medium-sized banks will increase their participation in the bond market in the future.

As previously discussed, the People’s Bank of China's shrinking balance sheet does not necessarily signal a tightening monetary stance. However, some observers have criticized the PBOC for failing to effectively manage market expectations during periods of deflationary pressure and failing to act swiftly to lower long-term bond yields.

In reality, the PBOC intervened four times in 2024, but each intervention triggered market anxiety due to the sharp declines in long-term yields that followed. The root of this concern lies in the actions of many small and medium-sized banks, which had unilaterally bet on falling long-term interest rates by purchasing large volumes of long-duration bonds. This strategy introduced substantial financial risk.

This issue has gained heightened attention in light of the collapse of Silicon Valley Bank in the United States, which suffered substantial losses due to its massive bond purchases.

The idle funds in M2 result from collective actions by large banks, large enterprises, and small banks. These institutions contribute to a rapid increase in M2 through leveraging and balance sheet expansion. However, much of this liquidity fails to reach the real economy, thereby creating a buildup of idle money.

The sharp decline in M1 reflects significant bottlenecks in China's economy. Notably, the PBOC has recently adjusted the definition of M1 to include household demand deposits. This adjustment is reasonable because, according to the IMF’s definition, any funds readily available for payment should be classified under M1.

With technological advancements and the development of financial engineering, the definition of money is constantly evolving. As a result, the quantity theory of money is becoming increasingly difficult to apply consistently. Monetarism rose to prominence because of inflation—and ultimately faltered for the same reason.

The high inflation of the 1970s propelled the rise of monetarism, but it was also high inflation that gave birth to new forms of money. Under interest rate controls, soaring inflation caused real deposit rates to fall sharply, prompting the emergence of financial innovations such as money market funds. These new products allowed households to earn higher interest while maintaining liquidity. The emergence of new monetary instruments has challenged traditional definitions of money and monetary policy.

Therefore, I believe that one should not focus solely on money, as money is not a determinant of inflation. That said, despite changes in the definition and form of money, it remains essential to monitor the impact of money on the economy, especially when there is a significant decline in M1, which signals a substantial number of bottlenecks in the economy.

There are three main reasons for the decline in M1:

First, a reduction in risk appetite has led both enterprises and individuals to favor term deposits over demand deposits. At the same time, the sluggish real estate market has weakened corporate demand deposits, a key component of M1.

Second, low inflation expectations have pushed real interest rates higher.

However, these two factors alone do not fully account for the sharp drop in M1 observed since April 2024. This decline can only be explained by a third factor: an unexpected episode of pro-cyclical fiscal tightening.

Looking back to this time in 2023, the central government announced the issuance of 1 trillion yuan in special-purpose sovereign bonds, raising expectations for fiscal stimulus in 2024. However, since April 2024, fiscal policy has shown a marked tightening trend.

This tightening is primarily reflected in two areas. First, there has been a decline in broad fiscal revenue—most notably, a sharp drop in land sales revenue. At its peak, land sales once accounted for up to 80% of local government income, but that share has now fallen to about 50%. The downturn is mainly attributable to the sluggish real estate market and weak housing sales, which have cooled activity in the land auction market. Land sales revenue has declined from a peak of 8.7 trillion yuan in 2021 to 5.6 trillion yuan in 2023, with 2024 revenues between 3 and 4 trillion yuan.

The second factor is mounting debt repayment pressure, driven by the maturity of earlier borrowings and by regulatory mandates to contain local government debt risks.

China’s unique “fiscal accelerator effect” has become evident in the current environment. Like the financial accelerator effect—where declining real estate prices reduce collateral values and constrain credit—the fiscal accelerator is triggered when local governments encounter budgetary strain. As revenues fall, the debt-to-revenue ratio rises, prompting local authorities to prioritize early debt repayment. This creates a self-reinforcing cycle of fiscal tightening.

This tightening trend caught many by surprise, as 2024 was originally expected to be a year of expansion. Instead, it has unfolded as a year of contraction, making it difficult for businesses to access credit. The sharp slowdown in M1 growth illustrates the shortage of funding now facing the real economy.

3. Other Theories Explaining Inflation

As discussed above, the quantity theory of money falls short in explaining inflation in China. So, are there alternative theories that might shed light on the phenomenon? For example, the wage determination theory posits that deflation may be linked to declining wages. However, changes in wage levels often fail to fully capture shifts in productivity or labor costs.

Cross-border factors may also play a role in shaping inflation dynamics. In his dual economy model, W.A. Lewis argued that global commodity prices are not set by end users but by the marginal cost incurred by the last producer. If a sector has access to an abundant labour supply, it can consistently suppress prices.

The Phillips curve, a classic theory describing the relationship between the labour market and inflation, has also exhibited non-linear characteristics. Over the past decades, it has evolved from a downward-sloping or V-shape in the 1960s to a U-shape during the 1980s through the late 20th century and, more recently, to a steeper form. These shifts suggest that the relationship between unemployment and inflation has grown more complex and non-linear.

In the post-pandemic era, the Phillips curve has steepened, meaning that even a slight decrease in unemployment can lead to a substantial rise in inflation. This shift presents policymakers with both challenges and opportunities.

I once had the idea to write an article titled From Bill Phillips to William Beveridge: Understanding the Rise and Fall of U.S. Inflation. However, during research, my assistant discovered that such an article had already been published in the Journal of Monetary Economics, and its line of thought was strikingly similar.

The Phillips curve and the Beveridge curve serve distinct roles in economics. The Phillips curve primarily describes the relationship between employment and inflation in macroeconomics, whereas the Beveridge curve focuses on the labour market at a microeconomic level. In the labour market, vacancy rates and unemployment rates are two key indicators.

In equilibrium, the labour market typically exhibits a 1:1 relationship between vacancy rates and unemployment—one job opening for each job seeker. However, in 2022, the United States experienced unusually high vacancy rates. In response, Federal Reserve Governor Christopher Waller proposed a potential path to a “soft landing,” whereby reducing vacancies through slower hiring—rather than through layoffs—could ease inflationary pressures without driving up unemployment.

This strategy ultimately proved a success. Contrary to widespread expectations, U.S. unemployment has remained low at around 4%. In my opinion, this outcome can largely be attributed to two factors: a resurgence in immigration and the return of individuals who had exited the labour market during the pandemic.

Another theory is the fiscal theory of the price level (FTPL). Christopher Sims, whose class I took at Princeton, founded this theory. According to FTPL, the success of the United States in curbing high inflation was not solely the result of labour market adjustments but also owed much to fiscal policy. Despite very low policy interest rates, the U.S.'s CPI did not rise significantly, largely because fiscal policy was contractionary at the time. It was only after the pandemic that fiscal policy turned expansionary, triggering high inflation. This case shows that price levels are determined not solely by monetary policy but by a combination of fiscal and monetary policies.

II. "What We Know" — The Joint Determination of the Price Level by Fiscal Policy, Monetary Policy, and Expectations

1. Distinguishing Credit from Money

Forecasting inflation is not easy for ordinary citizens, but one practical approach is to observe credit conditions. Loose monetary policy does not necessarily translate to loose credit—as the saying goes, “You can lead a horse to water, but you can’t make it drink.” However, loose fiscal policy can effectively promote loose credit.

From the perspective of banks, credit constitutes an asset, while money is a liability. In contrast, for households, credit represents a liability, and money is considered an asset. Thus, credit expansion is fundamentally a process of increasing leverage.

Before 2008, the United States did not experience inflation due to deleveraging. During that period, credit expansion consistently lagged behind monetary expansion. However, since 2020, the U.S. has implemented aggressive fiscal stimulus policies, and credit expansion began outpacing monetary expansion, triggering inflation.

China’s experience during that time was markedly different. Despite prolonged PPI deflation lasting over 50 months, the CPI did not enter deflationary territory. This was because China was in a phase of credit expansion, with significant capital flowing into the real estate sector, thereby sustaining economic growth. Under such conditions, deflation was virtually impossible.

Today, however, the situation has shifted. Credit expansion in China now lags behind the monetary expansion, and household leverage ratios have leveled off.

2. Recognising the Role of Inflation Expectations

At present, expectations have become a crucial determinant of credit activity. When confidence and optimism about the future are widespread, such expectations often become self-fulfilling.

The factors that shape expectations are complex and beyond the scope of this discussion. However, a marked increase in searches for the term “deflation” in the Baidu Search [the equivalent of Google Search in China] Index since 2023 reflects the surge in public concern about the economic situation. This growing anxiety is particularly pronounced amid falling housing prices and declining wage expectations.

3. How did low inflation come about in China?

Real estate constitutes a significant portion of assets in China, making changes in the property market highly influential on the overall economy. Currently, China is facing challenges from a downward real estate cycle, a stark contrast to the previous upward cycle. Compounding this is that the economy is also in the second half of a financial cycle, adding layers of complexity. While the precise duration of this downward phase remains uncertain, historical patterns and international comparisons provide some basis for forecasting its trajectory.

At present, monetary easing alone is insufficient to pull the economy out of the real estate cycle. To understand why, it is essential to distinguish between funding liquidity and market liquidity. The People’s Bank of China can inject funding liquidity into the money market, but if market participants remain pessimistic about the future and are unwilling to take on credit risk, funding liquidity, however abundant, can barely be converted into market liquidity. Consequently, asset prices may continue to fall. Even if housing prices appear to rise, there is a risk that this reflects price movements without real transaction volume.

III. How Can China Break Through its Low-Inflation Dilemma? By Strengthening Fiscal-Monetary Policy Coordination to Reverse Inflation Expectations

The result is idle funds if there is funding liquidity but no market liquidity. To address this issue, it is necessary to identify a method to restore market liquidity. One viable approach is for the PBOC and fiscal authorities to collaborate. As the lender of last resort, the PBOC can provide unlimited liquidity and leverage, while the fiscal authorities can assume credit risk and supply capital through mechanisms such as taxation.

This form of joint intervention has been successfully employed in the United States on multiple occasions. For instance, special-purpose entities have been established to inject funds into real estate firms, facilitate corporate restructuring, or directly acquire affordable housing. Such measures not only enhance the leverage effect of funds but also effectively resolve challenges in the property market.

In summary, the central challenge facing China’s real estate market today lies in identifying the funding sources and determining their allocation. Addressing this issue requires a coordinated policy mix of monetary policy, fiscal policy, and market mechanisms that restores market liquidity while ensuring stable economic growth.

Regarding funding sources, the primary reliance is currently on the PBOC, with fiscal funds yet to be deployed. On the allocation side, it is crucial to ensure a level playing field between private enterprises and state-owned enterprises (SOEs). If public funds are used only to support SOE-housing developers while neglecting those of private developers, confidence in the private developers will remain weak. In such a case, the market adjustment would undergo a slow and natural process.

Thus, the core issue concerns the source and allocation of funds, necessitating the design of more effective incentive mechanisms and institutional arrangements.

IV. Conclusion

To briefly conclude, my understanding of inflation and deflation has been shaped by oversimplified assumptions. I once firmly believed in the quantity theory of money, assuming that changes in the money supply would directly lead to inflation or deflation. However, reality has proven otherwise.

As discussed above, when market participants are unwilling to assume credit risk and expand borrowing, no amount of monetary expansion or funding liquidity can translate into actual market liquidity. The result is the accumulation of idle funds.

Therefore, the current imperative is clear: to break free from the trap of idle funds and enhance market liquidity.

Achieving this goal will require coordination between the PBOC and fiscal authorities. In this partnership, fiscal authorities must take on the roles of borrower of last resort and payer of last resort, while the PBOC serves as the market maker of last resort and lender of last resort.

Hopefully, steps will be taken in the right direction. As long as the direction is correct, no matter how distant the destination may appear, it will ultimately be reached.

Wang Yongli on U.S. dollar hegemony and digital RMB

Wang Yongli on SWIFT, sanctions, and sovereignty