Starting discussion to start your day, coming to a Bloomberg podcast near you.
Now we turn to a leading voice in global central banking. We've got Andrew Bailey, the governor of the Bank of England, coming on next. Since taking office in 2020, Governor Bailey has overseen monetary policy, financial stability, and credential regulation through some of the most uncertain economic times. So, please join me in welcoming him on the stage once again, in conversation with Tim Adams, president and CEO of IIF. A big round of applause, please.
Thank you very much. Thank you. I've changed seats. I'm back over here now. Andrew, thank you for joining us. Let's just get right to it. I know you've got a tight schedule. It says only three minutes. That's a really tight schedule. Maybe you can give me the right... There you go. I know you've got a full complement, and thank you for taking the time. It's always wonderful to see you. My pleasure.
Let's just talk about the challenges of understanding where the economy is going. Brent crude is at 106 this morning. 30-year gilts popped above 6%. They're back at about 5.7 this morning. But the econ print, the GDP numbers out from yesterday, the 15th, were pretty positive. So you've got a lot of mixed signals. You've got high oil prices. You've got a backup in interest rates, which we're all navigating, but you've got an economy that seemed, the Bloomberg headlines, was surprisingly strong growth. So how do you see the balance of those two things?
No, thanks, Tim. It's really good to be back. I think we're seeing a number of things. If I look at the building blocks of our thinking in terms of our decisions on interest rates. By the way, I'm only talking about the building blocks. So if you think... I'll borrow the old Alan Greenspan phrase. If you think you've worked out from what I'm about to say what we're going to do in a couple of weeks' time, you haven't. Anyway, so I think there's a number of building blocks. On the inflation side, it's enormously frustrating. I mean, I genuinely thought pre the conflict breaking out in the Gulf that we would be back at the 2% target late spring. And I've seen nothing subsequently change my view that that would have been the case. But of course it isn't because we've had the conflict breakout. But it is the case. So we've got the first round effects of energy prices. We've got inflation just over 3%. The question for us is, what are we seeing and what to expect to see in terms of it passing through into indirect effects and then into second round effects. And two things on that. One is it's early days, so we mustn't really form any very strong views on this, I think, because it's too early. But to the extent we have seen evidence, it's quite subdued. When I go around the country, as I do a lot, talking to businesses, there is a very, very big focus on cost cutting and a very big focus on the difficulty of passing through, but I'm not taking that for granted.
So that's the sort of the inflation story at the moment, and that's the judgement we have to keep making at every meeting. What are we going to see in terms of particularly of the pass through? And the problem we have is that the longer the energy price situation goes on, because firms are clearly squeezing margins as well as cost cutting, they're squeezing margins. The longer the energy situation goes on, the harder that gets.
Second thing is, as you said, and this is good news. I mean, described as surprising, I'd say it's good, is yes, the economy is actually turning out to be more resilient than we expected it to be. The numbers are coming in now pretty consistently above. But let me put that into context. It's not as strong a growth as frankly we need. It's not as strong a growth as we've had in the past. But there are signs, and I think there's two signs in there. One is that I think household finances are more resilient. We've had quite strong household and corporate balance sheets for the last decade in Britain, really, once the post-crisis adjustment happened. And that suggests that there probably is some room for smoothing consumption.
And we are seeing some greater resilience of investment, and I think there is now the beginnings of signs of investment connected with AI. I think it's more of the capital deepening sort, which is what you'd expect, of course, in the first instance. The final bit of the story then is what's gone on, as you say, in markets and how does that affect financial conditions.
Well, going back to my point about expecting to be back at target, pre-conflict we were expected to cut rates this year by the market. I thought that was reasonable. Market was pricing a couple of cuts in. I thought that was pretty reasonable. Obviously that came off the table back in March. So we have effectively tightened policy. We haven't raised the official rate, but we've effectively tightened policy relative to where it was expected to be. And you can see that in mortgage rates. UK mortgages price off two and five years. Actually our mortgage rates have gone up. They're right pretty much the top end of the G7 actually in terms of increases in mortgage rates. So we've seen tightening of...
Well, quicker than the US, they're two and five years. So our mortgage market has changed over the last 20 years. So we used to be a variable rate market, which reset immediately. You get the lags now, but then nothing like obviously the lags you see in the US. So it's a faster transmission certainly. So we are seeing tightening of financial conditions there. We've seen obviously bond markets respond. I think it's interesting. When I look at what moves bond markets or the bond rates around the UK, it is news from the Middle East as the biggest impact. When our staff do the diagnosis, the biggest impact is really that. I don't think the UK has been sort of front and center in the last few months actually of the sort of high frequency movements. I do think there's a somewhat greater degree of stability of the commitment to macroeconomic frameworks, which is a good thing.
The thing about interest rates generally, this move back up in rates to me seems like a normalization. If you think about the last 40 years of rate cycle, this is as much a normalization. Some people seem to think that what we've had in the last 15 years is the normal. To me, that's an aberration. Would you agree with that?
Absolutely. Absolutely. No, you're right. And I think there's a second part to that argument, of course, which is very much live now, which is if you take what's going on with the investment cycle, particularly with the AI cycle, are we seeing as a consequence of that an increase in the so-called R-star or the neutral rate of interest, which would be consistent with that, obviously it would be consistent with stronger productivity growth.
You mentioned AI. Every CEO, every suite I've been in the past four months, the first question I've received is about AI. Last night, we hosted a dinner of many of our chief risk officers ostensibly to talk about geopolitical risk in the US midterm elections. We quickly pivoted to AI. I mean, this seems to be the conversation in the hallways today. Let me take this at both a macro and a micro level. From a macro perspective, it's no longer an equity issue. It is a macro risk, right? It's no longer do we see some changes in certain brands of equity, but it is certainly a macro risk. How do you think about it from a macro perspective?
Yeah. Well, of course, I'm not sure whether we're supposed to talk about AI now, are we? I did use the phrase the other day, the artist formerly known as AI, but we'll see.
Alan Turing actually, who I think is one of the really founders of it, called it I think machine intelligence, but anyway let's stick with AI because it's sort of the one we know and love. Well, we know it. I'm not sure we have varying views on it. But the macro point is important, and I think I'm going to add a little bit to the story here.
For some time over the last few months, I have been saying I think there are really sort of three issues that we must focus on in terms of the risks around this macro story with AI. By the way, I'm positive about AI. I think some combination of AI and robotics is the next general purpose technology.
Historically, it is general purpose technologies that have actually been the sort of the large leg up in potential growth rates. And I think one of the reasons we've certainly in Europe had slower growth over the last 15 years is that we probably may have been between cycles. We had the ICT internet cycle, but its contribution to growth... People say to me, well, it must be big because the internet's everywhere. I say that's the whole point though. Eventually it sort of runs out in terms of the growth contribution. And I think this is the most likely next leg up. So I'm positive about it. But look, we could obviously see market corrections. And I think there are three stories I always tell. One, the market's job is to estimate the future stream of returns. The future stream of returns could be positive, but the market could overestimate them. That's quite possible. I mean, it's quite possible.
The second one is, and again, history is a little bit of a guide here. It tends to take longer for these big general purpose technology sort of waves to come through into productivity growth than people expect. Although I think it will speed up in the economy we have now, but you go back to Edison and I think it was 40 years between him wiring a light bulb up and then coming into US productivity stats. And you know, Bob the economist, Bob Solow, always said, you can see computers everywhere except in the productivity series. And I think that's because you often need quite a lot of complementary innovation to go with the basic innovation to put it to work.
The third one is the sort of everybody's priced to be a winner, right?
But again, everybody isn't historic. I always give the example, how many of us have forgotten about Netscape, right? So everybody doesn't actually win in the end. I mean, many people do, but not everybody. I think the four... I had a fourth one which I think is very relevant to sort of today. Well, they're all relevant today, but sort of newer in my mind anyway. And it's this.
I think there's a big issue with AI now which I really summarize in two parts. It's public trust, and I think behind that, it's sort of quite philosophical, but I think public trust relies on the feeling that humans are ultimately in control of the thing, and there are now big question marks. I mean, you see things like the French refrigerator story and you think, where's all this going?
The reason I give it this, I think this public trust argument now, the more the industry necessarily is having to finance investment, to leverage up to finance investment, I think we have to be careful. These two arguments are going to come together because in a way then the ante gets upped, as it were, the financial position of the industry and therefore some potential spillovers to financial stability get more closely linked to this public trust argument.
And then there's the other aspect of financial stability in terms of cyber resiliency, right? You mentioned the refrigerators in Paris, but think about that in the financial system.
And so the FSB, where you're putting on your FSB hat, is you put out some guidelines on how you think about... and it's just guidelines on how you think about the cyber.
Yeah, I mean it's interesting to think that the last time we had this conversation back in April in Washington was the week that really Frontier AI for me anyway sort of, you know, I had a conversation with Scott Bessent that week and he said, do you realize what this stuff can do? And then I think you talk to them, you know, you talk to the sort of OpenAI and Anthropic and so on, and they were clearly surprised.
Yes. I mean they always give these sort of exponential curves and they said we've gone into the vertical part of the curve and we thought we'd get there but we didn't think we'd get there now.
Right, so it is a surprise in that sense, to your point. So since April, since we last had one of these conversations, it has been a very big issue for us in terms of cyber business. I mean, we've obviously been on cyber for a decade or more, but again, we've had to increase the pace, increase the intensity of it, working very closely with the industry, because I think there are two things now. I think there is a sort of foundational issue coming back to my sort of trust issue, which is I hope we can move towards... I'm not going to use the word regulation because I actually have some sympathy for those who say, is it really regulation? I think you've got to know why you're intervening for a public interest point of view, but I think what I would call model testing standards is an important issue. This is quite a focus of the UK government, quite a focus of us at the Bank of England. We talked a lot to our government about this. UK's in a stronger position. We've got the AI safety institute, which is I think a very good precedent in this respect. And then the second leg sits on that, as it were, which is then how you drive the implementation of this stuff down into critical industries, of which banking and finance clearly is a very critical industry in this respect. In the UK, quite a bit of what we do is part of what we call the critical national infrastructure.
So I think these two things, what we need in terms of model testing standards, both pre-release and post-release, because the models obviously move on, and then how we drive it down into how we all are having to put it into effect. These two need to work together.
So do you think we have appropriate, and I don't want to use the word regulation, appropriate adult supervision at the frontier model level to understand how these systems are evolving, both in terms of our capacity to use them to stress test our own systems or where they might go awry? And I go back to what I said earlier about complex systems always suffer collapse.
