WEBVTT - The Fed Lag and Mortgage Rate Dilemma 0:00:03.920 --> 0:00:07.320 Welcome to Bloomberg Opinion. I'm valny Quinn. This week film 0:00:07.360 --> 0:00:09.600 new homeowners are feeling stuck. You know, they're lucky enough 0:00:09.640 --> 0:00:11.920 to have a low rate, but they're afraid to move 0:00:11.920 --> 0:00:13.880 where They're just worried that the costs are going to 0:00:13.960 --> 0:00:16.640 be so exhorbiting compared to where they are right now. 0:00:16.720 --> 0:00:20.200 A conversation with Alexis Leander's on homeowners feeling trapped by 0:00:20.560 --> 0:00:23.919 they're cheap mortgages. And if you're going to use the 0:00:24.040 --> 0:00:26.880 Chinese burdon or chet GPT, it's going to align with 0:00:27.160 --> 0:00:30.280 the Communist Party's view of the world. Right, So if 0:00:30.320 --> 0:00:34.480 you ask them politically sensitive questions about us or Taiwan 0:00:34.760 --> 0:00:37.239 or China's place at the world, you don't get that 0:00:37.479 --> 0:00:40.400 view of the walk. Tim Colvin on how chat GPT 0:00:40.560 --> 0:00:44.720 will perform partially according to its ideological underpinnings. First, though, 0:00:44.760 --> 0:00:48.920 to this extraordinary month in markets, particularly rates markets, chief 0:00:49.000 --> 0:00:52.880 rates correspondent to Garfield Reynolds joins Garfield, You'll start the 0:00:52.920 --> 0:00:56.080 month at four point one o nine. On the twos, 0:00:56.120 --> 0:00:58.600 we're now at four sixty nine, and on the tens 0:00:58.640 --> 0:01:00.760 we were at three forty two and now above three 0:01:00.920 --> 0:01:03.720 eighty seven. To some extent, that's just a regular repricing, right, 0:01:03.760 --> 0:01:06.039 the Fed moving by twenty five basis points and signaling 0:01:06.040 --> 0:01:08.760 more to come. Looking back on the month, has it 0:01:08.880 --> 0:01:13.400 been an easy repricing or did it cause pain in places? No, 0:01:13.520 --> 0:01:18.040 it's definitely not been an easy repricing. The market had 0:01:18.080 --> 0:01:22.679 to abandon the thesis that the Fed was close to 0:01:23.360 --> 0:01:27.240 calling halt and that then we could start anticipating when 0:01:27.280 --> 0:01:30.480 they're y cut. We had seventy five basis point hike 0:01:30.600 --> 0:01:34.640 in November, fifty in December, and then as expected at 0:01:34.680 --> 0:01:37.360 the beginning of this month, we had twenty five. So 0:01:37.480 --> 0:01:40.120 the market was sort of going, okay, next move is 0:01:40.319 --> 0:01:42.720 no interest rate hike at all, and the next move 0:01:42.720 --> 0:01:45.040 after that as a cut instigate. The FED has made 0:01:45.040 --> 0:01:49.240 it clear that it's going to do multiple hikes from here. 0:01:49.400 --> 0:01:51.800 In fact, by now the pricing is for three more 0:01:52.000 --> 0:01:54.520 at least, and that it's going to hold rates higher 0:01:54.640 --> 0:01:57.520 for quite some time. Now. That's a message the FED 0:01:57.560 --> 0:02:01.360 had been tending for some time. It took several punches 0:02:01.400 --> 0:02:06.040 to the gut for bond bulls before that message got across. 0:02:06.360 --> 0:02:09.720 Is the message now across? Yeah, I mean the message 0:02:09.840 --> 0:02:12.919 is now pretty much across you can partly see that, 0:02:14.080 --> 0:02:18.600 you know, there's a lessoning in volatility and a willingness 0:02:18.720 --> 0:02:24.160 for some investors to come back in at these elevated yields, 0:02:24.040 --> 0:02:27.799 a sort of buy and hold investors. But there's kind 0:02:27.840 --> 0:02:29.840 of a sense that the bond market at the very 0:02:29.960 --> 0:02:33.480 least is no longer fighting the FED. It's now much 0:02:33.520 --> 0:02:36.480 more closely in line with the FED. Starting with that 0:02:36.600 --> 0:02:39.440 rate prices that we talked about, where the expectation is 0:02:39.440 --> 0:02:42.200 for a pig rate of seventy five basis points high 0:02:42.240 --> 0:02:46.720 per year and probably no rate cuts until next year. 0:02:47.800 --> 0:02:49.920 Ten year yields seem to be reluctant to cross that 0:02:49.960 --> 0:02:52.760 four percent level again that they did last October November, 0:02:52.800 --> 0:02:57.400 but not since. The question might be if three point 0:02:57.480 --> 0:03:00.880 nine seventy four, which was the high most recently, does 0:03:00.880 --> 0:03:03.640 that represent the bearish extreme intends for the time being. 0:03:03.639 --> 0:03:06.760 Do you think no? I don't think it does. I 0:03:06.760 --> 0:03:09.520 think that's a level that has been attractive to buy 0:03:09.560 --> 0:03:12.200 and hold investors that had a lot of people talking 0:03:12.200 --> 0:03:15.160 to me about dipping their toes in at this sort 0:03:15.160 --> 0:03:17.680 of level. But that's the thing. They want to be 0:03:17.760 --> 0:03:22.920 sure the temperature is right there, So things like PC 0:03:23.520 --> 0:03:27.120 price indicator, next month's jobs data, well this month's job 0:03:27.200 --> 0:03:29.840 data report at the beginning of next month. Then also 0:03:30.240 --> 0:03:32.919 inflation and what the fat says that it's brass meeting. 0:03:33.280 --> 0:03:36.680 All of these things have the potential to provide fresh 0:03:36.800 --> 0:03:42.080 either data or central bank stance shocks that could readily 0:03:42.200 --> 0:03:46.280 drive you higher. There is a pretty strong anticipation that 0:03:46.320 --> 0:03:50.640 it does go higher, you know something. Four point three 0:03:50.680 --> 0:03:54.040 percent is still seen for the moment at least being 0:03:54.600 --> 0:03:58.040 probably the peak of this cycle. I'm a bit skeptical 0:03:58.160 --> 0:04:02.960 about that estimation, considering that historically ten year yields have 0:04:03.120 --> 0:04:06.400 topped out around about the same level as the fence 0:04:06.480 --> 0:04:10.240 cash rate, and the fence cash rate is already considerably 0:04:10.280 --> 0:04:12.600 above that four point three three percent pick we saw 0:04:12.600 --> 0:04:15.600 an October for the ten year yield, so potentially more 0:04:15.600 --> 0:04:19.640 pain to come. Who's feeding that pain our field? Bond 0:04:19.720 --> 0:04:24.039 investors are certainly feeling that pain. The global aggregate impact. 0:04:24.240 --> 0:04:29.120 It had its best January ever and those gains almost 0:04:29.240 --> 0:04:34.360 got wiped out in February, So bond investors are certainly suffering. 0:04:34.360 --> 0:04:36.600 At the very least, they're looking at the potential that 0:04:37.000 --> 0:04:39.360 this year, if it is going to be gains is 0:04:39.400 --> 0:04:42.640 going to be gains eked out rather than a strong 0:04:42.720 --> 0:04:45.800 bounce back which they were looking for. There's also been 0:04:45.839 --> 0:04:49.000 pain spreading across the equities as they have lost momentum. 0:04:49.080 --> 0:04:53.080 And again they initially they were holding up okay when 0:04:53.640 --> 0:04:56.680 done to turn south, and there was some thought that 0:04:56.839 --> 0:05:00.680 the strong correlation we had seen between the two usually 0:05:00.760 --> 0:05:05.200 not correlated asset classes was going to break instead that 0:05:05.440 --> 0:05:09.120 swinging back, and that's likely a bitter case until the 0:05:09.200 --> 0:05:12.880 tightening cycle is definitively over. Yeah, I mean, it's really 0:05:12.880 --> 0:05:15.520 interesting because we didn't really see the reaction in say, 0:05:15.520 --> 0:05:18.760 the mega tech companies that we might have expected in 0:05:18.839 --> 0:05:21.719 quote unquote more normal times. That has yet to come, 0:05:21.800 --> 0:05:25.120 I guess too potentially. Yeah, well, I mean, and that 0:05:25.240 --> 0:05:28.520 depends a lot on how the economy develops. If you 0:05:28.640 --> 0:05:33.880 look at consumer sentiment drawing higher, the City Group Economics 0:05:33.960 --> 0:05:39.400 Prize Index shows that most data releases are beating expectations. 0:05:39.520 --> 0:05:43.400 So and yeah, the Atlanta our Fair tractor is a 0:05:43.520 --> 0:05:46.920 decent level and it comes to GDP, So the economy 0:05:47.120 --> 0:05:49.440 is showing a fair bit of resilience. As long as 0:05:49.440 --> 0:05:52.520 the economy shows resilience, you can hold out hope that 0:05:52.720 --> 0:05:55.680 equities in particular, look get through this with decent games, 0:05:55.720 --> 0:05:58.120 or even better than decent games, especially when you're thinking 0:05:58.160 --> 0:06:01.880 of text. Worry has to be that there's still a 0:06:01.920 --> 0:06:05.160 strong expectation for a recession, so at some stage of 0:06:05.240 --> 0:06:08.960 recessions gain't hit. That's still the base case for most investors. 0:06:09.040 --> 0:06:13.120 And the question is, you can meg Tex and other 0:06:13.480 --> 0:06:17.160 companies hold on to the gains that they've booked so 0:06:17.200 --> 0:06:21.200 far this year if we do get a recession, especially 0:06:21.279 --> 0:06:24.560 if it's a deep one and not a shallow recession. 0:06:24.960 --> 0:06:28.400 But if the labor market tightness doesn't weaken a little bit, 0:06:28.720 --> 0:06:33.440 how can we possibly get a recession? Well, and to 0:06:33.480 --> 0:06:39.040 some extent that's defends reasoning and willing to go higher 0:06:39.040 --> 0:06:42.240 and higher on rates to make sure inflation comes down. 0:06:42.320 --> 0:06:45.400 You know, they regularly refer to the lessons of the 0:06:45.480 --> 0:06:49.279 nineteen eighties where they eased off too soon is the 0:06:49.320 --> 0:06:52.240 received wisdom, And they ease off too soon why because 0:06:52.279 --> 0:06:54.680 it look like the economy was going into recession. But 0:06:55.080 --> 0:06:59.120 why should say ease off now unless inflation is visibly 0:06:59.400 --> 0:07:03.479 on its way down sustainably to the target range, if 0:07:03.480 --> 0:07:06.279 the labor market is still this strong, if the signs 0:07:06.279 --> 0:07:09.440 of recession aren't there. They kind of said, look, we're 0:07:09.440 --> 0:07:11.440 not going to stop until inflation is down, even if 0:07:11.480 --> 0:07:14.000 we're risking a recession, and at the moment it doesn't 0:07:14.000 --> 0:07:16.720 look like the risking a recession, so why stop. We 0:07:16.760 --> 0:07:19.000 did get the minutes this week too, though, Garfield, and 0:07:19.040 --> 0:07:22.000 it seemed like because we didn't know as much then 0:07:22.040 --> 0:07:25.000 as we know now, FED officials were also willing to 0:07:25.040 --> 0:07:28.120 believe that things were going in their direction, when in fact, 0:07:28.320 --> 0:07:31.240 then we got this massive jobs reward and plenty of 0:07:31.280 --> 0:07:34.800 economic data that suggested inflation wasn't coming down. Our FED 0:07:34.840 --> 0:07:37.360 officials willing to believe that the lag is actually shorter 0:07:37.440 --> 0:07:41.040 than it is. Too well, I think FED officials are 0:07:41.440 --> 0:07:45.600 very much you watch and respond mode. Part of the 0:07:45.760 --> 0:07:49.480 point of the steep rate hikes from last year was 0:07:49.560 --> 0:07:54.280 precisely to get from extremely loose conditions where they were 0:07:54.400 --> 0:08:01.200 helping to exacerbate an overheated economy to restrict fift territory. Now, 0:08:01.240 --> 0:08:03.320 from the first point of view, now that they're in 0:08:03.400 --> 0:08:08.560 restrictive territory, they can take smaller steps as they increase 0:08:08.640 --> 0:08:12.560 that level of restrictiveness, or even pause for a while 0:08:12.880 --> 0:08:15.960 to see if that level is as restrictive as they 0:08:16.120 --> 0:08:20.640 need to get. But that's very different from a scenario 0:08:20.680 --> 0:08:23.240 where they say, oh, okay, we've done enough, therefore we 0:08:23.240 --> 0:08:25.680 can look to cut rate again. They're only going to 0:08:25.680 --> 0:08:29.760 look to cut rate if a inflation is back down 0:08:29.920 --> 0:08:32.640 very close to target and they can see that the 0:08:32.679 --> 0:08:36.559 economy is really really in trouble. Now in the short 0:08:36.679 --> 0:08:40.000 term of they're more alert for our signs that the 0:08:40.080 --> 0:08:43.040 economy is continuing to show more strength, and they expected 0:08:43.480 --> 0:08:46.959 then the Minute seemed to indicate fifty basis points might 0:08:47.000 --> 0:08:50.760 be back on the table, and certainly at least three 0:08:50.880 --> 0:08:54.120 twenty five basis point heights seems to be a consensus 0:08:54.120 --> 0:08:56.280 for what is the base case of what's going to 0:08:56.280 --> 0:08:59.200 be needed. And if the data continues to come in 0:08:59.280 --> 0:09:01.959 stronger than next fact that well then they would look 0:09:02.000 --> 0:09:04.880 to If they don't go fifty, then they might go further, 0:09:05.400 --> 0:09:08.040 just with their twenty five places point hikes to keep 0:09:08.160 --> 0:09:12.200 the pressure on Garfield. Since the market is still projecting 0:09:12.400 --> 0:09:15.480 a chance of a rate cause in December, does that 0:09:15.559 --> 0:09:17.400 imply to the market things there will be a month 0:09:17.440 --> 0:09:20.280 where we suddenly see the data completely fall off a cliff, 0:09:20.760 --> 0:09:23.240 because there's not that many months left between now and December. 0:09:23.320 --> 0:09:28.520 Really well, yeah, they do obviously see a fairly swift turnaround, 0:09:28.559 --> 0:09:32.600 and to be fair, previous cycles we have seen that 0:09:32.640 --> 0:09:35.400 sort of thing. We've seen a switch to ray cuts 0:09:35.480 --> 0:09:37.840 or at least to an easy bias within three to 0:09:37.920 --> 0:09:40.680 six months of the peak in the fad rate. Now, 0:09:40.840 --> 0:09:43.000 as people are in pointing out to me, the difference 0:09:43.120 --> 0:09:46.199 is this time around in casement is much more elevated. 0:09:46.360 --> 0:09:50.960 So that's going to stay the fair hands to some extent, 0:09:51.040 --> 0:09:55.360 you would think, unless the data really collapses when it 0:09:55.360 --> 0:09:58.079 comes to turning towards ray cuts. But there are a 0:09:58.120 --> 0:10:02.520 lot of difficulties in navigating the current climate because so 0:10:02.640 --> 0:10:06.560 much in the way of data and other phenomena are 0:10:06.600 --> 0:10:10.760 extraordinary in the because of the pandemic and the post pandemic, 0:10:11.040 --> 0:10:15.160 you never quite clear to what extent things can revert 0:10:15.240 --> 0:10:18.239 back to how they were in the years before the pandemic. 0:10:18.480 --> 0:10:22.280 But you know, markets and the climates inevitably do factor 0:10:22.360 --> 0:10:27.199 in historical modeling. So if you think that things are 0:10:27.360 --> 0:10:30.040 going to work the way that they worked in the past, 0:10:30.160 --> 0:10:34.200 and if you look the latest Bloomberg Economist Consensus recession 0:10:34.280 --> 0:10:37.000 indicator with it a sixty five percent chance of a 0:10:37.280 --> 0:10:39.600 US recession within a year. If you think there's a 0:10:39.640 --> 0:10:41.760 better than fifty percent chance of recession within a year, 0:10:41.880 --> 0:10:44.200 you probably think there's a better than fifty percent chance 0:10:44.280 --> 0:10:47.920 of racer within a year. Stay tuned. We continue our 0:10:48.000 --> 0:10:51.920 conversation with Garfield Rentals and moments on Bloomberg Opinion. You're 0:10:51.960 --> 0:10:54.880 listening to Bloomberg Opinion and Vannie Quinn. Let's return now 0:10:54.960 --> 0:10:57.920 to our conversation with Garfield Renals. We'll get as thoughts 0:10:57.960 --> 0:10:59.920 on the churn at the Bank of Japan and moment. 0:11:00.160 --> 0:11:02.719 But first, this process is likely to take quite a 0:11:02.760 --> 0:11:05.760 bit of time. It's not going to be we don't 0:11:05.760 --> 0:11:08.560 think smooth. It's probably going to be bumpy. And so 0:11:08.880 --> 0:11:10.760 we think that we're going to need to do further 0:11:10.880 --> 0:11:13.400 rate increases, as we said, and we think that will 0:11:13.440 --> 0:11:16.120 need to hold policy at a restrictive level for a 0:11:16.160 --> 0:11:19.720 period of time. Some final thoughts on FED chair Powells dilemma. 0:11:19.880 --> 0:11:23.160 How does FED chairs your own Powell reconcile saying we 0:11:23.200 --> 0:11:25.200 are going to get to a certain point and we 0:11:25.280 --> 0:11:27.800 are going to hold I'd also say your data dependent 0:11:27.960 --> 0:11:31.280 and you might need to move basically with a trigger finger. Well, 0:11:31.480 --> 0:11:35.600 the loadstone for the fad remains inflation, and as long 0:11:35.640 --> 0:11:38.880 as that is at the elevated level, that is you're 0:11:39.000 --> 0:11:42.240 well about the faith target. They can be clear that 0:11:42.360 --> 0:11:45.120 we are more insted in racing rates from here than 0:11:45.120 --> 0:11:50.199 in holding them alone lowering them. If it does come 0:11:50.320 --> 0:11:53.920 down strongly from here, you know, if we do get 0:11:54.640 --> 0:11:58.679 a return to the disinflationry trend that we've had, anything, 0:11:59.200 --> 0:12:05.040 that placement trend accelerates. Some way of saying, if we 0:12:05.160 --> 0:12:09.840 get inflation coming down faster than than it currently is 0:12:09.880 --> 0:12:14.199 and two levels that would be consistent, well, then the 0:12:14.280 --> 0:12:19.280 FEND could be willing to switch to an easiest to 0:12:19.400 --> 0:12:23.880 a less hawkish bias, especially if there are also signs 0:12:23.880 --> 0:12:27.520 that the economy is starting to roll over. However, if 0:12:27.520 --> 0:12:31.400 you've got inflation coming down rapidly so you can see 0:12:31.400 --> 0:12:34.640 it getting to the appropriate levels, but the economy is 0:12:34.679 --> 0:12:36.920 still burning, well, you're sort of not going to be 0:12:36.960 --> 0:12:40.640 interested in cutting rate because that would be repeating the 0:12:40.640 --> 0:12:42.680 mistakes of the verteen eighties, which is what they said 0:12:42.679 --> 0:12:45.520 they don't want to do. And you know, the feed 0:12:45.640 --> 0:12:49.200 is said the path they do think they can achieve 0:12:49.240 --> 0:12:53.000 a soft landing, but it's a very narrow path. That 0:12:53.280 --> 0:12:58.800 also signal that they're willing on the side of risking 0:12:58.880 --> 0:13:05.040 a hard landing, provided that ensure that inflation is tained 0:13:05.880 --> 0:13:09.680 this time round. They don't want to see elevrated inflation 0:13:09.760 --> 0:13:13.240 lasting for a long time and fitting into expectations. So 0:13:13.600 --> 0:13:18.280 that's part of why bond space the potential for further pain, 0:13:18.520 --> 0:13:22.160 because if the Fed is willing to stay harsher for longer, 0:13:22.920 --> 0:13:28.480 that that could keep yields higher for longer. And also 0:13:28.559 --> 0:13:33.920 why equities could face a bit of a double jeopardy, hey, 0:13:34.000 --> 0:13:36.560 because race might be higher than they think me because 0:13:36.840 --> 0:13:40.080 the feed might end up with that hard landing despite 0:13:40.120 --> 0:13:42.520 the best effort, and that's not going to be good 0:13:42.520 --> 0:13:45.560 for stops. It might be good for bonds, it's probably 0:13:45.559 --> 0:13:48.400 not going to be good for stops. Garfield Green pivoted 0:13:48.480 --> 0:13:51.680 Japan for a moment. How in control is the BOJ 0:13:51.920 --> 0:13:55.160 It appeared that traders were attacking the point five percent 0:13:55.520 --> 0:13:59.320 upper limits on the band again this week. Oh, they 0:13:59.360 --> 0:14:02.200 certainly were, and they're difficulty years. There just aren't many 0:14:02.280 --> 0:14:05.400 ten weird bonds around in Japan because the BJ's bought 0:14:05.400 --> 0:14:10.199 the more so it's not easy. The benchmark yield has 0:14:10.280 --> 0:14:14.320 closed at the two point five percent target every trading 0:14:14.400 --> 0:14:17.559 day since Jevary ten when it rose to that level, 0:14:17.960 --> 0:14:20.240 and then it's been stuck there, even if the last 0:14:20.280 --> 0:14:24.560 couple of days it's jumped above. Now, overwhelming consensus is 0:14:24.960 --> 0:14:27.200 that yield curve control is going to have to go, 0:14:27.480 --> 0:14:31.800 probably sooner rather than later. That it's not sustainable that 0:14:31.880 --> 0:14:35.720 who later will have to change things once he becomes 0:14:35.800 --> 0:14:38.760 BJA governor because the BOJA owns too much of the 0:14:38.840 --> 0:14:45.400 market already, it's dysfunctional. It causes your problems for banks, 0:14:45.480 --> 0:14:50.000 for other companies with everything that's going on, and also 0:14:50.200 --> 0:14:54.760 the concern that the end could resume steep declines unless 0:14:55.160 --> 0:14:59.400 the finds a way to buy fuel bonds. So those 0:14:59.440 --> 0:15:03.360 are all the issues and the difficulties. There's is very 0:15:03.360 --> 0:15:07.240 strong inticipation that Japanese year's going to go higher, don't. 0:15:07.280 --> 0:15:11.440 I question is when, and the secondary one is how, 0:15:11.600 --> 0:15:14.840 because there is a way. Yes. The reason worry that 0:15:14.960 --> 0:15:17.720 if you take the lid off again or even you know, 0:15:17.760 --> 0:15:21.120 shifted a little bit higher, that all sorts of volatility 0:15:21.200 --> 0:15:24.560 could could break out, and in particular that you could 0:15:24.560 --> 0:15:28.320 get foreigners fleeing the short end of the market where 0:15:28.360 --> 0:15:32.560 they've been looping the rewards of the cross currency basis trade, 0:15:33.240 --> 0:15:36.920 and that that could send yields jumping on a range 0:15:36.960 --> 0:15:40.200 of tenors. That that could then set off an evaluation 0:15:40.240 --> 0:15:44.240 shock for Japanese investors. You'll hold those bonds now suddenly 0:15:44.280 --> 0:15:47.400 have to market to market. So there's a lot of 0:15:48.400 --> 0:15:53.400 your concerned bubbling away beneath the surface. Even as a 0:15:53.480 --> 0:15:58.160 stretch of metaphor, the Bank of Japan pushes down hard 0:15:59.080 --> 0:16:02.520 on the lid of the borrowing clothing that it claims 0:16:02.560 --> 0:16:05.360 to control. Right, I mean, it's a terrifying prospect for 0:16:05.400 --> 0:16:08.440 casual way. The end national core CPI is at four 0:16:08.480 --> 0:16:11.600 point three percent, so it's not as if it's helping 0:16:11.760 --> 0:16:13.960 really in terms of inflation or what the longer run 0:16:13.960 --> 0:16:17.120 outlook for inflation will be. Obviously Japan wants it back 0:16:17.200 --> 0:16:20.520 up to two is casual way they're suddenly going to 0:16:20.560 --> 0:16:25.360 innovate in monetary policy and think of something new. Well, 0:16:25.520 --> 0:16:29.240 that would be one way to result. The difficulty is, 0:16:29.320 --> 0:16:33.000 you know, what can they come up with and how 0:16:33.040 --> 0:16:36.360 do they try and move back to a more normal 0:16:36.720 --> 0:16:39.920 way of doing things when they own so much of 0:16:40.080 --> 0:16:42.880 the bond market and may also have a government that 0:16:43.480 --> 0:16:46.600 season needs to spend, for example, on defense and on 0:16:46.640 --> 0:16:50.560 other reforms, and so it's going to go on borrowing. 0:16:51.200 --> 0:16:54.040 It's not going to want to see Japanese yields go 0:16:54.160 --> 0:16:58.000 too high. Japan in general is it a country where 0:16:58.040 --> 0:17:02.280 they prefer to see things moved at a measured pace 0:17:02.600 --> 0:17:06.359 most of the time. So you think he's going to 0:17:06.440 --> 0:17:10.120 try to be contatory, he's going to try to move gradually, 0:17:11.200 --> 0:17:14.760 But then he himself has said at times that if 0:17:14.800 --> 0:17:18.400 you're going to end yield, curve control and negative rates, 0:17:18.480 --> 0:17:22.080 that be better to do it in one fell swoop 0:17:22.760 --> 0:17:25.840 in order to avoid everybody betting against you, and in 0:17:26.000 --> 0:17:31.360 order to be able to control the process better than 0:17:31.440 --> 0:17:35.960 if you signal or allow people to assume that it's 0:17:36.000 --> 0:17:38.560 going to be something that takes place gradual. Right now, 0:17:38.560 --> 0:17:42.320 options markets are positioning for him to retain flexibility. What 0:17:42.359 --> 0:17:45.199 are the odds and have realized I'm asking you to 0:17:45.359 --> 0:17:49.200 answer something that's really unanswerable. When do you think there's 0:17:49.240 --> 0:17:51.600 any possibility that he might actually do away with it 0:17:51.680 --> 0:17:55.480 completely in April and to surprise everybody, that's a possibility 0:17:56.400 --> 0:18:02.040 is soon to be predecessor. Corona managed to surprise markets 0:18:02.119 --> 0:18:05.920 ten years ago when he took over, and everybody knew 0:18:06.080 --> 0:18:09.760 that some sort of extraordinary monetary easing was coming, because 0:18:09.800 --> 0:18:13.119 that was part of the three arrows. But the way 0:18:13.160 --> 0:18:15.159 he did, at the extent to which he did it 0:18:15.760 --> 0:18:20.439 course huge shots. Initially, and in fact, on the first 0:18:20.520 --> 0:18:23.919 week or two of a Japanese yields jump up and 0:18:23.960 --> 0:18:26.159 then down, there was a lot of strains, There was 0:18:26.200 --> 0:18:29.200 a lot of concern, a lot of volatively and again 0:18:29.560 --> 0:18:33.240 as well and then in twenty fourteen, So for a 0:18:33.400 --> 0:18:38.120 seconds to Rota also surprise again. Even though people were 0:18:38.160 --> 0:18:42.600 expecting him to increase easing, they weren't quite sure that 0:18:43.000 --> 0:18:45.679 he shocked them with the way he did it. So 0:18:46.200 --> 0:18:48.720 there's definitely an extential for a surprise that would be 0:18:48.800 --> 0:18:54.920 very surprising to think. The more likely scenario would seem 0:18:55.119 --> 0:18:58.720 that he tries to hold the fort for a while 0:18:58.760 --> 0:19:04.960 at least while he commissioned a review or takes measures 0:19:05.080 --> 0:19:09.200 to try and improve the way the market is working. 0:19:09.640 --> 0:19:11.240 Whatever he does, in a lot of ways, he is 0:19:11.280 --> 0:19:13.840 going to surprise, because if he does do essentially nothing 0:19:14.200 --> 0:19:16.680 at the first meeting, that's also going to be a shock, 0:19:16.920 --> 0:19:20.240 and a lot is going to depend on how well 0:19:20.320 --> 0:19:23.600 he sells whatever the message is that he's sending. This 0:19:23.840 --> 0:19:27.240 is his first chance to show what he has when 0:19:27.240 --> 0:19:32.040 it comes to being a communicator of policies or potential 0:19:32.080 --> 0:19:37.040 policies and how he sees things being developing. And in 0:19:37.080 --> 0:19:40.800 particular he's also obviously talking to a very important constituency, 0:19:41.560 --> 0:19:46.480 the Japanese governing PLATH Chief Rates correspondent Garfield Reynolds. There 0:19:46.840 --> 0:19:51.520 next is your chat GPT Benign well define Benign will 0:19:51.560 --> 0:19:55.560 speak with Bloomberg's Tim Colbyn and Taipei. This is Bloomberg Opinion. 0:19:56.200 --> 0:20:00.280 You're listening to Boomberg Opinion. I'm Vannie Quinn. China may 0:20:00.400 --> 0:20:03.679 have a chat bought advantage, according to Bloomberg Compinions. Tim Kulpin, 0:20:04.080 --> 0:20:08.000 we had an IRL conversation about the ethical questions surrounding 0:20:08.000 --> 0:20:11.120 this technology that's seeing a race for territory from many 0:20:11.200 --> 0:20:14.800 of the major tech companies across the globe. Generative AI 0:20:14.920 --> 0:20:18.600 has obviously suddenly become the revelation Dure. You've pointed out 0:20:18.600 --> 0:20:22.399 that domestic internet companies in China had an advantage in 0:20:22.600 --> 0:20:25.680 search because literally the Great Firewall or the companies weren't 0:20:25.720 --> 0:20:27.520 even there in the first place, or they left once 0:20:27.640 --> 0:20:32.280 censorship threatened their product. So now China has potentially a 0:20:32.400 --> 0:20:35.560 chance again to seize the territory, explained to was your thesis. 0:20:36.359 --> 0:20:39.160 So one of the biggest struggles we're seen with chat GBT, 0:20:39.600 --> 0:20:43.080 and that's essentially a brand name of open ai to 0:20:43.200 --> 0:20:46.600 start up. One of the challenges that's seen already is 0:20:46.640 --> 0:20:48.879 that it doesn't always get dry. And then when Google 0:20:48.920 --> 0:20:52.639 came out famously just recently and showed off their own 0:20:52.760 --> 0:20:56.159 version literally in their initial ads, they made a mistake, right, 0:20:56.480 --> 0:21:01.040 So the problem is that they're gathering from a huge 0:21:01.160 --> 0:21:04.919 corpus of information while this information is conflicting, right, so 0:21:05.080 --> 0:21:07.520 you know, it mixes up east and west, north south, 0:21:07.560 --> 0:21:09.760 you know, black and white, this is sky blue, right, 0:21:09.800 --> 0:21:11.800 that kind of stuff, and like, you know, a humanise 0:21:12.119 --> 0:21:15.280 that doesn't make sense. But because it is just about 0:21:15.280 --> 0:21:18.280 it's not smart. We see a lot of stuff spat 0:21:18.320 --> 0:21:21.200 out bra chat GPT that looks very confident, it looks 0:21:21.320 --> 0:21:25.040 very authoritative, but then uncloser inspection, you realize it's rubbish. 0:21:25.119 --> 0:21:28.360 It's just not true. And that's because they are working 0:21:28.359 --> 0:21:31.119 on a very large corpus of information out there in 0:21:31.160 --> 0:21:35.879 the world, including social media things like Reddit, and you know, 0:21:35.920 --> 0:21:38.720 there's not necessarily any fact checking going on in social media. 0:21:38.800 --> 0:21:43.280 Shock corra, and so that's the problem, whereas in China 0:21:43.359 --> 0:21:46.800 it's a bit dystope into savors. In China, if you're 0:21:47.200 --> 0:21:52.800 crawling the corpus of Chinese information because of censorship, then 0:21:52.840 --> 0:21:55.080 the information you're going to see on the Chinese Internet 0:21:55.240 --> 0:21:58.320 is I wouldn't say it's it's true or accurate, but 0:21:58.359 --> 0:22:00.800 you could say it's veted to an step. Now, not 0:22:00.960 --> 0:22:04.160 all information in China has been sensed. That we have 0:22:04.400 --> 0:22:07.720 so many years of this structure of censorship and control 0:22:07.720 --> 0:22:11.560 of information that if something is online in China, on 0:22:11.600 --> 0:22:15.400 a website on social media, even if it's not necessarily 0:22:15.440 --> 0:22:18.880 true in the factual sense of at least some sense 0:22:18.960 --> 0:22:22.080 or whatever has better and said yes, it's true in 0:22:22.119 --> 0:22:26.760 our eyes. And so that means that the chat GPT 0:22:27.080 --> 0:22:29.280 versions that are being developed by companies like bay Do 0:22:29.680 --> 0:22:32.680 natis in Auli Baba, if they assume that they call 0:22:32.840 --> 0:22:36.440 only the Chinese website, then they're going to be calling 0:22:36.920 --> 0:22:40.640 vetted information. And so the information may spit out when 0:22:40.680 --> 0:22:44.320 given a query is going to be somewhat more aligned 0:22:44.359 --> 0:22:47.680 with what sense it might call accurate, and so you're 0:22:47.720 --> 0:22:50.199 going to get information that does align to be somewhat 0:22:50.240 --> 0:22:53.560 generally correct, is it You're not going to have these problems. 0:22:54.080 --> 0:22:56.359 Is it possible though, that these versions will have their 0:22:56.400 --> 0:23:00.240 own biases built in because of the nature of China censorship? Oh? 0:23:00.280 --> 0:23:03.320 Absolutely so, if you're going to use the Chinese version 0:23:03.359 --> 0:23:05.880 or a chat GPT, it's going to align with the 0:23:05.880 --> 0:23:08.960 Communist Party's view of the world. Right, So if you 0:23:09.080 --> 0:23:13.399 ask them orthically sensitive questions about US or Taiwan or 0:23:13.600 --> 0:23:15.920 China's police of the world, you aren't don't get that 0:23:16.040 --> 0:23:18.639 view of the world, right. That's absolutely going to happen, 0:23:18.720 --> 0:23:23.080 Not because the chat GPT boss itself is necessarily biased, 0:23:23.080 --> 0:23:26.600 but it's coming from a biased view of the world. 0:23:26.760 --> 0:23:28.520 You know, the corpus that is dragged, you know, it's 0:23:28.520 --> 0:23:31.480 pulling its data problem is going to be aligned with that. 0:23:31.720 --> 0:23:34.679 So that's definitely going to be a problem. But in 0:23:34.720 --> 0:23:37.720 the initial phases of training, this stuff around very very 0:23:37.840 --> 0:23:39.679 very early days of all that is happening. This is 0:23:39.960 --> 0:23:43.240 a ten twenty year thing, and the very early days 0:23:43.240 --> 0:23:45.720 of it. The biggest risk you have with any of 0:23:45.760 --> 0:23:49.760 these chat bots is either keeps spinning out bad information 0:23:49.840 --> 0:23:52.200 people won't rely on and they won't trust it. They 0:23:52.200 --> 0:23:54.720 may move away from it, and the other thing is 0:23:54.800 --> 0:23:57.680 you can manipulate it. Right. It wouldn't take much for 0:23:57.840 --> 0:24:00.520