Tracking Cross-Border Capital Flows in Asia

Welcome to Pacific Exchanges, a podcast from the Federal Reserve Bank of San Francisco. I’m Alex Altman.

And I’m Paul Tierno. We’re analysts in the Country Analysis Unit, where we monitor financial and economic developments in Asia. Today, we continue our ongoing series Rethinking Asia, with a deep dive into capital flows in the region. We spoke with Brad Setser, a Senior Fellow for International Economics at the Council on Foreign Relations and a former deputy assistant secretary for international economic analysis at the U.S. Treasury.

Brad walks us through the major trends happening in capital flows across Asia, diving into which countries are attracting stickier foreign direct investment and which are subject to more flighty bank and portfolio flows. He also provides some great insights into capital moving in and out of China post-Global Financial Crisis, and how these flows have affected economic growth and policies domestically.

Yeah, he also offered some interesting perspective on the effects of the ongoing U.S.-China trade dispute. Compared to the renminbi sell off in 2015 and the impact on global markets, he suggests that spillover effects from the tariff war could be limited. But, let’s hear it straight from Brad.

Hi, Brad. Thanks so much for joining us today.

It’s my pleasure. Thanks so much for having me on.

Let’s start today with the 10,000-foot view. After the global financial crisis in 2008, Asia, and China in particular, received pretty sizeable capital inflows. In the past several years, however, this trend has reversed somewhat, and money has started flowing out of some of the Asian economies, again, I’m thinking China. There’s also some sense that tighter liquidity conditions are playing a part as the Fed here in the U.S. has started raising interest rates. Could you give us your sense of what’s going on with capital flows in Asia today?

Sure. I mean, I guess I would start by noting that throughout the post crisis period, Asia has been a net exporter of capital. In general, there’s been an outflow, or rather the outflows have exceeded the inflows. Certainly for China, and the period immediately after the Global Financial Crisis, inflows into China just added to China’s reserves, because China was still running a sizeable current account surplus. But there was no doubt a period of time when China did attract substantial inflows. That was, I think, a function in part of a decision on China’s part to liberalize its financial account, and allow more cross border borrowing. Then, also a function of interest rate differentials. The U.S. had also had relatively low interest rates when the money was really moving into China. And, in part, a product of expected currency moves. Up until 2014, the general expectation was that China’s yuan would appreciate and get stronger over time.

So, people were looking for ways to get into China. That really changed over the course of late 2014 and 2015. I think it changed in part because of developments inside China. China’s own economy had started to slow around then. Xi had launched his anti-corruption campaign, which sort of had as its by-product a sense on the part of some wealthy Chinese that their assets in China might not be as safe as they’d hoped. There was some pick up in hot outflows then. But above all, it was a function of the change in the dollar. The dollar really appreciated quite substantially against the euro and the yen in 2014. The yuan was, at the time, stable against the dollar and moved up with the dollar. All of a sudden, China’s currency looked stronger than some of its fundamentals, at least, warranted. I think that’s when you started to see money move out. Then the outflows really picked up as China made a decision, take your pick, to allow more flexibility, or to allow modest depreciation. You saw quite substantial outflows in 2015, latter part of 2016.

Then over the past couple of years, things, rather remarkably, have stabilized. Partially because China’s currency hasn’t been moving much. Partially because China tightened the screws and restricted outflows. In part, until very recently, because there was a sense that China’s economy had recovered a little bit from the slump in 2014.

You mentioned some policy shifts by the Chinese government, as well as some things happening in global liquidity, and how that’s affected flows into and out of China, but can you talk about it for Asia more broadly as well?

Well, the basic trend that you saw with China applies to many of China’s nearest neighbors. The biggest current account surpluses in Asia, as a share of GDP – so the biggest net capital exporters, are Korea, Taiwan, Singapore, now Thailand. Back when the dollar was generally globally on the weak side, so before 2014, they were uniformly intervening pretty heavily to keep their currencies from appreciating. The bulk of the financial outflow that is associated with large sustained current account surpluses was coming through the build-up of their reserves, and in some cases, through outflows tied to their sovereign wealth funds and their sovereign pension funds. Taiwan is a little different because Taiwan has let its insurance companies run wild. The insurers have been generating the majority of the outflow. After the dollar appreciated, pressure hasn’t been as consistently one sided for those countries. They still have surpluses, but private outflows have come closer to offsetting those surpluses. Outflows through Korean insurers, for example. You don’t necessarily see consistent intervention, but unlike in China where you really don’t see large scale intervention except to fight depreciation pressures after 2014. In some of these other Asian economies, there have been bouts of intervention to resist appreciation. It kind of makes sense when they’re running bigger current account surpluses. The structural pressures on their economy are a little different.

Then, I would say Japan is a category all of its own. Japan has had zero interest rates throughout. With the launch of quantitative and qualitative easing, and then its various intensification phases, you’ve seen significant outflows from private Japanese investors basically looking for yield abroad.

That’s really helpful in terms of giving us some good context of who the major players are in terms of countries. I’m wondering if we can go a little bit deeper and talk about any changes you see in the composition of capital flows. Any interesting trends in direct investment versus portfolio flows?

Well, on the inward side, I think portfolio inflows into Asia kind of have followed global trends– to Asia ex-China. When investors have wanted to hold emerging market debt, that generally has extended to wanting to hold the debt of a lot of emerging Asian economies. Indonesia, for example has at times seen fairly strong inflows into fix income markets, so portfolio inflows. I would say in general, after the crisis, there’s been stronger portfolio inflows than bank flows, in part because regulations on the banking side were tightened, and so a little harder for banks in Asia to directly take on as much foreign currency and short-term debt as some did, before the crisis. I think the exception there was China.

China, for reasons that I don’t fully understand, after the crisis selectively liberalized its financial account and essentially opened up to short-term banking flows before it opened up to portfolio flows. So just before China’s depreciation in 2015, there had actually been a really big rise in short-term bank borrowing, so-called carry trade. And the reversal of that carry trade was a big reason why China’s reserves fell so dramatically after the depreciation. Chinese firms, some wealthy Chinese individuals were borrowing in foreign currency, moving that money into the Chinese yuan, because, at the time, there were higher interest rates in China than outside of China, and they weren’t prepared for the risk of a depreciation. In the past couple of years, by contrast, China has tightened down, clamped down a bit on the banking flows and opened up to portfolio inflows. I think in the past year and a half, there’s really been quite strong portfolio flows into China. Some of that was reserve managers looking to hold Chinese yuan as part of their reserve portfolio.

Some of that is your traditional institutional investors who added Chinese exposure in anticipation of China’s index inclusion.

Brad, I wanted to unpack some of what you just said. Right after the crisis there was a short-term liberalization of short-term bank flows, then which recently was converted into stronger portfolio flows, and then you also referenced bond market inflows when there is more appetite for emerging markets, and you cited Indonesia. I’m wondering, for the benefit of the audience, can you spend a minute talking about how these different type of capital flows affect recipient countries. For example, are some types stickier than others? I’m thinking about the contrast between foreign direct investment and portfolio flows, and if you can talk a little bit about that relationship.

Foreign direct investment typically is a commitment. You’re buying a company, you’re building a factory, and it’s hard to unwind that investment quickly. You’re committed, the money is there, it’s going to stay there. You can try to hedge or otherwise get insurance against some of the risk that you face. But once you’ve made a big investment from the investor’s point of view, you’re stuck, and from the recipient’s point of view, the money isn’t likely to flee.

I think that can be contrasted with short-term bank flows, which are notoriously prone to reversal. Money flows in, money flows out. You take on a short-term loan, you’re committed to pay that back when it comes due. If there’s a big shift in market conditions or there’s some crisis inside your country, that money can flee quite quickly. We’ve seen that. We saw that in the Global Financial Crisis: the Korean banks, for complicated reasons, had actually borrowed a lot of short-term money in 06-07. That was a big reason why Korea came under such pressure in 2008.

Then we saw that with China. China didn’t liberalize immediately after the Global Financial Crisis, but they started in 2012.It kind of began to experiment with loosening up its financial account perhaps, because it wanted a bigger international role for its currency. It allowed more cross border borrowing. You saw half trillion go in in a very short period of time through banking channels. Then you saw about the same amount leave immediately after the depreciation. Portfolio inflows generally are a little less prone to reversals than banking flows. If you buy a bond, an Indonesian bond, or Chinese bond, in order to get out, you either have to find someone else to sell it to, or you have to wait until the bond matures. You can get out, but in order to get out, you have to find a willing buyer. It’s a little bit more difficult. Sometimes you have to accept a reduction in price. There’s a risk of reversal but it’s a little bit more sticky, and a little less risky from the borrower’s point of view than short-term interbank borrowing.

If we can step back for a moment to some of these structural changes, you were talking about Chinese banks borrowing to invest abroad, and participating in the carry trade. Last year, you mentioned in your blog, Follow The Money, that Chinese banks are increasingly borrowing abroad to finance foreign lending. Do you see that happening elsewhere in Asia in terms of changing foreign currency lending base, and what risks do you see with offshoring that funding base?

Well, I mean, it’s a particularly pronounced shift for China, because China, until say five years ago, China was looking to finance its offshore lending out of its own resources. It had too many reserves. It was looking to shift in some sense some of those reserves over to the banks, and have the banks do the lending, so that the central bank wouldn’t need to buy so many Treasury bonds. It is a little bit of a change to have the Chinese banks actively borrowing so that they can lend. But the scale, I would say, is still fairly modest relative to the size of China’s economy. I would characterize it as an interesting change. But not something that as of now, is really giving rise to any major sets of financial risk.

Some of the other Asian banking systems have been borrowing more abroad, like say, the Japanese banks. I think that’s, in large part, because they need what people in the market call dollar funding. They need to raise dollars. They’re intrinsically able to raise yen, because they’re Japanese banks, they collect yen deposits. But if they collect yen deposits, you have to lend yen. You don’t want to run a currency mismatch. You need to lend out yen, and there just isn’t a lot of demands for yen loans, and so Japanese companies don’t need to borrow much Japanese companies are cash rich. The interest rate is very low on yen lending. The difference between a Japanese bank, well it doesn’t really pay anything on yen deposits, but it doesn’t really make anything on yen loans.

In order to make money, a lot of the Japanese banks have been growing their dollar balance sheet. They need dollar funding, and you can get dollar funding by going out in the market and directly borrowing dollars in various forms, so that you can buy, either make a loan or buy U.S. securities, or you can enter into what’s called a cross currency swap and you basically exchange yen for dollars, and promise to reverse that trade in the future. In the process, you get dollars which you can invest. Japanese banks have been doing that, in some sense because they need to raise dollars, so they can lend dollars to make money when they can’t make money, or at least not as much money as they would like, on traditional intermediation in yen.

I think one of the things you see throughout Asia, actually, is this increase in purchases of offshore assets from financial institutions. Be it banks in Japan, or the life insurers in Japan, or the life insurers in Taiwan, or the life insurers in Korea that can’t get the yield they want at home, and so they’ve been looking towards the U.S., particularly after the Fed started raising rates. Although that’s complicated because of the slope of the curve type consideration. As long as there’s a difference between the return you can get in the U.S., and the low return they were getting at home, and as long as they could borrow dollars short-term at a lower rate, then they can get on their longer term lending or their longer term security purchases, they had an incentive to buy U.S. assets. That’s the big swing, I think, over the past five years, has been the rise of these private flows. As a result, Asia has been buying more corporate debt and fewer Treasuries than in the past.

You brought up insurers a couple of times. In terms of Korean insurers, and others moving abroad and investing in U.S. dollar assets, are they hedging that risk, or is this just looking for a higher return in a different currency?

The conventional financial market wisdom is that the Korean insurers largely hedge. There’s always an exception so it wouldn’t surprise me if there’s a Korean insurer or two that isn’t one hundred percent hedged at all points in time. But the general perception is that the Korean insurers do hedge, but sometimes they’re pretty adventurous in the kind of corporate debt that they’re willing to buy.

The Taiwanese insurers, they’re really big these days. They only partially hedge. That’s well-known. Their regulator has essentially not required full matching of their books. That’s encouraged them to buy more dollars than otherwise would be the case. Then the Taiwanese insurers, rather strangely, have been in some sense looking to raise dollars inside Taiwan by selling more dollar-denominated life insurance policies. I would say, in Taiwan, it’s looking for yield abroad, un-hedged. It is finding creative ways of passing the risk on to domestic policy holders. Then, the Taiwanese insurers because the volume, and the size of their dollar book is so big, they’ve also been pretty active in the hedging markets.

You’ve contextualized a lot of the Asian economies in terms of their current account position. In Treasury’s words, there are number of large and persistent current account surplus economies in the region. This, inevitably, brings us to the ongoing trade negotiations. How do you see the trade dispute affecting capital flows and are these changes going to prove lasting?

I mean, I think if you wanted to say, make a big picture point, I think relative to, say, the disruption that came with China’s depreciation in 2015, and then the fallout from that over the next year, the disruption that has come with the trade war has actually been much more modest. The real shock was 2015. Then, you really saw a sustained change in the overall composition and pattern of flows. Before then, pretty much China and all the others were fairly consistently intervening to avoid appreciation. After the Chinese depreciation in 2015, I would say that in general, some countries have had to intervene less, and China has been intervening on the other side. The trade war didn’t radically change that picture, China hasn’t had to burn through a lot of reserves during the trade war. But what you did see was during the period over the summer when tariffs were put on by the US, there was a depreciation in the yuan. That was echoed throughout the region, other regional currencies came under pressure.

Since the summit in Buenos Aires, and as a sense that the U.S. and China were working towards some kind of deal, I think you’ve seen a general stabilization. The yuan has started to appreciate a little bit. Some of the outflow pressure on China looks like it has faded. I think some of the countries that previously were depreciating in sympathy with China, no longer are under pressure to depreciate further. Some may be starting to see inflow pressure, and pressure to appreciate. But the general story is, in some sense, the dollar move in 2014, the initiation of the Fed rate cycle, and then the Chinese decision in 2015 that they couldn’t sustain the yuan’s value against the dollar after the dollar had appreciated, that was a much bigger shock on the financial side than the trade war.

On this discussion of the trade dispute, some multinationals may feel pressure to reorganize or reroute their global supply chains. Theoretically, would a negative outcome in this trade dispute affect FDI flows to some Southeast Asian nations that might become the substitute in the global supply chain?

I guess the first point is that FDI into China has been significant, but it is not an enormous inflow relative to China’s economy. China’s big, so even if China’s attracting just a normal amount of foreign direct investment, one percent or two percent of a big GDP is a big number. But there’s been actually, pretty good evidence in the latest IMF analysis, that China’s actually now, as opposed to the period immediately after WTO accession, attracting a little less FDI than a typical emerging economy. Just seems like a lot because China’s really big. Now, the question that came up on the trade war was, at its most intense, was sort of whether or not some of the multi-national companies that had centered their global supply chains, so building electronics for the world, building small appliances for the world, manufacturing clothes- whether some of those companies would look to Vietnam and others to replace China. The evidence that they are doing so to date is, I would say, modest.

Given the desire to have more diversity and resilience in your supply chains, some countries, and frankly, given that China is no longer as cheap as it once was, some Southeast Asian countries could start to receive bigger FDI inflows. Then, in some sense, they face an important choice. Vietnam has a current account surplus. If Vietnam receives large inflows of foreign direct investment, should it allow its currency to appreciate and start running current account deficits, or should it resist appreciation and add to its reserves? I think with stickier FDI, the risks are smaller, so some countries in Southeast Asia, if this shift materializes more strongly, could face pressure to make a decision about whether they want to maintain their current exchange rate regimes.

You’ve given us a really great lay of the land in terms of economic fundamentals in these countries, monetary policy, trade policy and how it’s affecting capital flows. If I can just ask one last question, what do you think are the biggest financial vulnerabilities? Where are you keeping an eye on, and what should we be paying attention to?

If economic data started to suggest that the Fed wouldn’t remain on pause, and would start, and expectations of a renewed hiking cycle would materialize. Obviously, that’s the really big risk. That would catch people off guard now. In terms of just Asia and some of the patterns of flows in Asia, I think there’s two things that I’m really watching. One is that China really slowed at the tail end of last year. The first quarter by every account is also going to be very weak. But there’s an expectation that China’s economy on the back of some new easing measures, new stimulus, will pick up over the course of the year. As a result, China’s currency hasn’t reflected the weakening of China’s economy. In some sense, you could argue that the yuan is now pricing in a little bit of a recovery, which creates a risk. A risk is that China’s economy is weaker than expected, even with the trade deal and that pressure on China to depreciate to get more support from exports returns. I think that’s one risk.

I think the other risk gets more technical. But a lot of the Asian insurers have financed their purchases of longer-term U.S. debt. Whether it’s collateralized loan obligations, which have more exposure to the short-term rates, or whether it’s long term fixed rate bonds, corporate bonds, but those investments have been financed by borrowing short-term or by hedging in the cross currency swap market, which is functionally very similar to borrowing short-term. When you’re borrowing short-term to buy longer dated securities, the difference between the short-term borrowing rate and the return on your long-term securities becomes important. If short rates keep rising and long rates don’t, some of these investments would be under water. I’m certainly watching that set of risk.

We’re looking at across the insurers complex in Asia well over a trillion dollars in foreign securities holdings, funded largely with various kinds of short-term money. Throw in the Japanese banks, you’re up to two trillion. There’s a real underlying vulnerability because of so much reliance on short-term funding and in some cases wholesale funding. That is mitigated by the size of the reserves in all these surplus countries, by the back stop provided by the potential availability of Fed swaps to the Bank of Japan, if liquidity pressures on the dollar side were ever to emerge. But I do think we have to pay attention to the dollar balance sheets in Asia. In some ways, for the same reason that we should have been paying more attention to the dollar balance sheets of the European banks before the global crisis. They’ve gotten to be big, they’ve become an important source of funding in the U.S. economy.

Great. This duration mismatch that the Asian insurers are carrying on their long term debt issuance, and the wedge between what the RMB has priced in for the Chinese economy versus where we actually end up. Those are really great.

That’s a really accurate technical description of my argument. Yeah.

Great. Well, thank you so much for joining us, Brad. This has been wonderful to have you with us. We really appreciate you taking the time.

Yeah. It’s been a lot of fun. There are not too many podcasts that can get excited about dollar funding and Asian insurers.

Well, you always find fellow wonks here. Thank you.

Thanks a lot, Brad.

We hope you enjoyed today’s conversation with Brad. For more episodes like this, you can find us iTunes, Google Play, Stitcher, and Spotify. If you like what you hear, please leave a review. Feedback from listeners like you will help more people find us. And for even more content, look up our Pacific Exchange Blog available at frbsf.org. Thanks for joining us.