How can policy makers respond to populist politics?

Sep 26, 2017

Sections Public Policy

Collections Monetary Policy

Tharman Shanmugaratnam, Singapore’s deputy prime minister and coordinating minister for economic and social policies; JPMorgan Chase International’s Jacob A. Frenkel; and Chicago Booth’s Raghuram G. Rajan spoke with Booth’s Steven J. Davis at the Fourth Asian Monetary Policy Forum, an annual event meant to surface research and insights relevant to economic and financial policy­. The forum, which takes place in Singapore, is co-organized and funded by Chicago Booth, the National University of Singapore Business School, and the Monetary Authority of Singapore. This is an edited version of their discussion.

When should we normalize monetary policy?

Davis: We’re going to cover a few topics, and I will open with Jacob Frenkel and ask him to tell us his views about monetary-policy normalization, the outlook for that in the United States and maybe other advanced economies, and in particular what that means for the rest of the world.

Frenkel: In order to know and to justify where we should go from here, we need to know why we are here. It all started with the financial crisis that erupted in 2008–09. There is a famous story written about 200 years ago by Irving Washington about a Dutch guy named Rip Van Winkle. According to the story, Rip Van Winkle, who lived in the US, went to sleep before the American Revolution and woke up much later after it. The story describes how surprised he was to observe all the changes that had taken place, as the world into which he woke up was very different from the world that he was accustomed to. If Rip Van Winkle were a central banker going to sleep in 2007, prior to the financial crisis, and woke up today, he would not recognize central banking. Central Banking today, 10 years after the crisis, is very different from what it used to be. That’s not a criticism. I only say it has fundamentally changed.

We are now talking about “normalization” of monetary policy. Normalization means that we have a concept of what normal is, and obviously the past 10 years, which have been characterized as unconventional policy, have been a departure from normal. There is an entire generation of people who have never witnessed positive real interest rates and who have never experienced inflation. Think about the fact that young people under the age of 18 do not usually read the financial press, and if they were 18 in 2007, they’re now 28. So everyone under the age of 28 has never experienced living with positive real interest rates or inflation. These observations imply that the concept of “normal” is rather complex. 

The financial crisis called for fundamental policy measures. Against the backdrop of fiscal policies being overstretched, of debt being excessive, and of leverage being high, monetary policies have become the “only game in town.” Central banks all over the world have lowered interest rates to unprecedented, low levels. When interest rates dropped close to zero, central banks ran out of conventional ammunition and turned to unconventional measures, which reflected themselves in a rapid extension of the size of the central banks’ balance sheets and also a fundamental change in the composition of their assets.

When Rip Van Winkle went to sleep in 2007, he knew that the asset side of the balance sheet of the central bank contained only high-quality papers, typically short-term government debt, highly liquid, with great ability to sell and buy in the market without creating havoc across sectors. But when the financial sector got into trouble, one of the purposes of the central banks was to remove toxic bad assets from the balance sheets of the banking system. Those assets, typically mortgage-based securities (MBS), found themselves on the balance sheets of the central banks. Hence, whereas in 2007 the Federal Reserve had about 95 percent of its assets in short-term Treasury bills, in 2010 65 percent of the assets became mortgage-based securities. These developments resulted in fundamental changes in the properties of the Fed’s balance sheet.

Where are we now? Much of the policy debate has focused on the cost of normalization. This discussion has not been sufficiently balanced because it has not factored in the cost of delayed normalization. In order to restore balance to the discussion regarding normalization of monetary policy, it is important to recognize that maintaining an exceedingly low interest rate and delaying the process of normalization entails economic costs. Some examples of such costs are:

  • The low interest rates induce investors to seek alternative ways to generate returns. By chasing after yield, investors end up assuming higher risk, which might be mispriced.
  • The low interest rates bring about an inflation of stock prices and may generate a financial bubble.
  • Corporations divert their efforts toward stock buybacks instead of allocating their resources to investment in their plant and equipment. 
  • The inflated financial markets create a disconnect between the real and the financial sectors of the economy.
  • The low interest rates encourage excessive leverage and thereby increase the vulnerability of the financial system.
  • The low interest rates and the flat yield curve result in negative consequences for the financial-services industry. This includes banks, insurance companies, and pension systems.
  • The transmission of the effects of monetary policy through the economy operates through the financial system; a weakened financial system thereby reduces the effectiveness of monetary policy.
  • The low interest rates provide an artificial stimulus to interest-sensitive sectors, such as housing. Since this sector is typically a low-productivity sector, it results in an overall reduction in the productivity of the economy.
  • The excessive reliance on monetary policy enables governments to postpone the necessary fiscal and structural measures. The postponement of these measures reduces the flexibility of the economy, and thereby reduces productivity and growth.

This partial list of the negative consequences of excessively low rates of interest suggests that delayed normalization is costly, and that one should always balance these costs against the cost of normalization. Obviously, most central banks are fully aware of these considerations, but on balance, it seems that the public debate of these issues puts a greater weight on the arguments highlighting the cost of normalization than on the arguments highlighting the cost of an undue delayed normalization. If such a bias exists, it is likely that when the process of normalization does take place, it will be implemented too late and might proceed too slowly.

My concern with the possibility that normalization might be excessively delayed is enhanced by the fact that recently all the major central banks put extraordinary emphasize on achieving their 2 percent inflation target. This is true of the Federal Reserve, the Bank of England, the European Central Bank, the Bank of Japan, and others. There are many reasons (many of them not under the control of the central banks) why, in spite of extraordinary efforts, the rates of inflation in the industrial world have been very low and below the 2 percent target. If interest rates are to be maintained at exceedingly low levels, for as long as the rate of inflation is below 2 percent, then there is a significant possibility that financial stability might be at risk. This concern has been forcefully voiced by the Bank for International Settlements, and I believe that it would be prudent to pay a significant attention to this concern. 

“One of the costs of excessive reliance on monetary policy is that it enables governments to delay necessary fiscal and structural measures. In this regard, by being the only game in town, the central bank exposes itself to the risk of losing central-bank independence.”

Are we ready to normalize? The Fed has been clear about that. It has said all along to look at three things. Number one, are labor markets sufficiently robust? Indeed, unemployment today in the US is the lowest it has been for more than a decade. Second, is economic growth sufficiently robust? In this regard, 2–2.5 percent GDP growth per year is not everyone’s dream, but it is clearly not a recessionary performance. In fact, since it seems that “potential” growth has declined recently, a growth rate of 2–2.5 percent is not bad, especially since it is now obvious that the precrisis high-growth rate was not sustainable. Third, while the inflation rate is currently slightly below 2 percent, estimates suggest that the path of inflation will converge to the 2 percent target before too long. In fact, it could be argued that if the Fed waits to implement its normalization measures until the actual inflation rate hits the 2 percent target, it will have waited too long, and the target will overshoot over time. The conclusion from these arguments that the US economy is ripe for normalization and therefore that it would be appropriate for the Fed to raise rates and to start gradually reducing the size of its balance sheets.

While the US economy is ready for normalization, Europe and Japan are not yet ready to normalize their unconventional monetary policies. As a result, there will be a period of time in which monetary policies in the US, Europe, and Japan will differ from each other. These diverging paths are likely to be reflected in corresponding changes in exchange rates, whereby the US dollar appreciates relative to the other currencies. These changes are desirable. They reflect a healthy component of the international adjustment mechanism and should not be resisted.

Davis: You’ve given us a clear sense of the dangers of waiting too long and also a sense of why we might have waited too long. Let me ask, Raghu, for your sense of whether the Fed in particular has waited too long to begin normalizing monetary policy, and whether you agree about the cost of delayed normalization.

Rajan: On the cost of delayed normalization, I think we’re exactly on the same page. It’s hard to tell whether the Fed has waited too long or not, and it depends on the hidden problems. There is also a concern that is held to some extent by the Fed, which is that they don’t want abrupt adjustments from the market if there’s a sense that the Fed is tremendously behind the curve. If, for instance, we get close to full employment—if the shape of wage growth changes from the fairly steady 2.5 percent or so it has been for the last few months in the jobs report to something much steeper—you can imagine this changes expectations about monetary policy as well as asset prices quite dramatically. I don’t think the Fed wants to be in the situation where, effectively, its hand is forced by data. Broadly speaking, forget what’s happened in the past, exit is probably a reasonable thing now at a measured pace. That’s exactly what the Fed is doing. I would support the current process.

Davis: Tharman, I want to get your remarks, and perhaps you also want to address the potential cost of delayed normalization leading to unhealthy growth, and whether that’s a big concern or has anything to do with monetary policy conduct in the US, here in Asia, and in Europe.

Tharman: One way of thinking about this is that we’re in a new situation, because central-bank intervention has been necessary for much longer than it has been in previous episodes. And we’re learning some things about the interaction of central-bank policy and the supply side of the economy—and not just the demand side of the economy, not just the cyclical, but the structural and the long term. Or if you like, we’re learning something about the interaction between macroeconomic policy and microeconomic responses. And the longer we find ourselves in a highly activist mode of monetary policy, the more we have to think about how this balance of considerations between demand and supply, micro and macro, cyclical and structural has to shift.

When we talk about the cost of not normalizing, it’s really about what we’re seeing in allocative efficiency especially. The supply side is just not vibrant. In most of the advanced world, innovation is concentrated. There are some superstar firms, but the data are more or less telling the truth on productivity: you’re not seeing that spread of innovation across whole sectors and economies. And there’s evidence that the capital markets are not working very well in allocating capital to new shoots that deserve it, or established firms that are more innovative or efficient than others.

We’re in a situation where there are no rules for how we move from demand to supply-side considerations, cyclical to structural. We’ve got to apply some judgment, and the judgment at this time has to be based not on today’s data, or any precise data, but on the stylized facts of how policy is shaping economic responses, especially microeconomic responses. And based on these facts, the judgment is that if we delay for longer, we’re likely to get a negative spiral of weak productivity growth, weak middle-income wage growth, and the political economy that comes out of that.

What are the consequences of unconventional monetary policy?

Davis: Do you think there’s a link from this extended use of unconventional monetary policy to capital misallocation to slow productivity growth?

Tharman: There are two sides to this. There’s a way in which conventional monetary policy aimed at stimulating demand does aid productivity growth. It hampers hysteresis, and there are lots of studies showing that innovation works well where there’s some demand growth. But there’s a fair amount of study of the way capital markets have been working less efficiently, when you’ve had low yields overall. To take an analogy, there was a risk of a forest fire and we flooded the forest, which was right at that time. But when you flood the forest, everyone gets the same amount of water at the same cost. You’re not focusing on the new shoots and the new sectors that deserve more resources. So it doesn’t help to have the yield curve the way it is—it stimulates some entrepreneurial activity, but it’s more by way of acquisition than innovation. The data are clear on that. There has been a slowdown in innovation, in the translation of R&D into commercialization. There’s been a slowdown in the number of new firms, and a slowdown in job churn within the labor market, in particular, from low-productivity to higher-productivity sectors and firms. So the markets aren’t working very well. One has to be tentative about the link between all of that and monetary policy, but with some judgment, you do see some relationships.

Frenkel: I would like to follow up on Tharman’s remarks. When we read the daily financial press, we see that much of the discussion focuses on the short run. The debate is whether the Fed will raise interest rates in this month or the next month, and whether the hikes will be by 25 or 50 basis points. I would like to emphasize that from a long-term perspective, it really doesn’t matter. The important thing is to get clarity about the medium run. What will Fed rates be in, say, 2019? We know that interest rates today are so far from their long-term levels as judged by historical standards, and we also know that we have negative real interest rates in most of the main financial centers. We know that this cannot be an equilibrium. In this regard, it really doesn’t matter much whether convergence toward the more sustainable long-term levels occurs a few quarters sooner or later, as long as markets receive a clear indication of the medium run.

One of the costs of excessive reliance on monetary policy is that it enables governments to delay necessary fiscal and structural measures. In this regard, by being the only game in town, the central bank exposes itself to the risk of losing central-bank independence. By entering into the territory that should typically be dealt with by the fiscal authorities, the central bank may find itself vulnerable to the criticism that it entered into the political arena, which is of course a minefield that ultimately may result in a reduced ability to carry out independent monetary policy.

“The financial crisis essentially sent the message that people in power really don’t know what they’re doing and don’t know how to find a way out of this problem—and that bankers look after their own.”

Rajan: We are somewhat guided by the growth rates that we had before the financial crisis. Across the industrial world, there’s a sense that these were the right growth rates, and therefore a lot of the evaluation of what’s happened since asks: How far below are we from the trajectory that continued? Is it possible in some way to make it up, and can we grow again at those rates?

Jacob asked the right questions, which are basically: Was the growth rate ever sustainable? Does it make sense to try and go back to it? Should we reexamine the whole process of growth? A number of emerging markets have been forced to ask: Was the export-led growth before the crisis sustainable? What should we do to reform our system to substitute for the sources of growth that have gone away? This discussion is going on in industrial countries, but it almost surely implies that if you want to reenergize growth, supply-side reforms have to take place. What are those? How do you get the political conviction? And how do you make sure that it creates a redistribution of growth opportunities across the spectrum so that we have more buy-in for this kind of process? These are the things we should be focusing on, and monetary policy is of limited use in this whole discussion.

How sustainable is Chinese growth?

Davis: I want to pivot to a different part of the world. Obviously China’s long-term growth and underlying growth prospects have been declining for some time. That’s a source of concern—and there are concerns about the increased use of leverage in China and some other emerging market economies. I want to get your sense of the extent to which there are serious dangers lurking, and ask what policy makers around the world should do to prepare themselves for an abrupt slowdown in some of the major drivers of the world economy.

Tharman: As different as China is from the economies we’ve just been talking about, the way I would frame the issue is very similar. It’s really about the balance between macroeconomic policy, and in this case especially policies influencing the growth of credit, and structural policy. If you read the speeches of Chinese policy makers over the last year, you know that structural reforms are the most important strategy for the future. But there’s in fact a balance in policies that we’re seeing, and that balance is probably a matter of political economy and timing. They overdid the credit engines a year ago, and they’re now correcting. They don’t want to correct too fast or too hard, but if they don’t correct at all, there’s a very serious risk of medium-term disruption, which will have a cost that’s more than economic. I think they’re managing this, if you ask me, fairly well. As in every other society, policy making operates within a political economy context, and they’re managing it fairly well.

What’s the potential? I would say China is still an exception among emerging economies of being able to continue a transition from a middle-income—or, by World Bank standards, upper-middle-income—to a high-income country. China today is at about the same level of productivity—or, say, per capita income—as Korea and Taiwan were in the mid-1980s. And Korea and Taiwan in the subsequent 15 to 20 years saw growth rates of about 6 to 8 percent. In China’s case, the negative is that they don’t have the labor-force-growth component of that 6 to 8 percent growth anymore. There’s barely any labor-force growth in aggregate, so it has to come out of productivity growth, especially multifactor productivity growth. The good news is that China still has substantial opportunity to reap the gains that come from reallocation of resources— from very low productivity sectors to medium and high productivity sectors, and from the coastal regions to the inner regions, where levels of productivity and income are still much lower.

When you’re talking about a continental-scale economy such as China or India, internal convergence is a huge game. And China’s inland provinces today have about half the level of productivity of coastal provinces such as Jiangsu. And there are some populous inner provinces such as Sichuan, and smaller ones such as Guizhou, where it’s well below one-third the level of productivity of Jiangsu. As they build infrastructure and as Chinese manufacturing, even with existing technologies, moves inwards, there’s scope for traditional structural change or reallocation of resources from low-productivity to high-productivity activities that’s going to give them a substantial kick in the next 15 to 20 years. So I’m an optimist for that reason, but it’s going to require policy changes, and changes especially to improve the allocation of capital. That’s a political economy question. They know what to do. They’re timing it carefully, and we’ll have to watch what happens after the 19th Party Congress.

Frenkel: China is still saving about 46–48 percent of its income. Whatever misallocation China may have, the country is bound to be an accumulator of assets. When we talk about the Chinese economy and its unique performance, it should be obvious that we witness a megatrend rather than an episodic performance. About 30 years ago, when I was the economic counselor and head of research of the International Monetary Fund, the Chinese economy was insignificant and was under the radar screen. In 1990, it produced about 4 percent of world output, which is very different from its current role.

Currently, in 2017, China produces about a fifth of world output—a fundamental change in the center of gravity of the economic world. China has also become a critical player in world trade, reflecting the shift in the center of gravity toward Asia. The volume of trade between China and the rest of Asia is about double the size of the volume of trade between China and the US and Europe taken together. Furthermore, China has become the most important trading partners of most of the major industrial countries in the world. For example, by now between one-fifth and one-quarter of US and European exports go to China.

Where am I concerned? The main lesson from the recent global financial crisis is that we must ensure a robust banking system. When it comes to China, it is important that shadow-banking activities are illuminated and taken out of the shadow so as to be transparent. Furthermore, it is important that nonperforming loans are handled properly. Notwithstanding these considerations, I am still positive about the role that China plays and can play in the global economic system. The recent initiative of China in launching the Asian Infrastructure Investment Bank (AIIB) is a positive and encouraging development. It is encouraging that most of the major countries have joined this new multilateral bank, and I can only hope that the US and Japan will also join before too long. Generally speaking, China should be granted its appropriate place in the formal international architecture and its quota shares in the multilateral organizations should be commensurate with its economic size.

Rajan: What happens in China is one of the most important issues that should concern us for the next few years, because what happens doesn’t stay in China, it goes to the rest of the world. The questions surround the tremendous transformation that China is attempting. They have already had some success in enhancing domestic demand and increasing domestic consumption, in moving away from manufacturing to services. They’re moving away from export-led to high-tech manufacturing. One of the things they’re trying to do, which is extremely important, is move away from government management toward a more market-dominated economy. And in attempting to do that, they have one serious constraint, which Tharman alluded to, but by not having an official position, I can be more explicit about: they’re trying to make this change even while the Communist Party dominates the economy.

In order to make this change successfully, the Communist Party cannot dominate through authoritarian means; it has to dominate based on public support. And this requires at least two things. One, the Communist Party has to maintain the reputation for economic management that it has built up, which means it has to keep growth going, prevent financial crisis, clean up the financial system, and all that good stuff that we have grown to expect. Two, it has to reduce the amount of corruption in the system, which is why there’s an anticorruption campaign, so that there’s some sense of legitimacy. The problem is that these things imply major trade-offs, and success implies getting these trade-offs right.

For example, last year China had to loosen the taps, get stimulus going, and get the Chinese economy moving fast. But again and again, you find that as you loosen the taps, the economy goes much faster than you expect. And as it suddenly expands, there’s a financial-sector component, which is an interesting part. The Chinese economy is growing old fairly fast. People don’t feel they have adequate savings. This means that as soon as they see opportunity for high-return investment, they run into it. So every sort of loosening of the taps is also accompanied by financial flows into various sectors. The government would like to allow the markets to take control, but it sees that every time it opens the taps, the markets go wild. So it has to restrain, and it probably overcorrects. It is a yo-yo process, and the government hopes very much that it gets it right. It has gotten it right so far, but there is always the possibility that it won’t get it right enough.

Every time it does this, the underlying economic structure also changes, most particularly with the extent of debt that is building up. One of the worries over the short to medium run is, does the increase in debt that has built up, especially in the nonofficial sector, come back to haunt China in this stop-go process? Does it convert what is a slowdown into a meltdown? Nobody has really been right betting against China, but it is a possible source of uncertainty.

Davis: To sum it up, I hear guarded optimism about China’s prospects from you three, but you’ve also pointed out some serious risk that Chinese policy makers need to contend with.

What are the roots of populist politics?

Davis: In the US and Europe, we’ve seen the rise of populist political pressures. And I want to get your sense of, why now? Why in the last few years? Why in some of the advanced economies but not in others that also have disappointing growth performance? And finally, how can policy makers productively respond to these populist political pressures that we’ve seen emerge so strongly in the US, in the United Kingdom, in France, and in a few other countries?

Rajan: The roots of current populist pressures and forces aren’t short term. They have been building up over a long period, and I would argue that it goes back to the technological revolution. We’ve seen automation on the back of increasing globalization. And as many have recognized, this has created pressure on middle-class jobs and middle-class incomes because routine jobs have been outsourced or automated. But that’s been going on for a long time. In fact, when you look at US manufacturing numbers, you see that in 1943, manufacturing reached its peak, at 38.5 percent of the labor force. Since then, it has fallen steadily to where it is now, at 8.5 percent of the labor force. Why the sudden angst about manufacturing?

I would say two things. First, it’s hard to rail against technology. In the short run, it is a problem in terms of job losses. Trade, on the other hand, tends to hit entire establishments. You have to close down because you’re not competitive anymore, and you move the jobs to Mexico or to China. And the manufacturing establishment is much more localized in semirural, semi-urban areas. To that extent, the damage to the local community is significantly greater. Nowadays everybody wants to travel to the “flyover states” in the US and see what’s going on. The damage has been considerable in town after town with little economic activity. The effects of trade are much more concentrated than the effects, more broadly, of automation.

“We just haven’t had enough spread of new technologies to be able to lift productivity, and thereby lift that broad middle in the wage ladder.”

—Tharman Shanmugaratnam

Second, the financial crisis essentially sent the message that people in power really don’t know what they’re doing and don’t know how to find a way out of this problem—and that bankers look after their own. When they got into trouble, nobody went to jail, and many of them went back to earning bonuses while Main Street basically got nothing. It’s a damaging picture of the market economy. When this comes on top of the fact that people are suffering, it causes people to say the system is broken, and they start moving against it.

Why didn’t this happen in Scandinavia? Well, Scandinavia and many countries have somewhat more flexible systems. Because they’ve been so open to trade in the past and because they’re small, they can deal with some of the problems of trade dislocation. Japan has not been subjected as much to the kind of localized dislocation. It’s more a steady decline. But what’s happened in rural or semi-urban America, what’s happening in places around Paris, what’s happening in places outside London, is that they’ve borne the brunt a little more.

Because the impact has been localized, while political authority is more concentrated in the areas that have actually benefited from globalization—whether it’s London, Washington, New York, San Francisco, or Paris—we’ve been able to overlook the building anger. The anger had not been voiced until the disgust with the system reached a peak and led to the popular support of politicians such as Donald Trump and Marine Le Pen. And then we recognized, ah, there’s a problem, and now everybody is finding out and trying to see what people in Granite City, Illinois, think. And that’s a good thing. It will make us focus on the neglected, the forgotten man, so to speak.

Davis: So, it’s a good thing to recognize the problem, but it took us too long. Jacob, can I get your thoughts on how policy makers can productively respond to these populist political pressures?

Frenkel: At the World Economic Forum held in Davos, Switzerland, there is a session organized by public relations expert Richard Edelman, who for several years has developed what he calls the “trust factor” or the “trust index.” This year’s results were rather dramatic. The lowest trust was given to the media and, of course, as in the past, to bankers and politicians. It is clear that people have lost their trust in what they read or listen to in the media. Everyone has his own truth. The truth that people believe in is typically the one that is shared by the group of people that they are associated with, and typically people associate themselves with others who hold similar beliefs and values. This is the way groups are formed in the social media world. Individuals who belong to a WhatsApp group or WeChat group or other groups within Facebook typically share the same beliefs and values. By talking to each other, they define their collective “truth.” At the very same time, the same happens in other groups that define their collective truth. Under these circumstances, it is typical that election results are not easy to forecast, since every group believes that everyone thinks like them because these are the views that they encounter within their own group. No wonder that most of the recent forecasting of election results failed dramatically.

One of the unfortunate results of the recent financial crisis has been the perception that “experts have failed.” As a result, being an expert is not viewed as an asset but rather as a liability. Likewise, having experienced persons is viewed as a liability rather than an asset. This has resulted in elections of inexperienced candidates and a rise in the prominence of outsiders, a phenomenon that, in the complex world of today, should be viewed with concern, especially as it has been accompanied by populism and antitrade sentiments.

The recent popularity of protectionism is a case in point. Most people who support protectionism would clearly prefer not to give up the gains from trade. They justify their protectionist stance on the grounds that occasionally opening to international trade may inflict hardship on some segments of the population. It is important to emphasize that the appropriate way to address this challenge is through fiscal measures such as trade-adjustment assistance, retraining programs, or appropriately budgeted safety nets to support the weakest segments of the population. The problem does not arise from trade but from failure to enact the appropriate fiscal measures. It will be tragic to forego the gains from trade just because governments are unable or unwilling to implement the appropriate policy measures that simultaneously secure the gains from trade while reducing the hardship that may accompany it.

Since the 1980s, in the industrial countries, the share of labor in GDP has been on the decline, which has been correlated with the rise in income inequality. And those at the higher end of the income distribution are also the owners of capital, so when the share of capital rises, it is associated with a rise in income inequality—but that, again, has nothing to do with trade or with technological advance. This implies that we need to focus on why the share of labor has declined, rather than on adopting a populist stance.

Tharman: There are two interesting, stylized facts coming out of the sociological studies of what’s been happening. The first is that people tend to extrapolate from what happened in the past in forming their expectations of the future. And when those extrapolations don’t materialize, particularly if they’re about rising incomes, they face grave disappointment. There have been lots of studies on this in the US as well as in parts of Europe. It was assumed that what happened to incomes during their parents’ time would carry on in their own lifetimes, and it didn’t. The white working class, in particular, has seen stagnation in incomes to a remarkable degree. If we talk about the Rust Belt states, more than 50 percent of people in their early 30s are earning less than their fathers did at the same age. That’s remarkable stagnation. But the source of disappointment isn’t simply in the fact of wage stagnation; it’s that there’s a deficit compared to what was expected.

The second finding is that people tend to weigh losses more than gains. Some people have been making gains. There’s been steady progress among blacks and Hispanics in the US, and, over a long period, among women. But these gains aren’t weighed as heavily as the losses faced by white males and the white working class generally.

When you combine those two stylized facts, they’re potent. Now, if that’s the problem, what’s the solution? The aim should be to get back onto a path of some form of income mobility over time—within realistic possibilities—and it has to involve a reigniting of productivity growth, which is what we’ve been talking about. And if that’s the aim, then I think we haven’t had enough technological change, enough automation or robotization, in a whole range of sectors.

The data speaks for itself. We know of that widening divergence between the leading firms and the rest, sector by sector. We just haven’t had enough spread of new technologies to be able to lift productivity, and thereby lift that broad middle in the wage ladder. There’s been some job loss, but you haven’t had that much job churn, and that’s a problem.

But that brings us to the second part of the solution: it takes proactive strategies. Help everyone losing their job get another one. Help them reskill and upskill. So, in the countries you mentioned, Sweden and Switzerland—or since we are on s’s, I’ll throw in Singapore—the basic motif in public policy is about reskilling and upskilling. It’s what my colleagues and I spend time on: finding the best way to close skills mismatches through traditional educational institutions or new and often untested programs.

If we can do that, spur more use of new technologies, especially in the services sectors that are ripe for innovation, and work proactively with everyone who loses their job to help them reskill and get back into a better job, we’ll not only cope reasonably well with this new era of globalization and technological change, we could break those stylized facts that I talked about.

Rajan: Different countries have different views on what makes more sense. Some countries are trying to prepare their labor force for the future, and there’s a lot to gain from doing that. Others seem to be a little more pessimistic about this and are focusing on redistribution or universal basic income as the solution. Those who are focusing on universal basic income certainly are reacting to the social cost of having unemployed people in the short run, but potentially they are giving up too quickly on the possibility of preparing workers for new, potential jobs, and may be creating a trap. If you give up too quickly and say, “We just have to pay people off to do nothing,” you may entrench a kind of structure that is hard to get out of. On the other hand, implementing the kind of training that small countries such as Singapore and Sweden are able to provide may not be that easy for a large country, so there is a question of what the transition path is and how you navigate that.

Davis: We need a mission from Singapore to the US to learn how to do upskilling and reskilling for older workers, because we don’t do it effectively in the US right now.

Tharman: It has to be said that there are some cities and metropolitan areas in the US that do it much better than others. And both in the US and elsewhere, the most innovative cities, the places that have done best to create new jobs to replace old jobs, tend to be those with a strong sense of partnership—between the mayor; educational institutions including community colleges, technical colleges, employers; and often the unions. It is not about whether the mayor is on the left or right, a Democrat or Republican, but about people feeling they are part of a community of practice. There’s a way in which the social compact enthuses everyone.