WEBVTT - Hyun Song Shin on the New Financial Stability Risks 0:00:10.119 --> 0:00:13.360 Hello, and welcome to another episode of the All Thoughts podcast. 0:00:13.400 --> 0:00:16.480 I'm Tracy Alloway and I'm Joe Wisenthal. Joe, we are 0:00:16.520 --> 0:00:20.200 still here at Jackson Hole. We are coming to the 0:00:20.400 --> 0:00:24.080 end of our sort of discussion series, however, and I 0:00:24.120 --> 0:00:26.960 think one of the themes that has emerged from a 0:00:26.960 --> 0:00:29.360 lot of the people that we've been talking to at 0:00:29.360 --> 0:00:33.519 this event is this idea of living with high levels 0:00:33.560 --> 0:00:38.240 of public debt and high inflation all while having basically 0:00:38.280 --> 0:00:42.080 a global financial system that is built on bonds, things 0:00:42.120 --> 0:00:44.240 that are supposed to be boring. They're not supposed to 0:00:44.240 --> 0:00:48.200 see their yields kind of swing wildly, and yet that's 0:00:48.240 --> 0:00:49.200 been what's happening. 0:00:49.520 --> 0:00:52.800 Two things. One is to your point on yields, I mean, 0:00:52.840 --> 0:00:54.200 it's the story of the last year, but it's really 0:00:54.200 --> 0:00:56.840 also like the story of the last month, sure, which 0:00:56.880 --> 0:00:59.000 is that you know, you think things have mellowed out 0:00:59.000 --> 0:01:01.760 and suddenly you get a new spike mortgage rates at 0:01:01.760 --> 0:01:04.040 the high since two thousand and ten year or thirty 0:01:04.080 --> 0:01:06.160 year year old's highest lease since like two thousand and seven. 0:01:06.520 --> 0:01:09.200 The other thing is, can I just say one fun 0:01:09.200 --> 0:01:12.520 thing about Jackson The whole is running into possible odled 0:01:12.560 --> 0:01:15.000 guests on a hiking trail, Yes, which is a really 0:01:15.000 --> 0:01:17.240 fun aspect of here. You notice, like when we're back 0:01:17.240 --> 0:01:19.160 in New York, we have to do all this work 0:01:19.440 --> 0:01:22.120 to try to find guests. You know, here you just like, 0:01:22.400 --> 0:01:23.600 oh hey, it's you. 0:01:23.600 --> 0:01:26.000 You just pick them off in the wilderness. Yeah, like 0:01:26.120 --> 0:01:26.760 creditors you. 0:01:26.800 --> 0:01:28.759 Want to come on the show? Yeah, all right. 0:01:28.800 --> 0:01:31.959 Well, one person we did actually run into while hiking 0:01:32.080 --> 0:01:34.040 is one of our favorite all blots guests. We're going 0:01:34.080 --> 0:01:36.080 to be speaking to Hyun Sung Shin. He is, of 0:01:36.120 --> 0:01:39.080 course economic advisor and head of research at the Bank 0:01:39.120 --> 0:01:43.360 for International Settlements, and we're going to try to synthesize 0:01:43.560 --> 0:01:46.560 a lot of the discussion points we've been having for 0:01:46.600 --> 0:01:48.240 the past day or so, right. 0:01:48.120 --> 0:01:50.000 And to your point, you know, I do think that 0:01:50.240 --> 0:01:52.320 and this was sort of part of the interesting thing 0:01:52.360 --> 0:01:55.120 we talked about with Darryl Duffy, which is that we 0:01:55.200 --> 0:01:57.560 came out of two thousand and eight very focused or 0:01:57.600 --> 0:01:58.960 two thousand and eight two thousand and nine like very 0:01:59.000 --> 0:02:02.800 focused on credit risk, like downside risk, etcetera. And I 0:02:02.840 --> 0:02:05.480 think by and large policy makers, we probably did a 0:02:05.480 --> 0:02:08.720 good job of de risking the system from the perspective 0:02:08.720 --> 0:02:11.920 of credit. Right then we got Silicon, we got this 0:02:12.040 --> 0:02:15.160 rate spike, we got the highest inflation in at least 0:02:15.320 --> 0:02:18.720 four decades, we got Silicon Valley Bank, and suddenly like, oh, 0:02:18.720 --> 0:02:20.480 there's other risks. It's not just credit risk. 0:02:20.560 --> 0:02:23.240 Well, we reduced credit risk at the expense of interest 0:02:23.320 --> 0:02:25.919 rate risk, it feels like, and we sort of buy 0:02:26.120 --> 0:02:30.400 design made bonds really fundamental to the way that, again 0:02:30.440 --> 0:02:33.239 the financial system works. So we need to dig into 0:02:33.600 --> 0:02:36.959 these financial stability risks as they are now. So without 0:02:36.960 --> 0:02:39.440 further ado, let's bring in Hun Hun. Thank you so 0:02:39.520 --> 0:02:41.800 much for coming back on. I think this must be 0:02:41.840 --> 0:02:44.840 your fourth or fifth opposite something like that, but. 0:02:44.919 --> 0:02:47.440 Our first person, which yeah, absolutely, I mean I was 0:02:47.480 --> 0:02:49.080 going to say, it's such a pleasure to do this 0:02:49.120 --> 0:02:52.560 in person, Tracy, and it's a great topic, of course, 0:02:52.600 --> 0:02:55.480 and I think you've had other people on during the 0:02:55.560 --> 0:02:58.040 day that have explored bits of this. But maybe the 0:02:58.080 --> 0:03:00.560 thing to the point to stop this is just to 0:03:00.600 --> 0:03:04.320 take a step back and have an overview of the 0:03:04.360 --> 0:03:09.440 way that the structure of intermediation has changed since the GFC. 0:03:09.760 --> 0:03:13.640 As you say, Joe, the GFC very much was around 0:03:13.680 --> 0:03:16.519 the banking sector and in particular credit risk. The idea 0:03:16.600 --> 0:03:21.440 is that you need capital there to absorb losses on 0:03:21.520 --> 0:03:24.200 the assets because they're risky assets, and that's the way 0:03:24.240 --> 0:03:27.919 that you protect the depositives and other claimholders. I think 0:03:27.919 --> 0:03:31.280 what we saw in March twenty twenty with the stress 0:03:31.280 --> 0:03:34.880 episode in the treasury market is that even safe securities, 0:03:35.240 --> 0:03:39.120 it can be at the center of a financial stress event. 0:03:39.760 --> 0:03:43.360 I think the UK guilt episode last year. Again, these 0:03:43.360 --> 0:03:46.200 are safe assets, but they were very much at the 0:03:46.200 --> 0:03:50.000 center of financial stress. And I think what that does 0:03:50.040 --> 0:03:53.680 point to is the shifting nature of risks, the different 0:03:53.680 --> 0:03:56.560 propagation mechanisms, and you know, the different set of players 0:03:57.120 --> 0:04:00.920 out there. So we're going from banks to non bank players. 0:04:00.960 --> 0:04:04.320 In the jargon they're called MBFI, it's non bank financial intermediaries. 0:04:04.800 --> 0:04:07.240 But it's not just the intermediaries. I mean, it's also 0:04:08.440 --> 0:04:12.800 the way that infrastructures, you know, CCPs, exchanges, they've also 0:04:12.880 --> 0:04:15.200 become very important as well. So this is I think 0:04:15.240 --> 0:04:17.440 a very very important topic for us to touch on. 0:04:18.560 --> 0:04:19.240 Well, talk to. 0:04:19.200 --> 0:04:23.000 Us more about who those non bank financial intermediaries are, 0:04:23.120 --> 0:04:26.279 because I think people often hear the term, the more 0:04:26.560 --> 0:04:30.480 colloquial term shadow banking, and it sounds kind of nefarious. 0:04:30.520 --> 0:04:34.320 But there are all these different flavors of non bank investors, 0:04:34.320 --> 0:04:36.839 So talk to us about who they are and what's 0:04:36.960 --> 0:04:39.839 happened to them in the years since two thousand and eight. 0:04:40.440 --> 0:04:42.600 Well, you can draw a kind of flow chart here. 0:04:42.800 --> 0:04:45.120 You've seen those in those New York Fed charts where 0:04:45.120 --> 0:04:48.640 you have ultimate creditors on one side and ultimate borrowers 0:04:48.680 --> 0:04:51.320 on the other side, and money flows from right to 0:04:51.400 --> 0:04:54.240 left following the balance sheet direction. We can think of 0:04:54.680 --> 0:04:57.320 something like that in this case as well. We have 0:04:57.480 --> 0:05:00.360 fewer of these bank based intermediaries, although we of course 0:05:00.400 --> 0:05:03.400 still have them, but they've shrunk in size and the 0:05:03.440 --> 0:05:06.760 heft if you like, within the system. Instead, what we 0:05:06.839 --> 0:05:10.240 have are many non leverage players in asset managers of 0:05:10.279 --> 0:05:14.240 various stripes, life insurance companies, pension funds. But I think 0:05:14.240 --> 0:05:17.320 a very very important class of players are the other 0:05:17.400 --> 0:05:21.440 hedge funds as well. There are non regulated market intermediaries there, 0:05:22.000 --> 0:05:25.960 and a very important part of the infrastructure here would 0:05:26.000 --> 0:05:31.880 be the new sexual counterparties other exchanges where rather than 0:05:31.920 --> 0:05:37.360 having intermediation go through a dealer balance sheet, you have clearing. 0:05:37.640 --> 0:05:41.680 You have the central counterparties that actually act as creditors 0:05:41.680 --> 0:05:45.120 and debtors to a wide range of participants. 0:05:45.480 --> 0:05:49.680 So you mentioned March twenty twenty, you mentioned the guilt episode. 0:05:49.720 --> 0:05:53.040 There's also, of course March twenty twenty three with Silicon 0:05:53.160 --> 0:05:55.880 Valley Bank. I guess we have had this run up 0:05:55.920 --> 0:05:58.479 in yields over the last month again, but I don't think, 0:05:58.560 --> 0:05:59.960 you know, nothings. 0:05:59.600 --> 0:06:01.279 Floated, nothing's broken yet. 0:06:01.320 --> 0:06:04.719 Nothing's broken just yet. But it feels like we might 0:06:04.760 --> 0:06:08.839 be in for multiple years of a very different direction. 0:06:08.880 --> 0:06:11.039 Whereas the twenty tens were all about you think great 0:06:11.120 --> 0:06:11.680 checking to go up. 0:06:11.640 --> 0:06:12.440 And they go back down. 0:06:12.440 --> 0:06:14.360 Do you think finally you're breaking out, they go back down. 0:06:14.480 --> 0:06:16.800 It will be in the other direction where we think 0:06:16.960 --> 0:06:19.760 rates and inflation are settling down, but it turns out 0:06:19.800 --> 0:06:21.800 they gather steam. We don't know, but maybe that is 0:06:22.000 --> 0:06:23.760 How do you think about that in terms of like 0:06:24.040 --> 0:06:27.599 systemic risk, financial risk? How is it different than a 0:06:27.640 --> 0:06:29.600 period of like credit risk and slow growth. 0:06:29.960 --> 0:06:32.920 It's worth thinking about the journey that we've been on 0:06:33.000 --> 0:06:35.320 for the last ten years or so, well maybe ten 0:06:35.360 --> 0:06:38.600 twelve years. You know, we've had a very long period 0:06:38.880 --> 0:06:42.520 of loaf of long interest rates and of course central 0:06:42.520 --> 0:06:46.359 bank acid purchases that has really compressed the Yel curve, 0:06:46.960 --> 0:06:50.279 and what that's meant is that borrowers have taken advantage 0:06:50.279 --> 0:06:53.239 of that and they've turned out you know, their borrowing. 0:06:53.400 --> 0:06:56.120 And we see it in the corporate sector. They've really 0:06:56.160 --> 0:06:59.760 turned out the bondissuance. We see it in the household 0:06:59.760 --> 0:07:03.520 sector as well. But I think especially important would be 0:07:03.600 --> 0:07:08.039 the government bond market. There's been an increased duration of 0:07:08.080 --> 0:07:13.280 the bond's outstanding. Not surprising really because government that managers 0:07:13.560 --> 0:07:16.600 would also be taking advantage of the low long term rate. 0:07:16.760 --> 0:07:20.200 So just to give you a number in the Biathaniel 0:07:20.240 --> 0:07:24.040 Economic Report this summer, we had a small discussion on this. 0:07:24.240 --> 0:07:28.480 If you look at the duration in aggregate of advanced 0:07:28.480 --> 0:07:31.920 economic government bonds, it was around seven years at the 0:07:32.040 --> 0:07:36.160 end of twenty ten. That's now nine plus years. So 0:07:36.160 --> 0:07:40.480 we've had a tremendous lengthening of duration, and so as 0:07:41.160 --> 0:07:44.080 central banks of race rates in response to inflation, what 0:07:44.120 --> 0:07:47.080 that's meant is that the price impact has been that 0:07:47.160 --> 0:07:50.640 much larger. So if you're not marking to market, if 0:07:50.680 --> 0:07:54.880 you're using ultimaturity accounting, we call that interest rate risk 0:07:54.920 --> 0:07:57.520 on the banking book, and this is what happened with SVB. 0:07:58.120 --> 0:08:01.200 You don't mark the market until you have too. If 0:08:01.200 --> 0:08:04.280 you're marking to market, then it's duration risk. So as 0:08:04.440 --> 0:08:07.240 long term rates go up, the price of your assets 0:08:07.600 --> 0:08:10.840 will go down accordingly. And the longer the duration, the 0:08:10.880 --> 0:08:13.840 bigger the impact. And I think in that sense there 0:08:13.880 --> 0:08:18.560 is a broad continuum here between what we saw in 0:08:18.720 --> 0:08:23.160 SVB the UK guilt episode. But also I think just 0:08:23.200 --> 0:08:26.800 in terms of emerging market bonds that I think we'll 0:08:26.880 --> 0:08:30.080 also talk about during the course of this conference. So 0:08:30.120 --> 0:08:33.400 I think that's if you like looking through the sector 0:08:34.240 --> 0:08:37.000 to the underlying exposures, and if we do that, I mean, 0:08:37.040 --> 0:08:40.120 that's what's out there. What's different, of course, is that 0:08:40.280 --> 0:08:44.480 inflation is high and so monetary policy has to respond, 0:08:45.000 --> 0:08:47.080 and after a very very long period of low for 0:08:47.160 --> 0:08:50.920 long we are seeing, if you like, the consequences of 0:08:51.000 --> 0:08:54.200 raising rates once the exposure has lengthened. 0:08:54.880 --> 0:08:57.200 So I mentioned this in the intro, and you just 0:08:57.200 --> 0:09:01.000 gave a very good outline of duration risk and this 0:09:01.120 --> 0:09:04.360 idea of additional sensitivity to interest rate moves. But I 0:09:04.440 --> 0:09:06.680 mentioned in the intro that a lot of this is 0:09:06.720 --> 0:09:10.560 by design. We have actively encouraged a lot of investors 0:09:10.559 --> 0:09:14.280 to buy more bonds, and maybe that wasn't an issue 0:09:14.320 --> 0:09:17.160 in the lower for longer era, as you just pointed out, 0:09:17.640 --> 0:09:20.760 but it seems to be problematic in an era of 0:09:20.840 --> 0:09:26.480 high inflation. If not bonds, what do you use as 0:09:26.559 --> 0:09:30.040 the sort of safety net for a lot of these entities. 0:09:30.679 --> 0:09:32.560 Well, I think you know, before we go there, Tracy, 0:09:32.600 --> 0:09:36.199 I think we have to point out that lengthening duration has. 0:09:36.040 --> 0:09:37.600 Also a lot of advantages. 0:09:37.920 --> 0:09:42.280 One thing that is very good with very long duration 0:09:42.679 --> 0:09:46.920 exposures is that with very long duration liabilities is that 0:09:46.960 --> 0:09:50.120 you're not facing the rollover risk that you used to 0:09:50.160 --> 0:09:52.880 if you were borrowing short So you know, for example, 0:09:53.400 --> 0:09:56.800 if you go back to the Asian financial crisis, there 0:09:57.080 --> 0:10:00.560 you have the combination of currency mis as well as 0:10:00.559 --> 0:10:04.000 in maturity mismatch, and if your liability is very short 0:10:04.080 --> 0:10:07.679 term and your creditors want their money back, then that's 0:10:07.720 --> 0:10:11.440 actually going to lead to a much sharper episode of stress. 0:10:11.520 --> 0:10:16.360 So lengthening duration has also a lot of advantages for 0:10:16.840 --> 0:10:20.200 financial stability, for mitigating financial stability risks. 0:10:20.280 --> 0:10:22.720 This would be one reason why we perhaps haven't seen 0:10:22.760 --> 0:10:26.400 as many bankruptcies as were expected as the FED raisers rates, 0:10:26.400 --> 0:10:29.360 because everyone spent the past two years terming out their debt. 0:10:29.720 --> 0:10:34.040 Absolutely, and that's especially true for those homeowners who actually 0:10:34.080 --> 0:10:36.840 refinance their mortgage. I think that's a very very good example. 0:10:37.280 --> 0:10:39.560 And what we also see is that in the household 0:10:39.559 --> 0:10:43.280 sector mortgage duration has also lengthened as well, not just 0:10:43.320 --> 0:10:46.400 in the US, and US is special because of its 0:10:46.800 --> 0:10:49.120 of its institution of a thirty year fixed rate, but 0:10:49.160 --> 0:10:51.440 it's also true in other economies that there's been this 0:10:51.600 --> 0:10:54.680 shifting out. I think on balance, I would say that 0:10:54.960 --> 0:10:59.000 the lengthening of the liabilities is a good thing on balance, 0:10:59.600 --> 0:11:03.120 but every you know, silver lining as a cloud, and here, 0:11:03.200 --> 0:11:05.960 you know, we have exactly this problem that it's not 0:11:06.000 --> 0:11:09.239 a free lunch. You have to pay for the additional 0:11:09.400 --> 0:11:11.160 risk that comes from greater duration. 0:11:12.320 --> 0:11:16.199 And what about the if not bonds, what question? Because 0:11:16.240 --> 0:11:18.720 this is what I struggle with, What is the next 0:11:18.760 --> 0:11:20.640 safe asset? Basically, well, I. 0:11:20.640 --> 0:11:24.000 Think the safest asset is just money, and I think 0:11:24.000 --> 0:11:26.600 this is where the money, in the sense of the 0:11:26.679 --> 0:11:29.200 high powered money that is issued by the central bank. 0:11:29.679 --> 0:11:33.400 So when a central bank conducts QE, what happens is 0:11:33.920 --> 0:11:37.720 the central bank takes out duration by purchasing the bonds, 0:11:38.120 --> 0:11:40.640 but then pays for it by creating reserves that are 0:11:40.640 --> 0:11:44.080 held by the sellers of those bonds, typically commercial banks, 0:11:44.240 --> 0:11:47.120 who then would pass that on to the sellers, you know, 0:11:47.160 --> 0:11:51.319 their customers. What that means, though, is that as interest 0:11:51.400 --> 0:11:55.800 rates rise, the assets that the central banks hold will 0:11:55.840 --> 0:11:57.760 also be subject to losses, and this is why we're 0:11:57.800 --> 0:12:01.000 seeing the spate of losses on set bank balance sheet. 0:12:01.559 --> 0:12:04.680 The important thing about central bank liabilities, though, is that 0:12:04.720 --> 0:12:07.880 they're not subject to redemption. The central bank doesn't have 0:12:08.000 --> 0:12:11.680 to repay the dollar with another dollar, because that is 0:12:11.720 --> 0:12:16.120 the ultimate that is the ultimate liability, and as we 0:12:16.720 --> 0:12:20.320 argue in a BIS bulletin recently, we should not be 0:12:20.360 --> 0:12:23.520 so concerned about central bank losses in the same way 0:12:23.559 --> 0:12:26.480 that we are concerned about losses suffered by a private 0:12:26.480 --> 0:12:29.960 sector entity who are subject to those redemptions. But it 0:12:29.960 --> 0:12:32.240 doesn't mean that there are no limits. I mean, clearly 0:12:32.320 --> 0:12:36.000 there are limits, and we see those limits especially in 0:12:36.640 --> 0:12:40.680 very fragile, emerging and developing economies that don't have the 0:12:40.720 --> 0:12:44.600 same credibility in the value of money as with the 0:12:44.640 --> 0:12:48.080 FED or with an advanced economy with other advant economy central. 0:12:47.880 --> 0:13:06.960 Bank, this is a good time to ask you a question, 0:13:07.760 --> 0:13:09.920 and I also pose this to very Eike and Green 0:13:09.960 --> 0:13:13.840 and earlier conversation balance sheet policy. I mean, I think 0:13:13.880 --> 0:13:19.160 we typically associate FED, QEY, Etceteram. Central banks are building 0:13:19.200 --> 0:13:21.880 out that capacity, and it seems like in the post, 0:13:22.160 --> 0:13:25.320 in the sort of COVID era, we are seeing more 0:13:25.440 --> 0:13:30.080 central central banks having some credibility or capacity enough to 0:13:30.160 --> 0:13:32.760 like be a buyer of last resort for their own 0:13:32.760 --> 0:13:34.440 government bond markets in a way that we didn't have 0:13:34.480 --> 0:13:36.560 in the past. Is that fair? Is that an accurate character? 0:13:36.679 --> 0:13:39.120 I think, Joe, what I would say is, if we 0:13:39.160 --> 0:13:42.280 look at the events of last year, if anything, it's 0:13:42.320 --> 0:13:45.000 been the emerging markets that have done better in some 0:13:45.040 --> 0:13:47.920 respects than the advanced economies. And let me explain what 0:13:47.960 --> 0:13:50.800 I mean by that. So you know, good comparison is 0:13:50.880 --> 0:13:54.640 between the events of last year with the earliest stress 0:13:54.679 --> 0:13:58.440 episode for emerging markets in twenty fifteen twenty sixteen. So 0:13:58.559 --> 0:14:01.920 back then, what we saw was a very strong dollar 0:14:02.040 --> 0:14:07.320 coupled with very low commodity prices capital outflows for emerging markets. 0:14:07.440 --> 0:14:11.440 We saw a sharp steepening of yeld curves. That combination 0:14:12.160 --> 0:14:15.560 of stresses are very typical of emerging market stress episodes 0:14:15.600 --> 0:14:19.800 that were familiar from the textbooks. What we saw last 0:14:19.880 --> 0:14:22.920 year was actually quite different. What we saw was the 0:14:23.040 --> 0:14:25.200 nature of the shocks were different. We had a war 0:14:25.440 --> 0:14:30.840 and a pandemic, and what that meant was when commodity 0:14:30.840 --> 0:14:35.360 prices rose, emerging markets that were commodity exporters actually did 0:14:35.880 --> 0:14:39.720 very well relative to historical experience. If you look at 0:14:39.800 --> 0:14:43.160 the Mexican peso or the Brazilian reality, it actually appreciated 0:14:43.640 --> 0:14:47.240 last year, and that's very different from later the Euro, 0:14:47.480 --> 0:14:50.560 the yen, even all the other advanced economy currencies. We 0:14:50.640 --> 0:14:55.040 see quite a resilient picture and so they didn't even 0:14:55.120 --> 0:14:58.960 need to enter the market like that. Now, Joe, your 0:14:59.080 --> 0:15:03.440 question raises very important issue, which is when can a 0:15:03.480 --> 0:15:07.840 central bank come into the market, intervene and play the 0:15:07.960 --> 0:15:12.000 role of a buyer of last resort. I think this 0:15:12.080 --> 0:15:14.680 is a very very important policy, Sue. Maybe we can 0:15:14.920 --> 0:15:17.080 get back to some of the discussions that's happening in 0:15:17.120 --> 0:15:20.000 the official world, But when we look at the emerging markets, 0:15:20.040 --> 0:15:23.760 in particular emerging market central banks policy makers generally are 0:15:23.840 --> 0:15:29.200 very reluctant to wade into markets in those stressed circumstances, 0:15:29.880 --> 0:15:33.880 and for good reason, because when you go into the market, 0:15:34.200 --> 0:15:36.920 you have to buy the bonds and you're creating reserves 0:15:37.120 --> 0:15:39.320 as the byproduct of that that's going to be held 0:15:39.320 --> 0:15:43.680 by some of the domestic participants. You're creating money, so 0:15:43.880 --> 0:15:47.800 liquidity injection. That has many, many good aspects, But what 0:15:47.840 --> 0:15:50.000 it means is you're also going to put pressure on 0:15:50.080 --> 0:15:52.840 exchange rate. So if you're pushing a lot of liquidity 0:15:52.880 --> 0:15:55.960 into the system and you're a central bank that doesn't 0:15:55.960 --> 0:15:58.840 have the credibility of a central bank like the FED, 0:15:59.560 --> 0:16:01.840 then that's going to lead to a very sharp depreciation 0:16:01.960 --> 0:16:05.280 of exchange rate that could in fact do more harm 0:16:05.360 --> 0:16:08.880 than good. So I think the short answer to your question, Joe, 0:16:08.960 --> 0:16:11.240 is that last year we saw a very different set 0:16:11.240 --> 0:16:15.360 of circumstances. Emerging markets actually did pretty well. I would 0:16:15.400 --> 0:16:18.880 say the major emerging markets did very very well. I mean, 0:16:18.880 --> 0:16:22.800 they are clearly more stressed developing economies that still have 0:16:22.880 --> 0:16:25.560 to borrow in dollars and so on, but emerging markets 0:16:25.600 --> 0:16:28.080 did particularly well. But I think there is a very 0:16:28.080 --> 0:16:32.760 important lesson here on what are the circumstances that mean 0:16:32.800 --> 0:16:36.360 that you can enter the market with impunity and you know, 0:16:36.520 --> 0:16:38.680 when can you enter the market and get away with it? 0:16:39.680 --> 0:16:42.200 Well, why don't we go back to central banks that 0:16:42.320 --> 0:16:45.960 can create reserves without putting necessarily downward pressure on their 0:16:46.000 --> 0:16:48.760 own currencies. I wanted to talk to you about leverage 0:16:49.120 --> 0:16:52.440 and bonds because I think we're used to talking about 0:16:52.480 --> 0:16:56.280 leverage in the context of credit, and everyone remembers credit 0:16:56.320 --> 0:16:59.040 derivatives from their starring role in the two thousand and 0:16:59.040 --> 0:17:03.560 eight financial crisis. But I guess what's perhaps underappreciated, although 0:17:03.560 --> 0:17:06.080 it feels like more people are becoming aware of it now, 0:17:06.160 --> 0:17:09.040 is that there's also a lot of leverage attached to 0:17:09.160 --> 0:17:11.879 the bond market, and we've seen sort of flashes of 0:17:11.920 --> 0:17:17.000 it emerge, most notably in the March twenty twenty treasury event. 0:17:17.119 --> 0:17:20.040 But how are you thinking about that issue and how 0:17:20.040 --> 0:17:23.080 do you see leverage emerging in the world of bonds. 0:17:23.840 --> 0:17:26.639 I think that's a very important question, Tracy, and I 0:17:26.640 --> 0:17:30.399 think it goes to the point made at the outset, 0:17:30.880 --> 0:17:34.560 how is it possible that a safe asset can still 0:17:34.560 --> 0:17:36.440 be at the center of a stress event? And I 0:17:36.480 --> 0:17:38.840 think I think here we have to think about the 0:17:38.880 --> 0:17:42.960 possible reasons for perverse demand responses, and I will sort 0:17:42.960 --> 0:17:46.400 of make it more concrete shortly. By a perverse demand response, 0:17:46.400 --> 0:17:49.800 what I mean is when a price falls, we typically 0:17:50.080 --> 0:17:52.879 would think, well, that means that it's more attractive to buy, 0:17:53.200 --> 0:17:56.000 and so people would come into the market. So when 0:17:56.040 --> 0:17:58.000 the price falls, you expect people to come in and 0:17:58.080 --> 0:18:03.600 pick up the cheap assage. But in these stress episodes, 0:18:03.640 --> 0:18:07.679 which you typically find is that a price decline actually 0:18:08.520 --> 0:18:12.280 generates more sales and that actually, of course leads to 0:18:12.320 --> 0:18:14.520 further price declines and you can have this loop. Now 0:18:14.560 --> 0:18:18.080 why would you have this perverse demand response. Well, leverage 0:18:18.720 --> 0:18:20.639 is one way that you can have that. So if 0:18:20.680 --> 0:18:23.359 you're leveraged. If you're levered and the price of your 0:18:23.400 --> 0:18:26.359 assets fall, well, your creditors want their money back, you 0:18:26.440 --> 0:18:29.800 need to meet margin calls, and you have to sell, right. 0:18:29.880 --> 0:18:32.000 This is something we spoke with Daryl Duffy about that 0:18:32.200 --> 0:18:35.440 typically the thing you sell first is your safest, most 0:18:35.480 --> 0:18:38.480 liquid asset, which would usually be a bond, probably a 0:18:38.600 --> 0:18:39.399 US Treasury of so. 0:18:39.440 --> 0:18:42.119 Absolutely, and so this is how a safe asset can 0:18:42.160 --> 0:18:44.480 still be at the center of a stress event. But 0:18:44.520 --> 0:18:47.720 it doesn't have to be leverage as such. It could 0:18:47.720 --> 0:18:50.080 be leverage like behavior. And what I mean by that 0:18:50.200 --> 0:18:53.400 is what other ways can you have where a price 0:18:53.480 --> 0:18:57.359 decline would beget more sales. I think a typical and 0:18:57.400 --> 0:19:01.120 a very very classical example, and it's in the mortgage market. Actually, 0:19:01.480 --> 0:19:04.200 a well known historical episode is what happened in nineteen 0:19:04.280 --> 0:19:07.159 ninety four when you have this rapid steepening of the 0:19:07.200 --> 0:19:10.000 Yelk curve. The way that the embedded option in the 0:19:10.040 --> 0:19:13.560 mortgage market, in mortgages means that you know when you hedge, 0:19:14.000 --> 0:19:17.320 so when rates go up, the duration increases. Actually because 0:19:17.359 --> 0:19:18.600 people stop refinancing. 0:19:18.920 --> 0:19:20.600 This was the big convexity. 0:19:20.720 --> 0:19:25.639 This was the convexity issue. Now there's something similar also 0:19:25.880 --> 0:19:29.560 in I mean, something similar can happen even in very 0:19:29.600 --> 0:19:32.840 boring sectors like you know, pension funds or life insurance. 0:19:33.000 --> 0:19:36.600 So if you're trying to match duration, and you know 0:19:36.640 --> 0:19:39.600 you have liabilities to your policy holders which are let's say, 0:19:39.640 --> 0:19:41.840 you know, thirty years, but you have assets that are 0:19:41.880 --> 0:19:45.640 ten years, liabilities are much longer duration than your assets. Now, 0:19:45.680 --> 0:19:49.879 when rates rise, the duration comes in both on your 0:19:49.920 --> 0:19:53.080 assets and your liabilities. But because your liabilities are much 0:19:53.160 --> 0:19:57.159 longer duration than your assets, liability duration comes in much faster. 0:19:57.840 --> 0:19:59.720 So what ends up happening is that you find that 0:19:59.760 --> 0:20:02.720 you've got too much duration on the asset side, so 0:20:02.760 --> 0:20:05.480 you have to sell. So what's just happened, you know, 0:20:05.560 --> 0:20:08.800 race rows, but you're ending up selling. That's another example 0:20:09.280 --> 0:20:09.680 of level. 0:20:09.760 --> 0:20:11.760 This is this is the UK story, right, I mean, 0:20:11.760 --> 0:20:14.119 this is similar. This is what happened. I think the UK. 0:20:14.359 --> 0:20:17.720 It's a combination of both leverage and this story because 0:20:18.080 --> 0:20:22.240 there was the LDI fund aspect, which actually gave it 0:20:22.280 --> 0:20:25.200 a further amplification boost. But the underlying exposures I think, 0:20:25.240 --> 0:20:29.160 are you know, really all the same. Depending on exactly 0:20:29.160 --> 0:20:31.119 how it will play out depends on you know, who 0:20:31.280 --> 0:20:34.520 are the main players, but the principles are very similar. 0:20:34.760 --> 0:20:39.159 So even very conventional, very supposedly boring sectors can still 0:20:39.200 --> 0:20:41.720 be the source of these kinds of you know, perverse. 0:20:41.880 --> 0:20:42.680 That's a fun thought. 0:20:43.800 --> 0:20:45.879 Forgive me if like this is like too big picture 0:20:46.000 --> 0:20:49.000 or too meta of a question, but like for a 0:20:49.080 --> 0:20:52.480 sort of market capitalist system, like someone's got to hold 0:20:52.480 --> 0:20:55.720 the risk people need anycome, you know, it seems to me. 0:20:55.880 --> 0:20:57.440 I don't know, maybe there's not even a question here, 0:20:57.440 --> 0:20:58.760 but it seems to me it's like you know, two 0:20:58.800 --> 0:21:00.560 thousand and eight, Okay, if there's too much credit risk 0:21:00.640 --> 0:21:04.080 and too much bad lending to households and German banks 0:21:04.119 --> 0:21:07.200 somehow financed construction in the sun Belt in some way. 0:21:07.200 --> 0:21:10.480 And then now we're talking about like pretty boring in 0:21:10.760 --> 0:21:14.600 boring bonds. Treasury is going up boring sort of simple 0:21:14.680 --> 0:21:19.199 life insurance companies, pension companies not particularly exotic. Is this 0:21:19.240 --> 0:21:21.320 our fate that for the rest of our careers is 0:21:21.400 --> 0:21:23.480 just every ten years we just have to like pinpoint 0:21:23.520 --> 0:21:25.280 a new area. I mean, because it's not like risk 0:21:25.320 --> 0:21:27.000 can ever go away. Someone's got to hold the sse 0:21:27.040 --> 0:21:29.200 that someone's got to make the income to pay the savers. 0:21:29.359 --> 0:21:31.520 I think the key here is that we have a 0:21:31.600 --> 0:21:35.600 diversity of investors. Okay, so whatever we do, we have 0:21:35.640 --> 0:21:38.320 to make sure that when someone is selling, someone is 0:21:38.320 --> 0:21:41.480 actually willing to come in and buy. And the diversity 0:21:41.680 --> 0:21:45.040 of the market participants is absolutely key for this. And 0:21:45.160 --> 0:21:47.520 this is all about finding the right price, and finding 0:21:47.520 --> 0:21:49.560 the right price means that you know, we have buyers 0:21:49.920 --> 0:21:54.320 as well as sellers. The reason why these financial stability 0:21:54.480 --> 0:21:59.240 channels of propagation can be so corrosive is because sometimes 0:21:59.520 --> 0:22:02.480 one part of the market just becomes too big. They 0:22:02.520 --> 0:22:06.120 grew without our knowing it, without our really noticing it, 0:22:06.720 --> 0:22:09.159