This is the transcript of “Roundtable IV: Domestic and Global Rebalancing” at the 2026 Tsinghua PBCSF Chief Economists Forum.
Participants were
Huang Yiping, Boya Distinguished Professor and Dean of the National School of Development at Peking University;
Zhang Xiaojing, Research Fellow at the Institute of Finance and Banking of the Chinese Academy of Social Sciences and Director of the National Institution for Finance and Development;
Wu Ge, Chief Economist and Assistant to the President of Changjiang Securities;
Shi Kang, Chair Professor at the Tsinghua University PBC School of Finance and Director of its Center for International Macroeconomic Policy.
Huang Yiping called on the central government to take on more debt to repair the balance sheets of local governments and households. He also argued that stimulus policy needs to be more imaginative. Despite successive rounds of new policies over the past three to four years, downward pressure on the economy has persisted. At this point, he suggested, somewhat more aggressive stimulus is needed to break that cycle. (My first thought was another package on the scale of the measures announced at the joint press conference by several major government agencies on September 24, 2024.)
Zhang Xiaojing emphasized that fiscal and monetary policy should be supplemented by “asset policies”: putting public assets to more productive use, improving their returns, and supporting households through mortgage relief, stronger social protection, and better public services. He also highlighted the tension between “investing in people” and central–local government relations—specifically, how governments are held accountable for results. Local governments may invest in people, but those people can move elsewhere. The benefits are therefore difficult to capture in local officials’ performance metrics, while the expenditure remains clearly visible. This weakens local governments’ incentives to make meaningful investments in people. To address this problem, Zhang argued that the central government should take the lead, with performance assessed over a longer time horizon.
Wu Ge, a veteran of Chinese monetary policymaking who spent many years in the People’s Bank of China’s monetary policy department, reiterated that economic weakness cannot be attributed entirely to structural factors such as population aging. Countercyclical policy still has a role to play. Fiscal, monetary, and other policies should work in concert, and the transition from old to new growth drivers should follow the principle of “establishing the new before dismantling the old,” rather than allowing policies to offset one another. He also called on policymakers to do more to guide market expectations. Monetary and fiscal policy should have a clear “reaction function”: excessively low inflation or weak nominal growth should automatically trigger interest-rate cuts, fiscal expansion, and measures to stabilize the housing market. If policymakers fail to act even as nominal indicators deteriorate sharply, markets will inevitably begin to question their resolve.
Shi Kang likewise supported greater central government borrowing to ease pressure on local governments. His particular concern was distribution: as industry shifts from labor-intensive to capital-intensive production, the gains from exports and technological progress may no longer flow automatically to ordinary households. Stronger policies to improve income distribution are therefore needed to ensure that households share more broadly in the benefits of growth.
Below is the full English version of the transcript I made with the help of AI. You can find the Chinese ver on the WeChat blog of New Economist 新经济学家智库
Shi Kang: Let us now move into the discussion. Rather than introducing another question, let us first ask Professor Huang to finish the fourth point he was making.
Huang Yiping: I would be very happy to continue. We are indeed facing some new issues, but imbalances are not necessarily all bad. My own thinking is evolving as well. We had global imbalances in the past, and now we are seeing new developments and some unusual factors.
If artificial intelligence develops in the direction I envisage, the mismatch between aggregate demand and aggregate supply may not disappear anytime soon. It could persist for some time.
At the beginning, Wu Ge asked: What problem does economics actually seek to solve? Our assumption is that resources are scarce. Economics studies how to use limited resources to produce more output and meet people’s needs, while also addressing the relationship between efficiency and fairness.
My preliminary assessment is that the mismatch between aggregate demand and aggregate supply may persist, and that this persistent imbalance would be detrimental to sustainable economic growth. In the past, China was still a relatively small economy, and external markets were more receptive to its output. Now that China has become a large economy, greater resistance in international markets would make domestic capacity pressures more pronounced. Going a step further, if AI affects supply and demand asymmetrically across national economies, the problems China faces today could become global problems in the future.
This has an important policy implication: for some time to come, we may need sustained policy support for aggregate demand, rather than just a one-off countercyclical stimulus when the economy is temporarily weak.
We can already see two things. First, the economy faces a degree of downward pressure every year. Policy support stabilizes it for a while, but the underlying downward pressure does not go away. Second, international imbalances seem to be widening.
Perhaps this is what Wu Ge was getting at earlier. Although I do not fully agree with his specific proposals, I agree with the general direction: we may need to consider more forceful ways to increase aggregate demand. This has become a very pressing issue, and it is not something we can resolve with a single intervention today.
More specifically, I would divide this into two broad areas. The first is structural measures, primarily to address domestic demand. Fundamentally, China’s insufficient domestic demand is mainly a problem of insufficient consumption. Investment also matters at the margin in the short term, but looking at the overall structure of the economy, the imbalance between strong supply and weak demand is quite evident.
We cannot rely indefinitely on consumer credit or trade-in programs to increase consumption. These measures can provide a short-term boost, but they cannot sustain a lasting increase in spending. Ultimately, it comes down to income and confidence.
Structural issues are also connected to industrial upgrading. We need to think further about what kinds of market mechanisms facilitate an efficient allocation of resources. Businesses and capacity that should exit the market need to be allowed to do so. We also need to discuss what kinds of industrial policies can promote upgrading without creating excessive pressure from surplus capacity.
I believe it is important to push ahead with market-oriented reform. On the one hand, we should allow the market to play a greater role in allocating resources. On the other, we should leave more of both resource allocation and income distribution to the market. This would help increase the household share of income, benefiting both consumption and the development of new quality productive forces.
The third structural issue concerns international cooperation. I broadly agree with what Shi Kang said about the pressures being felt overseas. China produces goods well, and other countries benefit from buying Chinese products. But China is now a large economy, which is different from being a small one.
Singapore, for example, has a very large current-account surplus relative to its economy, but its global impact is limited. When China exports a little more or a little less, it affects other countries’ economic structures, industrial development, employment, and income distribution.
We therefore need to build a new framework for international economic cooperation. As we pursue our own development, we also need to consider whether other countries can benefit from it. For example, we could engage in more investment cooperation and industrial cooperation. When companies manufacture overseas, they do more than export technology and products: they can also create local jobs, income, tax revenue, and GDP, contributing to shared prosperity. In the long run, I believe this is the direction we should pursue.
The immediate challenge, however, is that we are already facing this situation. We already have a policy framework, and many frameworks and measures to promote consumption have been introduced. On market-oriented reform, the Third Plenary Session of the 20th CPC Central Committee also called for the market to play its role in resource allocation.
Even if these measures prove effective, they may take time. They will not necessarily deliver results right now. In the meantime, do we need some countercyclical—or, we might say, aggregate-demand—policies to lift overall demand?
I have heard many of your analyses of fiscal and monetary policy: which is better, fiscal policy or monetary policy? I do not have a particularly strong preference. But when choosing policies—for example, whether it is more important to stimulate investment or consumption—the question is not whether we happen to prefer investment or consumption.
Rather, if we already have strong supply and weak demand, will investment made today ease that imbalance tomorrow, or make it worse? That is something we need to consider.
Both fiscal and monetary policy can be very effective. I will not discuss today whether fiscal policy works more directly than monetary policy. I still think investment is important, and stimulating consumption is not that easy. That is why I particularly agree with the point Zhang Xiaojing made.
If the short-term economic weakness and widening external imbalance we discussed earlier are related to a relative contraction in investment, how do we change that? If a whole range of issues—local government debt, real estate, excess capacity in traditional industries—are connected to this, then one possible solution, in my view, is balance-sheet restructuring.
My suggestion is to use increased leverage at the central government level to repair or improve the balance sheets of local governments, financial institutions, businesses, and households.
In essence, this is what the United States did during the subprime mortgage crisis: the government increased its leverage while businesses, institutions, and households deleveraged. When the crisis ended, the economy rebounded strongly.
If the balance sheets of these individual economic actors are not repaired, stimulus policies will ultimately be much less effective, because those actors will lack the capacity to undertake new activities.
I will not go into the specific mechanics. We need to think about what Wu Ge has been saying and come up with new approaches—to “think outside the box.”
We have been using these traditional policies for three or four years now. Each year, we provide some stimulus, the economy stabilizes for a while, but the downward pressure remains. If we adopt somewhat more aggressive measures, perhaps we will have a chance to break out of this cycle.
Shi Kang: That was very well put, Professor Huang, and it brings us back to Professor Zhang’s point. Professor Huang’s position essentially supports increasing central government leverage.
Those of you who were here this morning will remember that Professor Ju also raised this. His presentation was on renminbi internationalization and three dimensions of rebalancing in the Chinese economy. His three dimensions were somewhat different from yours: his were more macroeconomic, while yours are more focused on the micro level.
Professor Justin Yifu Lin disagreed with his proposal, but I think the general direction is right. Finding ways for the central government to increase leverage, and thereby relieve pressure on local governments, is an important direction.
Professor Huang also made an excellent point about moving beyond the traditional framework and finding new approaches. We cannot keep focusing only on trade-in programs or debating cash transfers—although cash transfers are still a good thing.
Xiaojing’s proposals and views are relatively new to most people. Please elaborate.
Zhang Xiaojing: Let me offer a few more thoughts, including on Professor Huang’s suggestion that we adjust the framework.
In the past, we relied on fiscal and monetary policy. Wu Ge also reminded us not to overlook monetary policy. Everyone talks about fiscal policy all the time. I think both are important, but neither is sufficient. Keynesianism is not enough.
I would like to add another category. I have already published an article on this—I believe it was in the eighth issue—arguing that we need “asset policy.”
That may sound new, but think about the proposal at the Third Plenary Session of the 20th CPC Central Committee to explore the management of the national macroeconomic balance sheet. This is essentially what it means.
We need to recognize that, in addition to the three main pillars of macroeconomic policy we usually discuss—planning, fiscal policy, and monetary policy—we should add a new one: balance-sheet policy. For simplicity, I call it asset policy.
What does that mean? Its most important element is the point we just discussed: a moderate rebalancing and redistribution of wealth between the government and household sectors.
How should we do this?
First, as we have said, China has substantial public-sector net assets. International comparisons show that our net assets are far greater than those of advanced economies. But all countries have public-sector assets—the United States has them, and Sweden has them too.
This has led to a new strand of thinking about public wealth and the wealth of nations, emphasizing how public-sector wealth can be accumulated and how its returns can be improved.
So my first proposal for asset policy is to put our many existing public-sector assets to productive use and increase their returns.
A one-percentage-point increase in the return on public-sector assets worldwide would be enough to meet global public infrastructure needs. Just one percentage point would suffice.
Imagine what it would mean if we increased the return on our own public-sector assets by one percentage point. I will not go into the specific figures, but it would be highly significant.
We must not leave this wealth sitting on paper. That is particularly important: we need to put it to work. Unlocking the value of existing assets has now become one of the most important tasks for local governments.
The second point is transfers. We should not look at all these assets and conclude that we must keep expanding production. In many sectors, further expansion is unnecessary.
I think we should make appropriate transfers. There are many ways to do this, including the fiscal and financial methods mentioned earlier. Residential mortgages are also relevant.
I have heard an argument that makes a great deal of sense, and I have raised it myself on several occasions: household mortgages are one of the most important sources of support keeping our banks going, because their interest rates are quite high.
Why do so many banks see consumer lending as a potential new opportunity? Because interest rates on consumer loans are also relatively high. But the scale of consumer lending is far smaller than that of mortgage lending.
If, in a sense, households have been subsidizing the financial sector—and even the state-owned sector—through their mortgages, we now need to find ways to reduce that subsidy and instead provide support to households.
How do we do that? Cutting interest rates seems difficult; we will not discuss that issue today. But even without cutting rates, we can subsidize interest payments. We have had many discussions with the Ministry of Finance about this.
I believe it will be extremely important to use fiscal and financial measures to reduce this burden on households. Reducing it is itself a form of wealth rebalancing.
Then, of course, there are pensions, healthcare, social security, and related public services. We absolutely need to strengthen these.
The central leadership has called for “investing in people.” In the past, we did not really have a clear concept of what that meant. Today, I think taking a balance-sheet approach represents a major change in how we think about both investment in people and debt risk.
In many past discussions, I have pointed out that, over more than 40 years, China’s assets and liabilities grew and expanded together. We incurred liabilities, but assets were created in return.
Whether through industrialization or urbanization, we could see the buildings, infrastructure, factories, machinery, and equipment. The physical assets were right there. Debt on one side, assets on the other: they appeared well matched, with no risk.
But if we invest in people today, we have investment and debt, yet no visible asset formation, because human capital is largely absent from our national balance sheet. This is an omission under the System of National Accounts, or SNA, framework.
Investing in people will be a national strategy for a long time to come. If we borrow to invest in people but cannot see the resulting assets, the two sides will appear out of balance. Unless we find a way to address this, governments will not be able to commit fully to investing in people. It simply will not happen.
Many local governments are already complaining: “You are asking us to invest, but once we have done so, nothing tangible has been created. We will be held accountable, and the assessments will show that our risks are very high.”
And that is before we even consider another problem: what if a local government invests in someone, and that person then leaves? The externalities are substantial.
I therefore have two points about this shift. Investing in people is itself a form of balance-sheet rebalancing.
First, the central government should take responsibility for this. That would address the externality problem.
Second—and this is very important—our future assessments of fiscal balance and debt risk must take an intertemporal perspective.
We cannot say: “Today we invested in someone. They attended a training course, listened to a concert, went to a live performance, and improved their capabilities in various ways. Tomorrow we should be able to see higher human capital on the balance sheet, along with patents and innovations.”
That is impossible. The assessment horizon must be extended. We need to look across time.
How should we measure the performance—the KPIs—of investment in people? This is a challenge for countries around the world. Some have established satellite accounts for human capital. Denmark and many other countries are doing this, and we should too.
My third proposal relates to Professor Huang’s point about the impact of AI.
AI’s effects are so substantial that asset policy must take them into account. Its impact is economy-wide. This is not simply a case of one industry suddenly running into trouble, followed by some targeted central government measures. Do we have enough money to support the process of so-called creative destruction on this scale?
One very important consideration is that local governments—and the central government as well, which has established a fund to guide investment in innovation—are investing in many AI companies.
Much of this investment takes the form of equity stakes, because many of these companies are privately owned. Through the returns generated by innovation, including AI innovation, governments will acquire substantial new assets and wealth.
My proposal is that, because AI’s impact is broad and systemic, the wealth obtained from these innovation returns should be earmarked specifically for the people adversely affected by AI.
This is my third asset-policy proposal: governments should accumulate new kinds of assets related to AI and technological innovation, and use those assets to support the people harmed by creative destruction.
Through these approaches, I believe we can gradually establish an asset-policy framework that goes beyond the traditional combination of fiscal, monetary, and planning policies. It would allow us to put our accumulated wealth—and we have a very substantial stock of it—to genuine use.
If we fail to put that wealth to work, then having it amounts to very little. Thank you.
Shi Kang: Professor Zhang has further explained how we might improve or reform the way we compile, assess, and even measure balance sheets. I find this particularly thought-provoking in the age of AI.
Around 2000, we had global imbalances. China ran large surpluses between 2000 and 2007. At that time, China’s economy was labor-intensive.
What did that mean? It meant that when China ran an export surplus, exporters of all sizes benefited. Workers and their families benefited too. It was a surplus whose benefits were broadly shared.
Today, when we conduct field research, we find that many companies—in semiconductors, integrated circuits, or new-energy vehicles—are capital-intensive.
This raises an issue in the context of AI: as industries diverge, the gains from external demand, or the benefits associated with these imbalances, are no longer broadly shared.
We therefore need a range of distributional policies, and I believe that need is urgent.
Now, let us invite Wu Ge to continue discussing monetary policy. I have a very specific question, and I wonder whether you agree.
The current policy stance is essentially described as “moderately accommodative” monetary policy, together with proactive fiscal policy. Would you favor removing the word “moderately”? Or would you add another qualifier to describe the monetary policy we need now? That is my question.
Wu Ge: The Chinese language is rich and nuanced, so you can choose whatever wording you like. But what people care more about is whether they feel a tangible financial benefit.
Speaking of tangible financial benefits, may I ask those of you who have stayed until the end a question? The central leadership attaches great importance to your income and your income expectations. So let me ask: How do you prove that your income is high, or that you have money?
That might sound like a pointless question: “I know perfectly well whether I have money.” But from a macroeconomic perspective, or from the perspective of money and banking, what does it mean to have money? Where is the money? Where has it gone? How does the amount of money increase? How does society as a whole acquire more money? Which comes first—the chicken or the egg?
From the perspective of money and banking, the answer is actually quite straightforward. All our income—including corporate income and even government revenue—exists as deposits within the banking system’s payment network.
Very few people can operate outside a payment or account system dominated by commercial banks.
I do not know whether everyone agrees with this principle, but it is a basic principle of money and banking: our money cannot escape the reach of the commercial banking payment system. That is the first point.
If all our money, or the vast majority of it, is in the banking system, what do we mean by the money people have? How much money do they have in their pockets? At the macro level, we call this M2.
What is M2? It is people’s deposits; it corresponds to how much money they have.
Where does that appear on the balance sheet of a bank, or of the commercial banking system as a whole? On the liability side.
All our money is in the banking system. Banks have assets and liabilities, and their liabilities consist primarily of M2—that is, people’s deposits.
And on the other side of the balance sheet? Banks can use funds to make loans, buy bonds, do other things, or even purchase foreign exchange. All of these can create money.
So, from a macroeconomic perspective, imagine treating the whole of society as one large system. Put together all the money people hold at commercial banks, and consolidate all commercial banks into a single bank.
Then ask: If people are to have more money and feel that they have more money—if M2 is to rise—what must happen?
The asset side needs to expand.
This is what macroeconomics and the theory of money and banking mean when they say that loans create deposits, or that credit creates deposits.
Where does money come from? It is generated through credit expansion.
So, if we want to meet people’s income expectations and aspirations, then from a money-and-banking perspective, we need a mechanism for credit creation. And that mechanism requires expansion on the asset side: lending and bond purchases.
In that sense, talking about income in isolation is not very meaningful. At its source, income—or deposits—is closely related to credit.
What, then, can support a steady expansion of credit, rather than the slowing year-on-year growth in outstanding credit that we have been seeing?
Some people would object: “No, this is simply what happens as a society ages. People’s desires diminish.”
But that is a matter of debate. The Nobel laureate Modigliani argued precisely that an aging society can stimulate consumption demand. The most rational elderly person would spend their last dollar at the very moment they leave this world.
So old age, especially the final stage of life, should be a period of substantial consumption, not contraction. Of course, this is debatable.
There are also many countries whose populations are much older than ours, and whose low-birth-rate problems are much more severe. Yet we have recently seen many positive changes in those countries.
Structural factors, long-term factors, and potential growth certainly affect the trajectory of credit creation, the opportunities for it, and income expectations. But they are not destiny.
Some countries are more deeply aged than China, yet when they get their countercyclical policies right, we can see employment rise—and even, in some cases, birth rates.
In this sense, cyclical and structural issues are not opposites. Handling cyclical problems well can help ease structural problems, and addressing structural problems well can help ease cyclical ones.
China faces multiple challenges and seeks to achieve multiple objectives. According to economic theory—including frameworks such as the Mundell–Fleming model and the Tinbergen rule—multiple objectives require multiple policy instruments.
What we need, therefore, is a package of policy measures with at least three arrows in the quiver. Trying to advance with a single instrument will not work. That is my first point.
Second, Chinese thinking, and Eastern philosophy more generally, tends to emphasize balance and moderation.
In this process, we really need to follow the central leadership’s principle of “establishing the new before dismantling the old.”
Take local government finances. If local fiscal activity is contracting, some other source of support must make up for it. If that replacement is not yet strong enough, the contraction should be smaller.
The same applies to the transition from old to new growth drivers, and from traditional industries to new industries.
The central leadership says we should establish the new before dismantling the old. We are very enthusiastic about new things—but does liking the new mean that we should reject the old?
How should we balance the two?
If they are moving at different speeds—if new industries are growing quickly while old industries are declining sharply, but the old industries still account for a very large share of the economy—should we reconsider the balance of support between them?
In this regard, the central leadership has placed considerable emphasis in recent years on assessing policy consistency: which policies encourage credit expansion, and which cause credit contraction?
If I remember correctly, the National Development and Reform Commission had a dedicated department to assess whether different policies were working at cross-purposes and whether they were consistent.
This is critically important. As we often say, macroeconomics contains many fallacies of composition: what is rational for an individual actor is not necessarily rational for the system as a whole.
My understanding is that we need to guard against precisely this fallacy.
Over the past few years, pressure on local finances has led many local governments to pursue additional tax revenue through practices that I believe the central leadership strongly opposes.
In reality, however, local governments still engage in them for various reasons. I think these are short-sighted methods that exhaust the very sources of future revenue.
They have a very substantial impact on people’s expectations and confidence, especially the confidence of private entrepreneurs. This deserves particular attention.
My final point is that policy needs rules—especially countercyclical policy.
Countercyclical policymaking is an art rather than a science. But whether we are talking about fiscal or monetary policy, I think we need rules and benchmarks.
I am not saying that we must mechanically adopt a rule used in another country, such as the Taylor rule. We can be more flexible. But we still need objectives.
For example, macroeconomic policy should take into account both real GDP and nominal variables. When nominal variables decline very sharply, policy should respond countercyclically.
Rules are crucial because credible rules can influence people’s expectations without every policy necessarily requiring immediate, concrete expenditure.
To borrow the image of Zhuge Liang’s empty-fort strategy: you can display your rule on the city wall without moving a single soldier, and people will still believe you.
But that credibility depends on whether you have actually followed through on the objectives you set in the past—whether your actions match your words.
For example, if we say we want to promote a moderate recovery in prices, then we must genuinely work to achieve it. If we say we want to halt the decline in the property market and stabilize it, then we must actually do so.
I believe the market would then work alongside the central government to move the economy toward a better, higher-quality balance. Thank you.
Shi Kang: Wu Ge has offered some excellent insights. I would like to ask another question, perhaps one that particularly interests me personally.
I recently read an article you wrote about exchange-rate movements following Federal Reserve rate hikes.
This is something many people are concerned about now. Perhaps the Fed will raise interest rates once more, or even enter a new tightening cycle. What impact would that have on the imbalances we are discussing?
After the financial crisis, those imbalances clearly narrowed. But what might the subsequent effects be this time?
Wu Ge: That question really should go to Professor Ju, who is an expert on international issues. But let me offer a brief response.
First, the market’s general view is that this rate increase would not necessarily mark the beginning of a full tightening cycle.
As everyone knows, the US economy has its own problems. An important backdrop to this round of rate increases is the situation in the Strait of Hormuz and, subsequently, the straits around the Red Sea. Following the closure of both passages, crude oil prices rose by $100.
Against that backdrop, when policy could otherwise lean either hawkish or dovish, such a powerful cost shock calls for a response. It may not be a repeat of the Volcker era, but I think a policy response is understandable.
Yet that is not what markets are most worried about—or at least not their main concern.
Over the past two years, we have seen a very unusual phenomenon in the United States: interest rates, especially market interest rates, have kept rising.
In principle, when US market interest rates rise, the dollar should appreciate. But over the past two years, we have instead seen yields on 10-year Treasuries and other US government bonds rise while the dollar has depreciated. The higher rates go, the weaker the dollar becomes.
Why?
Because people’s assessment of the US economy is no longer based only on the Taylor rule—growth and inflation—or on whether there will be one rate hike or two. Those considerations still matter, of course.
But a more important question has emerged since the Trump administration took office: Is the United States still a beacon for investors? People have begun to doubt that.
Over the past two years, overseas investors have therefore been reducing the share they allocate to US Treasuries and other US investments.
Selling US Treasuries or dollar-denominated assets is what allows the two developments to occur together: a rise in 10-year Treasury yields and a depreciation of the dollar.
Of course, the United States still has its technology sector and its AI industry. But, objectively speaking, technology and AI are themselves undergoing an adjustment.
Like Professor Huang, I believe that AI holds tremendous promise for the future. But from an investor’s perspective, what kind of business is AI, exactly?
At present, it is an industry characterized by high costs, high debt, and rising marginal costs—but also great promise.
From a business or investment perspective, that does not make it particularly well suited to a short-term, frenzied rush of investment.
In this sense, perhaps it resembles the railways during the Industrial Revolution. Railways could carry us toward distant horizons and a better future, but very few investors may actually have been able to make money from them in the short term.
Taken together, these factors mean that markets will certainly react to Fed rate hikes or cuts, but investors are more concerned—especially since Trump took office—about the sustainability of US public finances, the durability of central bank independence, and even the reshaping of the international political order.
There was a news report this morning that Trump had renewed his interest in Greenland.
I think all of these issues influence how investors view the United States. Their concerns go well beyond whether the next short-term rate move is 25 or 50 basis points.