Interview: Former Indian Central Bank Chief on a Global Economy in Flux

By Wang Liwei

With the global manufacturing landscape undergoing seismic shifts, both China and India’s factory sectors are changing fast. To make sense of these developments, not many are as well positioned as Raghuram Rajan, a former governor of the Reserve Bank of India, who also served as chief economist of the International Monetary Fund from 2003 to 2006.

In a wide-ranging interview with Caixin, Rajan addresses the slow relocation of manufacturing to India, expresses skepticism about reshoring efforts in the U.S., and delves into China’s pivot toward a service-led economy. He also assesses the potential for technological competition and cooperation between India and China, offers a measured take on cryptocurrency regulation, and provides a cautious analysis of artificial intelligence’s (AI) effect on productivity.

Now a finance professor at the University of Chicago Booth School of Business, Rajan unpacks these global changes, providing a clear-eyed analysis of the risks and opportunities in a rapidly changing world.

Caixin: From your vantage point, what has been the progress of relocating manufacturing to India over the last one or two years?

Rajan: I think there’s been some relocation, and Indian manufacturers started by assembling the final product. I think the ambition is to also manufacture more components over time so as to increase the value added. Of course, this is a slow process and India needs to learn from China. And in terms of some of the manufacturing techniques, etc., Chinese engineers have some capabilities that India needs to acquire. But of course, it has to be a win-win situation for both. And I think we need to see how that can happen.

You have lived in the U.S. for many years. We have recently seen lots of layoffs in the tech sector. The most pronounced layoffs have been in tech companies and finance. Why is that?

Some of it is because we’re seeing some productivity improvements coming in from AI, but it can work both ways, right? If you can program in a cheaper way, there’s more demand for programming. And some of this is also because the tech sector has hired a lot in recent years and is taking the opportunity for a breather. That said, I think that going forward, there will be lots of opportunities to apply AI to firms in their operations. We haven’t even begun to touch that. And that would mean more engineers, software engineers, etc., to do that work. It would also mean more data scientists to collect the data and get it into shape for AI to use, and so on. So I think it’s early days yet.

We have recently seen companies like Apple and Micron pledging to invest billions of U.S. dollars in the U.S. Do you think this policy of attracting manufacturing back to the U.S. will succeed? Earlier we sort of all dismissed it, given the personnel shortage, but there are increasingly many companies making these commitments.

Remember that putting in an investment takes years, many years. So you can announce an investment, but when do you actually know that the investment has happened — maybe five, 10 years down the line. As you know, there are some companies which make a habit of announcing investments, and nothing actually happens. So I think one has to be a little cautious about investments that are being announced, whether in fact they will happen and what the timetable is. And to the extent that the investment is being announced just for political reasons, when the administration changes, they might find it not so warranted to make the investment.

In an environment of greater protectionism, there will be a rationale to manufacture more within the United States to save on tariffs. Whether it’s going to be a huge amount — I think not necessarily — because labor costs in the United States are also relatively high. That’s one reason why a lot of manufacturing left the U.S. So, it may be that at the margin, what comes back might be more automated manufacturing, where you don’t have to employ so much labor, and some of the U.S. ingenuity in making sophisticated production chains might be beneficial.

So we have to see. I think the first thing that manufacturers want is the volatility around tariffs to come down. Because today you don’t know what the tariffs are going to be on Chinese goods. It’s up in the air. And some of these tariffs applied to countries are also changing. So I think we need a more stable tariff regime. Then companies can see what I need to bring in, and what I can manufacture outside. At that point, we will see how much manufacturing comes back to the United States. But I would expect that much of the manufacturing that comes back is going to be much more automated, so we’re not going to see a lot more new jobs being created.

China has traditionally emphasized manufacturing, right up to this very day. But on the margin, it’s putting more emphasis on service sector growth. How important is the development of services? What are some cases where this could be really helpful for growth and employment?

You can see this with DeepSeek, which has made a big impression in China. It’s a service business and it has had a large impact. If you look at employment in industrial countries, it’s about 70% in services, maybe 10% in manufacturing, and 10-15% across the rest. So, I think countries move naturally toward a more service-oriented economy. And employment in manufacturing becomes more productive, but also becomes a smaller and smaller fraction of the economy for a while. I don’t think China is different. I think over time its service sector will increase substantially.

Now the big question is to what extent Chinese service exports will take off. And that’s an important question that the Chinese authorities must be asking. There are some issues. Because services are much more data-oriented nowadays, there are data privacy issues. There are issues around the extent to which the government can intervene in some of the data and acquire the data that companies get. So, many emerging markets and developing countries need to think about how they can facilitate service exports, knowing that issues like data privacy are really important for other countries. For people who are buying these services, they want to have confidence that the data will not be widely misused. So services bring a whole new set of issues that need to be dealt with.

India, having good engineers, is doing very well in some of the consulting and IT service areas. China now has many good AI researchers. In Silicon Valley, many of the top scientists or engineers at tech giants are of Chinese origin. Do you think there will be any competition between China and India in this area, or that there will be different areas that they excel at?

That’s possible. I hope that’s how it works out. I do think that China is ahead of India in improving the quality of its universities. And, you’ve got some fantastic universities like Tsinghua, with some technical universities comparable with the best in the world. But I’m hopeful that these two very large Asian countries can work together as opposed to working against each other. And there are many areas where they don’t need to go head-to-head.

A question about crypto and stablecoins: You recently mentioned that the practical way to regulate them shouldn’t be to ban them, but to contain them. Are there any successful cases or practices from other countries that come to mind?

Singapore has been quite reasonable on crypto regulation. It’s trying to give them some room to experiment, some room to play, so to speak, without being overwhelmed by mass participation in a market that you don’t fully understand. So there are places that have taken a middle road. A lot of innovation in the financial sector in Singapore has to first be tried out on a small scale. And then the regulators will decide whether you can roll it out in a bigger way. That may be an approach that we should think about. Of course, the problem is sometimes until it’s fully rolled out, you don’t know what the problems are. And sometimes you don’t even know the benefits until it’s fully rolled out. So I would be instinctively averse to banning, but you’re right that managing a new product without it exploding is also not that easy.

Back in 2023, you mentioned that U.S. productivity was not up much. How’s the situation right now in terms of productivity figures, and is that mainly from AI’s impact or other factors?

Productivity has been reasonably good in the last few quarters. Now, is that AI? It’s hard to say. I mean, lots of things are called AI — and some forms of automation also sometimes fall under the term. I think part of what has also been happening is something we talked about earlier — because of the cutdown in immigration, the supply of labor has also grown more slowly, which sometimes may mean that people are holding on to workers and don’t want to let go. That then would tend to diminish productivity. The fact that productivity has been reasonable would suggest maybe something is happening, but you won’t really know until you look back three or four years down the line.

This interview has been edited for length and clarity.

Contact reporter Wang Liwei (liweiwang@caixin) and editor Michael Bellart (michaelbellart@caixin.com)

caixinglobal.com is the English-language online news portal of Chinese financial and business news media group Caixin. Global Neighbours is authorized to reprint this article.

Image: Pixels Hunter – stock.adobe.com