Chief Strategist: “Most Scenarios Point to Recession” ft. Peter Berezin

[00:00:07] Peter

You’re trying to get the temperature of the water just right, but there’s a long lag between when you turn the knob, and the temperature adjusts. So, either you’re too hot or too cold. Maybe the Fed just moves the knob just right and we have an economy that doesn’t overheat, doesn’t cool down to the point that unemployment starts to rise and feed on itself. It’s possible, but if you’re an investor, I think you need to sort of have a scenario analysis in mind, and in my view, most scenarios point to recession.

[00:00:40] Intro

Imagine spending an hour with the world’s greatest traders. Imagine learning from their experiences, their successes and their failures. Imagine no more.

Welcome to Top Traders Unplugged, the place where you can learn from the best hedge fund managers in the world, so you can take your manager due diligence or investment career to the next level. Before we begin today’s conversation, remember to keep two things in mind. All the discussion we’ll have about investment performance is about the past, and past performance does not guarantee or even infer anything about future performance.

Also understand that there’s a significant risk of financial loss with all investment strategies, and you need to request and understand the specific risks from the investment manager about their products before you make investment decisions. Here’s your host, veteran hedge fund manager Niels Kaastrup-Larsen.

[00:01:36] Niels

Welcome, and welcome back to another conversation in our series of episodes that focuses on markets and investing from a global macro perspective. This is a series that I not only find incredibly interesting, as well as intellectually challenging, but also very important given where we are in the global economy and the geopolitical cycle. We want to dig deep into the minds of some of the most prominent experts to help us better understand what this new global macro-driven world may look like.

We want to explore their perspectives on a host of game-changing issues, and hopefully dig out nuances in their work through meaningful conversations. Please enjoy today’s episode, hosted by Alan Dunne.

[00:02:20] Alan

Thanks for that introduction, Niels. Today I’m joined by Peter Berezin of BCA Research. Peter is BCA Research’s Chief Global Strategist and Research Director.

He’s been with BCA Research since 2010 and has served as the Managing Editor of the Bank Credit Analyst. Prior to joining BCA Research, he was a Senior Global Economist with Goldman Sachs, and previously worked at the IMF. And Peter has also been a previous guest with us here on Top Traders Unplugged.

Peter, great to see you again. How are you?

[00:02:47] Peter

I’m doing very well, thank you.

[00:02:49] Alan

Good stuff. As I say, we’ve had you on before, so we won’t do a full recap on your background, et cetera, so anybody interested can refer back to our previous conversation about a year ago.

But I do recall last time you were on, you were reasonably upbeat on the economy, and you were kind of making the point that we probably wouldn’t get a recession until we got to the point where most people weren’t expecting a recession. And I see you now are very firmly in the recession camp. So, at a high level, what’s changed?

[00:03:22] Peter

I’ve turned heel, as they say in the wrestling business. Back in 2022, we were very bullish on the economy. We didn’t buy into the whole narrative about how the US was about to sink into recession.

And then last year, when most people were still expecting a recession, we said, we’re not going to get a recession. Rather, we’re going to get this immaculate disinflation where inflation would fall, but growth would still remain okay. But then this year, we changed our tune.

We initially moved to neutral earlier this year, and then just at the end of June, not long ago at all, we turned bearish on stocks, and we expect a recession in the US to start at the end of this year or early next year.

[00:04:13] Alan

Yeah. And I guess we got a taste of what that might look like. For a few days in early August, it looked like that was a very prescient call.

Obviously, recession fears got very elevated very quickly, and equities declined. Not an enormous decline, really, by historical standards, but there was certainly a lot of noise around it. And obviously, in Japan, we did see a very significant correction.

But maybe, I’m curious, just to get your perspective on what we saw in markets in that period. Lots of theories being bandied about; yen carry, trade unwind, dispersion trade, or was it global growth concerns or AI skepticism? How did you see that particular episode in terms of what was the main driver?

[00:05:03] Peter

Well, I think it was kind of a one-two punch. The first punch was this kind of light jab from increasing skepticism over the whole AI story. Maybe not so much that AI would boost productivity growth, but that firms would be able to actually capitalize on that and make some serious money.

Because what we’re seeing now is that a lot of these models are sort of converging in terms of what they’re capable of doing. So, where’s the competitive moat? Where’s the monopolistic power that’s going to come from that? It might not exist.

So that was kind of the first thing that markets became concerned with. And then we had that payroll report, which was preceded by a very weak initial unemployment claims print. And then those recession worries really began to surface.

Now, I don’t think we’re in a recession now. In fact, if you look at GDP tracking estimates for Q3, they’re still pretty good. They’re running at around 2%. But keep in mind that growth in the fourth quarter of 2007, right before the recession began, was also over 2%. So, recessions can often start pretty quickly and seemingly out of the blue.

A couple of years ago when we were bullish on the economy, the reason we were bullish was because we thought that there was a lot of insulation around the economy. You had these job openings, a record high level of job openings, which ensured that people who lost their jobs could just walk across the street and find new work. And that prevented unemployment from going up.

You had all these pandemic savings, over $2 trillion in excess pandemic savings. You had the fact that people were able to lock in the mortgage rates during the pandemic and so weren’t as exposed to rising interest rates. All of that kind of kept us bullish on the economy a couple of years ago, but the problem now is that those sources of insulation have really thinned out, if not completely disappeared.

The excess pandemic savings are gone. More people are faced with higher rate mortgages. And critically, job openings now have sort of normalized to very low levels, or at least low levels comparable to where they were prior to the pandemic.

So that at this point, people who lose their jobs are not going to be able to find new work easily. And hence, that means if labor demand continues to decline, we will have this rising spiral of higher unemployment.

[00:07:32] Alan

Yeah, so obviously, I mean, it seems like the labor market is central to the views at the moment. And obviously, as I say, we got an exaggerated response to the last payroll numbers, which I want to say the last one, the numbers released in early August. And obviously, I think that was partially with the Sahm rule being triggered, maybe seemed to trigger some panic.

I mean, 114,000 payrolls are by no means a disastrous outcome. I mean, reading your research, it sounds like it’s not so much the current conditions are so weak, but a lot of the leading indicators seem to be pointing to weakness. Is that the main thrust of your view on this?

[00:08:14] Peter

Yeah, that’s right. So, the Sahm rule is not a leading indicator for the economy. On average, it gets triggered after the recession has already begun.

The reason it got triggered this month was because unemployment was artificially inflated by increased labor supply. I mean, keep in mind that unemployment can rise for good reasons when people enter the labor force looking for work but haven’t found a job yet, or it could rise for bad reasons when people lose their jobs.

About half of the increase in unemployment was due to increased supply, which is a good thing. And so true underlying unemployment hasn’t increased by as much as the Sahm rule would suggest, would be problematic. But nevertheless, the other half of that increase in the unemployment rate has stemmed from people losing their jobs and not being able to find new work. So, we are seeing some signs of weakening in the labor market.

And the problem is that if this weakening continues, then we are going to see more people lose their jobs and unemployment rise. And of course, when that happens, it feeds on itself because people who don’t have a job anymore spend less, which means other people start to lose their jobs. We are moving in that direction.

[00:09:32] Alan

Yeah. I mean, I think everybody would probably agree, you know, to use the Fed’s terminology, that the labor market’s coming back into better balance. I suppose where people will disagree is, you know, is this a soft landing or is it, as you say, if the trend continues, we’re looking at something more serious.

So, what’s the key factor that gives you the confidence to say, actually, I think it is going to deteriorate from here?

[00:09:57] Peter

Well, I think the problem is that the momentum right now is on the downside if you look at kind of what’s happening to labor demand. I don’t see that momentum changing very much. In fact, it could actually accelerate because we are seeing, you know, clear evidence of weakening in key areas of the economy.

So, we just got the housing data last week. Homebuilder confidence has plunged again. Building permits, housing starts, all of that has dropped significantly. So, the housing market is weakening.

That’s obviously a key interest rate sensitive sector of the economy. One of the leading indicators for the economy typically is a weakness in the housing numbers. We’re seeing that.

You know, CapEx Expenditure Surveys look pretty weak. And so, firms outside of like the whole AI area are not really spending a lot on new equipment.

And I think perhaps most troubling, we are starting to see some cracks on the consumer front. Anecdotally, there was some evidence that companies are reporting more stress amongst lower income consumers. And you see that in the data as well.

The credit card delinquency rates are back to where they were in 2011. Back then the unemployment rate was like 8 percent. It’s like a little bit over 4 percent now.

We have the same delinquency rates on credit cards as we saw back then. The savings rate right now is like three and a half percent or so. It’s like half of what it was in 2019. So that’s unsustainably low.

People are going to have to save more. And the problem is that if they try to save more in aggregate, what that’s going to do is reduce demand across the economy.

Demand falls, then there’s less employment. If there’s less employment, there’s less income. And so, it ends up being a difficult situation where people are kind of chasing their own shadows and trying to save more, only to discover that more of them are becoming unemployed.

[00:12:17] Alan

Sure, fair enough. I mean, all valid points. I mean, if I was to play devil’s advocate and take the other side of a few of those points, you know, obviously around the consumer spending, retail sales, the last couple of reports have held up pretty well.

I think Walmart were out with a fairly upbeat assessment of the consumer. And, you know, other commentators would argue, yes, granted, delinquencies are rising, particularly for consumers at the lower income thresholds. But in aggregate, it’s wealthier consumers that drive the economy.

And we’ve had this big wealth effect from the stock market and from asset values. And obviously, you know, incomes have slowed somewhat, but income growth is fairly solid. What would you say to those points?

[00:13:04] Peter

Well, I think, you know, in all those cases, there’s reason to be somewhat pessimistic. So, take income growth, the last thing that you mentioned. We know that wage growth is coming down.

We also know that employment gains are starting to decline, which I think is inevitable at this point because labor force participation has reclawed its way all the way back to where it was prior to the pandemic. The participation rate for prime age workers is now at a multi-year high. So, it’s not going to increase anymore.

We’re not going to see the same sort of labor markets supplied. So, employment growth is going to slow. And what’s income growth is just wage growth times employment growth.

If both wage growth is slowing and employment growth is slowing, then income growth is going to slow. And if the savings rate has to rise, and as I said, the savings rate is unsustainably low, then arithmetically that means that consumption growth is going to be even lower than income growth. And so, if income growth is slowing, the consumption growth is going to slow even more. That’s 70 percent of the economy.

Now, it is true that people are wealthier now than they were before because the value of their homes has gone up. But that’s kind of a bittersweet thing because real home prices are 20 percent higher than where they were in 2019. And so, people just can’t afford to buy new homes. And because they can’t afford to buy new homes, home builders aren’t building as many homes as they were before. Existing home sales are very, very, very low right now.

So, it’s squeezing out consumption by people who are kind of stuck, want to buy a home, but can’t afford to do so. And I would also say that the ability of people to monetize the wealth in their homes is not what it once was. So, prior to the global financial crisis, people used those home equity lines of credit in order to spend.

The demand for HELOCs now is very weak and it’s not really increasing. In fact, if anything, banks are still tightening lending standards around home equity lines of credit. So, yes, people are wealthier than they were before. But taking that wealth and actually monetizing it so that you can spend it is harder than it once was.

[00:15:28] Alan

And I mean, what about the stock markets? And maybe we had the great resignation during COVID and, you know, talk of baby boomers retiring and having a huge pile of assets to liquidate. Obviously as the stock market goes up, it gives them more spending power. Is that a factor?

And then on the savings rate, I take your point, you know, at three and a half percent, that’s well below the average. But I think if you go back to kind of the ‘05, ‘06, ‘07 period, the savings rate stayed low for a few years before it eventually adjusted.

I mean, with something like the savings rate, how do you have a sense of when is that adjustment going to be this year or next year or in two years time?

[00:16:06] Peter

Yeah. I mean, the last two times we’ve seen a savings rate as low as it is now was in the late 1990s and then, you know, around 2005, 2006. Now, both of those environments featured an asset market bubble. The first one being the dot-com bubble, the second one being the housing bubble. So, yeah, I guess you can sort of bet on there being another stock market bubble maybe fueled by AI. But I think if you’re a prudent investor, you shouldn’t be betting on bubbles.

You should be looking at current valuations and becoming rather concerned because current valuations are quite stretched. The equity risk premium now is as low as it’s been since 2000. And yes, maybe it continues to fall, but for longer term investors, the risk reward of being in stocks is not very favorable at the moment.

[00:17:05] Alan

Fair enough. It’s interesting because, you know, we talked about the jitter we saw in stocks at the start of this month and it reminded me of a period in kind of, I think it was October ‘97, where the stock market fell about nearly 7% in one day. But, you know, back then that was a big move, but it seemed to have much more hysteria around a, you know, 2% or 3% move this year.

But my thought was that, you know, ultimately the turn down in that cycle was still a couple of years later. And then equally, obviously, as you say, these bubbles can ramble on for a bit; for one, two years. I know you have a quant model that is also kind of, I suppose, helps to kind of time these things as well as your subjective assessment and that also kind of downgraded stocks recently.

So, what’s the model seeing to kind of signal that maybe we are closer to a turn in equities now?

[00:18:00] Peter

I think the model is seeing just a confluence of negative factors. So, the leading indicators in the model have turned down. And again, for the reasons that I discussed, housing is weakening, manufacturing sector seems to be weakening.

We saw a drop in the ISM. We saw a drop in the manufacturing PMIs in most major economies. Some of the leading indicators for the labor markets are souring, not initial claims, they’re still OK. But certainly the trend down in job openings, the decline in the hiring rates, those are all fairly concerning.

So, could the economy just manage to achieve a soft landing? And of course, it’s possible. But I think for that to happen, it would really require that the Fed kind of calibrate monetary policy really well.

I’m kind of reminded of that Milton Friedman analogy of the fool in the shower, where you’re trying to get the temperature of the water just right. But there’s a long lag between when you turn the knob, and the temperature adjusts. So, either you’re too hot or too cold. Maybe the Fed just moves the knob just right. And we have an economy that doesn’t overheat, doesn’t cool down to the point that unemployment starts to rise and feed on itself. It’s possible.

There’s no law in the universe that prevents that from happening. But if you’re an investor, I think you need to sort of have a scenario analysis in mind. And in my view, most scenarios point to recession, not a soft landing.

And by the way, I just invite anyone who’s listening, go to, like, FRED and pull up a chart of the unemployment rate. It’s an incredibly mean reverting series. It’s usually going up or it’s going down. And when it gets down to low levels, it normally starts rising as it already is. And look at that chart, like spent 30 seconds just looking at that chart and tell me whether it’s likely that unemployment will just sort of move sideways from here when it’s never done so in the past.

[00:20:06] Alan

And from, you know, a big picture perspective, is it now, would you say that we’re seeing the kind of delayed impact? You know, to use the Milton Friedman point, are the lags now kicking in? I mean, you could also make the point. Well, recently, financial conditions have eased somewhat. Obviously, bond yields have come down. Credit spreads are still tight. Equity markets are continuing to move higher.

So, on aggregate, presumably financial conditions are weakening. So, people who look at the economy from that lens might be saying, well, maybe, you know, there is the risk of higher for longer or maybe a reacceleration of growth next year.

I mean, that was a common viewpoint earlier in the year. You don’t hear so many people saying it now, but you don’t pay much credit to that view by the sound of it.

[00:20:55] Peter

Well, the lags are clearly showing up in commercial real estate. So, the latest data that I track very clearly shows that delinquency rates are rising materially. The vacancy rate in the office segment is at a record high by a long shot at this point.

So, we’re seeing deals get made where like offices that were once worth 300, 400 million dollars are being sold for like 10, 20 million dollars. So, this is clear signs of distress, which is problematic for regional banks which are highly exposed to commercial real estate. And unfortunately, at this point, it’s not just offices.

We’re seeing stresses in multifamily construction, apartment buildings. Delinquency rates are going up there. There are stresses in shopping, even in hotels. So CRE does look very, very challenged at the moment. And I suspect we’re going to have another sort of flare up of banking stresses on the back of that.

[00:22:02] Alan

OK. And so, is the sense then that we’ll have challenges in the regional banks? I mean, is that kind of the scenario and kind of a flight at it, you know, from a deposit perspective and a significant pullback in lending or is that still a big driver given the kind of growth of private credit in the larger banks? Are those regional banks still key to kind of the functioning of the economy?

[00:22:28] Peter

I mean, it still matters. And, you know, today there was a story in the FT about how a lot of startups are going bust. So, they’re simply not able to access the credit that they need.

And I would also say that even if the Fed does cut rates, and they probably will starting in September, that might not have the therapeutic impact on the economy that a lot of people expect. For one thing, what kind of drives the economy are longer-term bond yields, and those yields already incorporate substantial amounts of easing. So that’s the first point to make.

But the second point to make is that, paradoxically, if mortgage rates do end up coming down, what that could do is unleash a lot of new supply into the market, because, of course, what happened was that when mortgage rates went up, people who had homes didn’t want to sell them because you can’t take your mortgage with you if you sell a home. And so that limited supply. But as those mortgage rates decline, we could see more housing supply.

And again, there is a fundamental imbalance in the housing market on the price side. Real home prices are 20 percent higher than where they were in 2019, even though mortgage rates are higher than where they were in 2019.

And so that increased supply down the road will be a good thing. It’ll help improve affordability. But in the near term, it’s going to strain bank balance sheets because housing is an important source of collateral for a lot of their lending. And the negative wealth effect from falling prices will also have an adverse effect on spending.

So, it’s simply not true to say, as many people have said, that there are no imbalances in the US economy. There are imbalances in housing, certainly imbalances in CRE. There are imbalances on the consumer side where you have an unsustainably low savings rate.

And there definitely are imbalances on the fiscal side. And the budget deficit is very, very large. It’s going to be difficult to do significant countercyclical fiscal policy next year, given the starting point of the deficit and the very unfavorable debt dynamics.

[00:24:40] Alan

Yeah. So obviously, you touch on the fiscal side, and that’s been one of the factors that probably surprised people in the last kind of 12 to 18 months in terms of the economic impact and, I guess, some of the spending in relation to the Chips and Infrastructure Act. Are we not, will we not continue to see a tailwind from kind of fiscal support in terms of the infrastructure bills for new chip plants? Is that a trend that kind of once it gets started will tend to continue or not?

[00:25:13] Peter

So, we’ve seen a peak in the number of projects subsidized by the Chips Act in the IRA. There’s still a large number, but it’s coming down. And so, what matters, of course, is what happens at the margin.

And at the margin, the tailwind from the Chips Act and the IRA is actually going to fade. At the same time, you know, if you look at kind of the budget proposals of state governments, they call for a substantial decrease in spending. So, the fiscal stimulus at the state and local level is going to potentially come down.

Now, what happens at the federal level? Obviously, that matters a great deal on what happens in November. But I would say most plausible scenarios, we’re not going to get a huge amount of stimulus next year.

Certainly, if Harris wins, and the Republicans retain control of the House, they’re going to block any stimulus bill. Rather, what will happen is that the personal income tax cuts, which are set to expire at the end of next year, will expire. So, there’s going to be an increase in tax rates stemming from that.

Another potentially very unfavorable scenario is where Trump wins, but the Democrats retain control of the Senate. And so, Trump won’t be able to pass any tax cuts. The only thing that he’ll really be able to do is jack up tariffs. And that’s not going to be a favorable outcome for the economy or the stock market.

So, there are definitely risks on the political side. And again, if the stock market was trading at like 15 times earnings, I would say, yeah, maybe those risks are price dead.

But by our calculations, the stock market is more than 50 percent overvalued right now, the highest level of valuation since August 2000. So that good news is well discounted. The bad news is not.

[00:27:14] Alan

OK, and that figure, 50 percent overvaluation is a fairly stark number. Where does that come from? I mean, I guess, you know, obviously any valuation model interest rates are a factor.

Is that reflecting that jump in rates that we’ve seen? And then is it that you’re much more pessimistic on future revenue growth and profits, presumably, than the market? Is that it?

[00:27:41] Peter

You know that calculation, it’s a very simple calculation. It’s just based on the assumption that profits grow in line with GDP over the coming years, and then we discount those profits using kind of real long-term bond yield with a kind of average equity risk premium.

Now, in some sense, that could actually be a very optimistic calculation because profit margins are super high right now. They’re much higher than they have been in almost any time in history. And if profit margins mean revert to the downside, then earnings growth could actually lag GDP growth, in which case the stock market might be more than 50 percent overvalued.

So, again, we’ve got this very bearish target for the S&P of 3750. But just to get down to that target would require two fairly conservative assumptions. One, that the P multiple drops to 16, which is not a big decline. The average P multiple on the S&P 500 was 15.8 between 2012 and 2019, a period which didn’t include a recession. So, we’re saying we’re only going to go down to 16, which is above the average. And of course, that 16 is going to be in a context of a recession. And the other thing that we’re saying, to get to 3750 on the S&P, is that profit estimates will fall by 10 percent.

If you look at what happened during the pandemic, during the 2008 recession, during the 2001 recession, which, by the way, was a very mild recession, earnings estimates fell a lot more than 10 percent. So as bearish as our target is for the S&P 500, it might actually be somewhat optimistic.

[00:29:29] Alan

And I mean, so much of the analysis of the S&P now has, I suppose, I don’t know if change is the right word, but you’ve nearly got two components. You’ve got the Mag 7 and the rest. And, you know, there is an argument that these companies are fundamentally different. I don’t know if that’s the right expression, but in terms of the moats that they have built, the cash flow generation, the international span, that they have market caps bigger than many countries GDP. And obviously you’ve seen, you know, a compositional shift in the index.

So how do you kind of grapple with those types of arguments? Or is that just the typical this time? It’s a different kind of argument that should be dismissed, or do you think there is a merit to those points?

[00:30:18] Peter

No, I think there’s certainly merit. And I think the current level of profit margins can be justified by the compositional shift in the index. But you still have to make kind of this assumption that these Mag 7 stocks, Mag 5, whatever number is these days, can maintain the current levels of profitability. And that’s not obvious.

Again, going back to the example of AI, like if you use ChatGPT, if you use Claude, if you use Gemini, any of these LLM models, you kind of ask a question and the answer often comes back more or less the same, which is not surprising because the architecture of these models or the neural nets and the transformers, it’s very similar. The training sets are basically the same. And so, are you surprised that the answers are the same?

Well, here’s the thing. If the answers are kind of the same and these models are sort of indistinguishable from one another, how do you make monopoly profits off that?

Perhaps these large language models will be more like airlines, which are very useful, right? You know, people can fly around the world, facilitate global globalization, but they never make any money because they’re kind of indistinguishable from one another. If you want a tech example, maybe they’ll be like browsers.

And do you really care whether you use Explorer or whether you use Chrome, kind of the same or like memory chips? So, there’s no guarantee that the profits from these models will actually materialize. And that’s a problem right now for current valuations, because when NVIDIA sells a chip, it books a profit. But when Google buys that chip, or when Microsoft buys that chip, that’s not a corresponding cost. That’s CapEx, right?

[00:32:11] Alan

Yeah.

[00:32:12] Peter

And so, the issue is perhaps that CapEx at some point will have to be written off, in which case you could get big losses appear all of a sudden.

[00:32:22] Alan

Yeah, it’s a fair point. And I know you referenced, in one of your pieces, this debate in the Goldman piece recently, which touches on some of the points you’re making. I think Jim Cavallo is the Goldman guy and an MIT economist or academic also making the point about, you know, that AI, in contrast to the Internet which was a low-cost solution and was clearly superior to kind of bricks and mortar in the case of Amazon. And you can’t say that about AI. It’s an expensive technology.

And yet, we haven’t found the big breakthrough that it’s solving. So, it sounds like you’re sympathetic to that argument.

[00:33:04] Peter

Yeah, I mean, in the late 1990s it was very obvious, right from the get go, that Amazon had a superior model to many brick and mortar stores and they could offer services faster, a wider range of products, lower price. That was obvious from the get go. It’s not so obvious right now where all the big profits from AI are going to come from.

Maybe some applications will eventually be created that give those companies that provide these AI models with a lot of monopoly power or maybe not. Right. We don’t know.

But the problem is that if you’re an investor, you’re kind of betting on the most favorable outcome and you’re kind of betting on that outcome occurring relatively quickly. What we saw with the Internet, for example, was that the productivity gains from the Internet began to be visible in the data starting in the mid 1990s, and then around 2005 those productivity gains disappeared.

But it was really only around 2005 that the profits showed up. So, it took a long time, 10 years, for Internet profits to materialize. If that’s going to be the case for AI, a lot of investors are going to be left very, very disappointed because it’s going to take a long time to make money off these products.

[00:34:27] Alan

How do you see that playing out? I mean, we’ve already seen it, I think, with some companies like Salesforce getting marked down, you know, on AI spend, but no revenue generation.

Is it we see disappointment with NVIDIA or is it that we get a recession and then the CapEx spend on AI chips gets pulled back? Or I mean, how do you see that scenario playing out?

[00:34:52] Peter

Yeah, I think it’ll be similar to what happened during the dotcom era. You know, back then, the ‘picks and shovels’ were made by Cisco and kind of other networking stocks. And they kind of held up longer than the rest of the Internet stocks.

But even they eventually crashed back down to earth when the demand for their hardware receded. So, my guess is that we’ll see the same thing happen with these large language models. The demand for those chips will come down as it becomes evident that there’s just not a huge amount of money to be made in the creation and maintenance of these models.

And then on top of that, competition will arise. Like NVIDIA has profit margins of 70 percent. That’s very tantalizing for a lot of companies to try to take some of NVIDIA’s market share.

It might take a while, but eventually I think it will happen.

[00:35:53] Alan

And from an economic perspective, there’s obviously a wide variety of views, you know, and some people saying it’s kind of going to boost productivity by one percent per annum over 10 years. And there are others saying it might be one percent over the total 10 year period. What’s your sense?

I mean, are you an AI optimist in terms of the overall economic impact longer term?

[00:36:14] Peter

I think there’s just a lot of uncertainty over this. It could be the case that these large language models are stepping stones to AGI, like artificial general intelligence; intelligence that can sort of learn from itself. If that’s the case, then, you know, these models are the last inventions humans will ever create because AI will be able to innovate by itself going forward.

If that happens, and we don’t kill ourselves in the process, of course the productivity benefits could be huge, but we don’t know that right now. We don’t even know whether large language models are the steppingstone to AGI or whether they’re kind of an evolutionary dead end. What does seem clear is that they don’t really reason in the way that humans reason.

They synthesize existing knowledge, which is great. It’s useful, but there’s a limit to how much value that can create. And in fact, what we’re seeing now is that the quality of the output of some of these models is deteriorating because they keep on adding more and more training data to the models.

And the quality of that training data isn’t particularly good. I mean, if you take the existing training sets and add like all the stuff on Twitter or Reddit, like is that going to improve the quality of the models? It might not. It might actually lead to a deterioration. So, there’s certainly scope for some disappointment along those lines.

[00:37:48] Alan

So just coming back to the economic outlook and, you know, your call in terms of a recession. Another factor I’ve seen some people talk about just of late is on the inventory side. And there seems to have been obviously a boost to was it Q1 or Q2 inventories were an important contributor to GDP and some data showing Port of LA kind of backed up with supply coming in.

And so not sure if suppliers have just, you know, ordered in anticipation of consumer demand, or there’s also some suggestion that it may have been kind of pre-ordering ahead of potential tariffs later in the year. Is that something on your radar as a possible risk factor of, you know, an inventory adjustment being a tipping point for the economy?

[00:38:33] Peter

Yeah, I mean, if you look, inventories did help boost growth in Q2, they’re likely going to be a bit of a drag in Q3 (at least based on the tracking estimates by the Atlanta Fed and others). So, you know, the Census Bureau does publish data on manufacturing inventories as a share of sales. They’re still fairly elevated, which kind of tells me that we’re not going to get like a restocking cycle anytime soon.

I think what’s going to be more kind of important is whether underlying demand for manufactured goods begins to increase. Because what happened during the pandemic, of course, was that people splurged on manufactured goods early on. They got all those stimulus checks, they couldn’t buy services. So, they bought electronics and other things.

And then we had this big air pocket of demand that appeared once the economy opened up and people shifted to buying services. You know, that air pocket hasn’t fully disappeared yet.

If you bought like a laptop or a washing machine, you don’t need to replace it after a couple of years. That’s going to come maybe like five years down the road. So, I think the risk is that the manufacturing sector kind of just moves sideways.

And meanwhile, housing weakens, consumer weakens for the reasons that I described earlier. The pandemic savings are gone. Delinquency rates and credit card loans are going up. And so, we’re seeing less borrowing there. And all of that kind of comes together and causes the unemployment rate to rise. And then when that happens, it sort of takes on a life of its own due to these sort of negative feedback loops.

[00:40:19] Alan

Okay. And in terms of how this plays out from a Fed perspective, I guess, you know, after the last employment report, you had people looking for, you know, emergency rate cuts and 75 basis points and all of this kind of stuff. But things have calmed down where it’s recording on 19th of August today.

So, we still, you know, have got Jackson Hole coming up. We’re still a bit away from the September FOMC. But I guess a near universal expectation of a rate cut of some description.

I mean, in terms of Fed policy, what’s your base case and how do you see that playing out as a recession unfolds, which is your base case?

[00:40:58] Peter

So, my base case is that the Fed will start cutting rates slowly. And then once the recession begins, they’ll cut rates aggressively. So, our target for the Fed funds rate for next year is actually 2%, which is well below kind of market expectations.

And our target for the 10-year treasury yield is 3%. So that’s a substantial decline in yields from current levels. And the investment implication, that is obvious. You want to be underweight stocks. You want to be overweight, longer-term bonds.

[00:41:32] Alan

Yeah. And do you think, you know, obviously, I think around in the, I’m trying to remember now, but, you know, in prior to the kind of global financial crisis, you know, certainly in the 2000, kind of 2002 rate cutting cycle, Fed funds bottomed around 2% or might have been slightly below that. Do you think the Fed will be keen not to go all the way back to zero, if possible, or would that be a potential scenario?

[00:42:03] Peter

Well, I think the Fed probably won’t need to bring rates all the way to zero because I don’t think the next recession will be of the same magnitude as we saw in 2008, 2009. That was more of a balance sheet recession, which required years of deleveraging. This will be more of an income statement recession where people just need to cut back on their spending, raise their savings, and that causes some problems for the economy.

In that context, the Fed probably won’t need to bring rates down to zero. But, you know, a decline in rates of 2% would still be quite substantial from current levels and certainly would create some opportunities for investors in the fixed income arena.

[00:42:49] Alan

Yeah. And I mean, we’ve had the disinflation that you were expecting. It’s been a bit stickier this year, certainly in the first quarter, but seems to be on track again.

Obviously, your view is that it’s falling sufficiently to allow the Fed to reduce rates. Have you any concerns that, you know, with an aggressive easing cycle of rates going back to 2%, that could, at some point in the next cycle, re-ignite an inflation surge again?

[00:43:22] Peter

Yeah, I think, you know, you have to be somewhat concerned about that outcome. I don’t think it’s an imminent risk. If you look at kind of underlying inflation and then you take out shelter, which you kind of have to do because shelter is a very lagging component of the CPI basket, then basically inflation is already down to 2%.

As I mentioned earlier, the wage growth numbers are pointing down because the labor market is softening. Wage growth is the key driver of services inflation. So, I don’t see a lot to worry about on the inflation side yet, which does allow the Fed to cut rates.

But keep in mind, you know, the Fed began cutting rates in late 2000. The recession started a few months later. The Fed began cutting rates in mid 2007. The recession started a few months later.

Historically, at least, the first rate cut has often been a harbinger of a coming recession because the Fed usually ends up somewhat behind the curve. And I think they are behind the curve right now.

[00:44:31] Alan

So, would you expect 25 initially and then at some point move to 50s?

[00:44:37] Peter

Well, let’s see what the next payroll report brings, if it’s another bad number then the Fed could do a 50 basis point cut in September. But we’re still so far above neutral that even if they do that, it’s not going to matter very much.

And again, you know, the interest rate that matters for the economy, it’s not the Fed funds rate. It’s longer-term bond yields. They’re already discounting a lot of easing.

So, ironically, what could happen next year is that the effective borrowing rate that households and businesses face could actually rise, even if the Fed is increasing rates, because you still have people who are buying homes, not as many as in the past, but they’re still buying homes at fairly high interest rates.

And so, the average interest rate is going to rise. It’s like, I don’t know if you remember from Economics 101, if marginal is above average, then average will rise, even if marginal is falling. It’s exactly the same issue with interest rates.

The average interest rate is going to rise, even if the Fed cuts rates, simply because the current interest rate is so high at the moment.

[00:45:44] Alan

I mean, in terms of housing markets, obviously, you’re pointing to a lot of indicators of housing weakness. Now, in many countries around the world, you know, there’s been an under supply, under investment in housing. So, kind of an expectation of a structural bull market.

Is that the same in the US, would you say? Like, would you expect that the housing market to quickly pick up on a rate cycle, a rate cutting cycle?

[00:46:12] Peter

Yeah, which is probably one reason why one shouldn’t expect a long drawn-out recession, because unlike in, say, 2008, we don’t have a glut of single family homes. But here’s a couple of points. One, as I mentioned, that disequilibrium is more on the price side than on the supply side. Prices are very, very high. They do have to come down to make homes more affordable. And two, outside of single-family homes, there definitely is a glut of apartments. We’re seeing that in a lot of the data.

Apartment list has this series on the average vacancy rate that has increased materially. And we’re seeing now, for the first time, overall construction spending in the US decline.

Now, what’s interesting is that this has happened despite an increase in the number of construction workers. Typically, construction employment lags when it actually happens to underlying construction. The decline in underlying construction over the last few months is a very, very negative signal for what’s going to happen to construction employment later this year and into next year.

And that’s another reason to expect the economy to weaken.

[00:47:27] Alan

And obviously, the implication, from your perspective, is bullish on bonds. And it makes sense from a rate perspective. I mean, the other factor that’s been influencing the bond market over the last year has been concerns about supply.

If we went last summer, it came into the market as a key driver. And then, obviously, we seem to have a change in issuance strategy from the Treasury. Whether it’s tactical or not can be debated.

How much of a concern is that for you? And do you think that can be managed? And would you be in that camp of the steepening trade, partially related to that viewpoint?

[00:48:06] Peter

Well, I think it’s definitely a concern down the road. Is it a big concern over the next 12 to 18 months? I mean, probably not.

What matters for bond yields, it’s not the budget deficit per se. It’s the budget deficit plus the private sector balance. So, if we’re going into recession and the private sector is saving more, or at least trying to save more because they’re worried about the economic outlook, that will generate the savings necessary to finance the budget deficit.

Of course, the Fed will be cutting rates, so interest rates are going to come down. That’ll make it easier to service the debt. But down the road, clearly, the current budget outlook is unsustainable.

We’re either going to need some tax hikes or some spending increases or some combination of the two. But the status quo can’t be maintained.

[00:49:04] Alan

But that’s a scenario for the next economic cycle. I mean, you don’t see, obviously, as you go into the current point of going into recession when the deficit is already at kind of six, seven percent of GDP, it’s probably unprecedented. So, and obviously, you would expect more fiscal support in the downturn. But you would see that influence in the market kind of further out. Is that it?

[00:49:29] Peter

Well, I don’t think we’re going to get a lot of fiscal support next year, partly because the deficit is so large and also probably partly because we’re probably going to have gridlock in Washington. And so, the onus will be on the Fed to ease policy. I don’t think it’s going to come from the fiscal side.

Down the road, you know, either we’re going to have like a true debt crisis or more likely we’re going to have a period of fiscal consolidation. If the Democrats win, we’re going to have higher taxes. If the Republicans win, then there are certainly some areas in social spending that they would like to cut. That’s actually not a horrible outlook for bonds to have kind of tighter fiscal policy. What would be a killer for bonds is if we end up with kind of debt monetization, where the Fed is forced to keep rates low in order to suppress interest rates.

But debt monetization would entail higher inflation. That’s very politically unpopular. So, it’s actually not that obvious to me that this will be the path that we take. It could just be more of a kind of a scenario where the budget ends up having to tighten. And that could be a long-term positive for bond yields.

[00:50:47] Alan

And I mean, you touched on, you know, how the private sector could have a willingness to take on more treasuries in that scenario. I mean, the other factor in this is the dollar and external demand. You know, historically, I guess, in a recession the dollar sometimes goes up. But at the same time, lower rates would probably be a negative. But you could have lower rates elsewhere. So, from a dollar perspective, how does that influence things in the short term and then more longer term?

[00:51:16] Peter

Yeah, I mean, you’re quite right. Like normally in recessions, the dollar does strengthen because it’s a counter-cyclical currency. And so, risk sentiment weakens during recessions. People look for safe alternatives.

What’s a safe alternative? U.S. treasuries. And so, money flows into treasuries. And that bids up the value of the dollar.

I would guess that something like that will happen during the next recession. But I don’t think the dollar will strengthen a lot. One, because it’s quite expensive now relative to purchasing power parity exchange rates.

These are sort of the exchange rates that equalize the price of a basket of goods and services across countries, like the Big Mac Index, but apply to all prices. The dollar looks quite expensive based on these PPP measures. And also, as you mentioned, the Fed is going to be cutting rates.

And the Fed has more scope to cut rates than other central banks. And so those interest rate differentials could actually move against the dollar. So, I would see some dollar strength, not a lot.

And then once the recession ends, probably a period of dollar weakness.

[00:52:21] Alan

And something, you know, obviously, we saw the carry trade unwinds to an extent. Obviously, the BOJ raised rates. And at the same time, we’ve had expectations of rate cuts.

I mean, do you think there is more of that to go? You know, if these themes of US economic weakness were to come through, you know, there’s different views out there in the market. Some people say the unwind is only 40% complete.

Have you any sense on that?

[00:52:49] Peter

Well, if you look at like speculative positioning in the yen, that has normalized to some extent. Speculators are still slightly short the yen, but not nearly as much as they were a month ago. But nevertheless, the yen remains a very, very cheap currency.

It’s close to 40% undervalued relative to PPP. So, it’s one of the biggest undervaluations in its history. And my guess is that what will happen over the next 12 months is that the US will fall into recession. The Fed will be cutting rates aggressively. The ECB will be cutting rates. The Bank of England, the Bank of Canada, everybody’s going to be cutting rates other than Bank of Japan, which barely has raised rates.

And so those interest rate differentials are going to move in favor of the yen, and the yen will strengthen. And so, our target for the yen for next year is 115 against the dollar, which does suggest quite a bit of upside for the yen.

[00:53:55] Alan

Okay. And I mean, you touched on the fact that everybody will be cutting at the same time, which is going to be my next question. I mean, we have seen a bit more divergences in the last year or so, you know, I think we had Fed raising rates and the Swiss National Bank cutting rates at the same time.

And obviously we’ve had expectation of Fed easing, at the same time, BOJ raising. And we’ve already seen like the ECB, I think the Bank of Canada cutting rates. So, something of a more of a divergent kind of picture across global central banks, you think that’s going to come to an end and everybody will fall into the same cycle once a significant downturn takes hold?

Is that fair to say?

[00:54:37] Peter

Yeah, I think once we have a global recession, then those who can, will cut rates. But there are a few central banks, like the Bank of Japan, and also Swiss National Bank to some extent, who don’t have the capacity to cut rates very much. And so those currencies kind of by default will end up being the winners, not because their central banks are hiking rates, just because everybody else will be cutting rates aggressively.

And so, I think you want to be in some of these more defensive currencies over the next 12 months.

[00:55:09] Alan

And obviously, what we’re talking about here is very much on the demand side. It’s a more conventional demand driven downturn, whereas, you know, the concerns we had in markets back in 2021, 2022, which were on the supply side, obviously, we’d had COVID, but then, you know, we had the geopolitical concerns and the war in Ukraine. At that time, a lot of people were talking about a structural shift in markets, we’re going to see more supply shocks.

What’s your take on that? I mean, you don’t hear as much about that now. Do you think that’s still a theme for the coming decade or not?

[00:55:42] Peter

Yeah, I think it’s a longer-term theme. I think kind of the whole trend towards increasing globalization has run its course. You know, globalization was able to flower during a period when the US basically controlled the global economy, where the US was the hegemon, militarily and politically.

That’s not the case anymore. Now you’ve got competing spheres of influence, China in particular, and typically, in these sort of geopolitical environments, where there are competing spheres of influence, globalization does tend to stall out. And we’re seeing some of that right now with kind of increasing regional trading blocks being crafted.

And that’s, in some sense, you know, a fairly bearish development for equities because globalization was a great tailwind to stock market performance. That’s going to become less of a tailwind over the coming years.

[00:56:42] Alan

From a commodity market perspective, you know, I guess a demand-driven downturn is not going to be favorable. I think you’ve written positively on gold recently, but you’ve kind of got those structural tailwinds to commodities, but it sounds like then a more cyclical challenging period looking in the year ahead.

[00:57:03] Peter

Yeah, I mean, I think the structural outlook for metals is generally better than for oil because metals will benefit from the green transition. You need a lot more copper to build out all the infrastructure for electricity generation, whereas oil will obviously be hurt by the green transition. So, the relative trade here is to be long copper versus oil over like a multi-year horizon.

Now over a short-term horizon, obviously the recession is going to be the dominant driver. And here I think things look pretty bad for both metals and for oil. On the metal side, you’ve got the issue that copper demand is very, very cyclical.

Chinese housing is weakening. The Chinese housing market has already undergone a major, major correction. New construction projects initiated have completely nosedived.

Now what’s interesting is that there’s still a lot of construction going on because the government has been giving developers the funds necessary to complete unfinished projects. And typically, copper is needed at the end of construction projects when you’re installing the wiring, when you’re installing the air conditioning units and things like that. So that has kind of helped support the demand for copper, but that’s going to disappear once those projects are done.

A lot of demand from China for electric vehicles. What we’re seeing now is that the whole EV market is becoming saturated. There’s a price war going on.

So, there’s going to be less need for copper there. So, over a 12 to 18 month horizon, I’m bearish on copper. I’m bearish on oil. Gold, I think, will hold its own because it’s more of a defensive asset rather than a cyclical asset.

[00:58:53] Alan

Good stuff. Maybe just before we wrap up, in terms of signposts, obviously the labor market data will be key. I mean, you touched on the regional banks and commercial real estate. Is there anything else that people should be focusing on apart from the obvious kind of payrolls, jobless claims, rates, all of that stuff? Or anything else particularly on your radar?

[00:59:14] Peter

Yeah, I think focus on housing. I think that’s very, very important. As we discussed, the recent data has been quite negative. Let’s see if that changes.

Focus on some of the manufacturing indicators, the PMIs, the ISM, et cetera. And also, in terms of the labor market, I look at job openings pretty closely. The news here, I think, has been sort of mixed. Job openings as measured by JOLTS have sort of stabilized over the last couple of months. The number of job openings on Indeed – a popular job posting site, that has stabilized.

Other measures of job openings, such as the one maintained by LinkUp, which kind of canvases the 10,000 largest companies in the US tabulating all the job openings. There the numbers are coming down. So, if we see some stabilization in job openings, that’ll be a positive development. But I want to see those numbers first before I turn more constructive on risky assets.

[01:00:17] Alan

Okay, great to have you back on to get your perspective and a lot to digest there. And definitely food for thought for investors with that quite dramatic forecast for the S&P 500.

So, Peter, thank you very much for coming back on today and make sure to follow Peter’s work because as we can tell from today’s conversation, it’s a potentially very interesting environment in the months ahead.

From all of us here on Top Traders Unplugged, stay tuned. We’ll be back with more content soon.

[01:01:15] Ending

Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today’s episode, the best way to stay updated is to go on over to iTunes and subscribe to the show so that you’ll be sure to get all the new episodes as they’re released. We have some amazing guests lined up for you.

And to ensure our show continues to grow, please leave us an honest rating and review in iTunes. It only takes a minute and it’s the best way to show us you love the podcast. We’ll see you next time on Top Traders Unplugged.