Market Narratives and the Unpredictable Alpha ft. Mark Rzepczynski

[00:00:02] Intro

You’re about to join Niels Kaastrup-Larsen on a raw and honest journey into the world of systematic investing and learn about the most dependable and consistent yet often overlooked investment strategy. Welcome to the Systematic Investor series.

[00:00:24] Alan

Welcome back to the latest edition of Top Traders Unplugged where we take the pulse of the market from the perspective of a rules-based investor. It’s Alan Dunne sitting in again for Niels this week and delighted to be joined by Mark. Mark, how are you?

[00:00:38] Mark

Not too bad, always good to be here.

[00:00:41] Alan

Good stuff. Yeah, we haven’t spoken in a while, so looking forward to hearing what you have to say and particularly as the elections are in the rear mirror now. So, we get your perspective on that. But I mean in general, what’s been on your radar of late?

[00:00:57] Mark

You know the one thing, and I sent you a chart I think the other day, is that the equity risk premium has turned negative. So, if you think that the stock market is going to continue to go higher over the next couple of years, think again. This doesn’t bode well for the longer-term view for stocks. I know there’s post election euphoria, but if you look at some of the longer-term indicators that’s not overly positive.

Beyond that, I guess I’d sort of say that, you know, if I was like a Rumpelstiltskin or someone who went to sleep and then woke up a week later after the election, you’d sort of say like the world is still here, it’s not on fire. You look at markets. We have some indications for markets, but it’s not as though that the world has gone to a catastrophe mode with a new US President.

[00:02:00] Alan

No, absolutely. It’s been a very stable handover, but that obviously is a good thing. Interestingly, I did see that chart you sent on the equity risk premium. The equity risk premium is a bit of a funny one. I mean, depending on who you talk to, there are different ways of measuring it. I know in that particular chart it was the S&P earnings yield minus the 10-year yield, isn’t that right?

[00:02:24] Mark

Yes, yes.

[00:02:25] Alan

Yeah.

[00:02:26] Mark

So, I think yes, you have to be careful about the equity risk premium argument. That’s one chart that’s gone negative. You almost have to look at an ensemble of tables for that valuation discussion. And when you look at the ensemble you’re going to get mixed reports, but generally all of them are moving in the same direction. The equity risk premium is declining.

All of the data sort of proves that when the equity risk premium declines, and then you look at what is the forward behavior of stocks over 1, 2, 5 years, it’s usually lower. So that doesn’t mean that it’s, you know, we’ll say that it’s always the case. You have to think in probabilistic terms. But generally, this is going to tell you that there’s less potential upside for stocks as the equity risk premium starts to fall.

[00:03:26] Alan

Yeah, I had Aswath Damodaran on the podcast going back a couple of months and he’s an expert in valuation now. He measures it slightly differently because I think that just taking the inverse of the S&P earnings yield, it doesn’t account for earnings growth. So, it’s just taking the current earnings, dividing it as a ratio of the price.

But that said, his equity risk premium was still positive, but I think it was at the lowest since around 2000 or at least the gap between kind of what it was pointing to and people’s expectations were kind of quite far apart in the sense that expectations of returns are high, but forward returns don’t look as attractive. And the last time they were of a similar nature was around 2000. So, that would be consistent with the same kind of analysis, albeit the numbers are slightly different.

[00:04:27] Mark

Right. And I think we’re going to talk about this later on in the podcast because one of the pieces of research that I found most intriguing, which is usually one of the things I always bring up, and what catches my attention on the research side is how analysts come up with their forward expectations or valuations for stocks.

And I think that, you know, everyone was taught, in your business school classes, your MBA classes, is that you’re always going to discount forward earnings. And what we find, and we’ll talk about this, is that when you look at reality, that’s not the way analysts behave. They’re actually somewhat backward looking. And this is done by some of the top Wall Street analysts.

So, this issue of how do you look at discounting forward earnings versus using current earnings or extrapolating from past earnings is an ongoing issue. And in spite of all of the education of MBA students of analysts, there’s still a preponderance of evidence to show that analysts look back in time as opposed to trying to forecast in the future.

[00:05:45] Alan

Yeah, interesting. Well, we’ll get on to that, I think. I know we’ve a lot of topics to get through, but a bit of housekeeping. Well, let’s talk about performance first before we get into the main topics. And so, we’re halfway through the month and it’s shaping up to be a better month for managed futures.

The SocGen CTA index up 1.6% month-to-date bringing the year-to-date figure to plus 0.9%, and the SocGen trend index up 2.7% on the month, taking the year-to-date to plus 0.55%.

So obviously, as you say, the world has remained stable in the aftermath of the elections and some of the trends have continued, particularly the rising equity trend and rising bond yield trend. Obviously, bonds had been choppy for a lot of managers this year in the trend following side, but I think probably the aggregate managers are short bonds now and benefiting from the rising yield.

Obviously, currencies have had quite strong moves, particularly in the last week with the dollar moving up sharply. All of those, I would guess, are positive for trend following. Commodities are a bit more choppy. Gold, obviously, reversing quite sharply since the start of the month and other markets, in the grains a bit choppy, wheat continuing to move lower, but copper less so. Yeah, so I would say a bit of a choppier picture on the commodity side but certainly currencies and financials trending. Any observations on performance Mark?

[00:07:16] Mark

Well, these are the things that really surprise you and the oddities that you see in markets. I think we could tie this back to how you think about building models or trend following models but first, my observations.

It’s amazing the amount of volatility that’s been sucked out of the market. So, we’ve probably had about a 20% decrease in the VIX from a level of 20% to 15%. Now, what is interesting is that the VIX volatility actually started to decline before the election.

Some of that is because of construction but we’ll sort of say that there is seen that people were starting to be less uncertain about the outcome before the actual election occurred. The second thing we notice is that if you look at the SPX volatility curve, the term structure was in backwardation – frontend volatility was higher than backend.

Now it’s gone back to contango which is the normal view in that volatility, further out in the future, should be higher because there’s more uncertainty than it is today. So that’s good. The other thing we notice on stocks is that, in spite of all of the fervor over the Magnificent Seven in 2024, we saw that the Russell 2000, sort of smaller cap stocks, actually did a lot better in this last week.

And the interesting part about that is that past macro research has shown that if there is a looser tax policy, or expectation of lower taxes, or lower regulation, that’s usually better for small cap stocks versus large cap stocks because the cost of regulation, you have to spread that over the size of your business. If you’re a smaller business, the fixed cost of regulation can actually be higher and that eats into your profits. So, there’s the expectation that regulation is going to be lessened, especially for small cap stocks.

Now the real oddity is when you look at the bond market because we still have this continued movement. We have the Fed actually lowering rates. They did it right after the election for another 25 basis points. And yet the long end of the bond market saying is, well, we’re not buying what you’re selling. What we’re seeing is that trend is continuing.

So, if you base your trades on macro relationships, classic relationship, if the Fed is loosening, so you got easy monetary policy, we should sort of see that there should be a decline in rates along the entire yield curve. And that’s not the case.

[00:10:19] Alan

Yeah, absolutely. And we had some comments from Powell last night, I think kind of dampening expectations of further rate cuts in December and beyond, given the strength of the economy. So yeah, interesting, we’ll see how that plays out.

But yeah, I mean certainly the move up in bond yields, it feels like every time bond yields go up, people start talking about debt and deficits again, and then when they go down they kind of forget about it. But you know, we’re still probably back in the range. Like the peak has been about 5%. Okay. They’ve been below 4%. And where are we now? About 4.6% or so, I guess. So maybe towards the higher end of the range. But yeah, it’s definitely one to monitor.

[00:11:05] Mark

And I think part of this is that you could talk about what the data is telling us versus what policy prescriptions that they think that is going on at the Fed. And I think from a policy point of view, I don’t know if you remember, I think most people forgot that the Fed actually came up with this idea that they were going to have the average inflation rate policy.

So, they said, well, if we’re below the 2% target, what we could do is we could allow inflation. It could have some flexibility, it could go above the 2% target, but vice versa is, let’s say we’re above 2%, where we want to have on average 2%, we could allow it to go below 2%, and it could get close to maybe zero because we’re thinking of what is the long-term average. So, whatever happened with that?

[00:12:04] Alan

I have a feeling, I think Powell got asked about that recently, you know, is there a scope, or an allowance for that? And it was a no. I think that has been abandoned, if not completely publicly, but certainly the framework no longer seems to apply.

[00:12:21] Mark

And I think that there has probably been more emphasis placed on like uncertainty measures. So, I think that there’s Professor Bloom, out of Stanford, who has the news and policy uncertainty indices that he’s developed for all across the world. I’ve looked at this from a research perspective, and it has some usefulness. It’s also sort of mixed because it really does, you know, bounce around a lot.

But it’s clear that if there is a lot of policy uncertainty, and even from the comments from Powell of what will happen, what you see is that there should be a greater risk premium, or term premium in bonds, especially in the longer end. And second of all, then you could have a divergence between what the Fed would like to see the markets do and what they actually do. So, I get to bring up one of my favorite terms here that we often talk about. The question is, the bond vigilantes, are they back?

[00:13:32] Alan

Yes. Well, it’s interesting, I was thinking about this one because I saw somebody writing about the bond market vigilantes and I didn’t like the whole idea of the bond market vigilantes and didn’t buy into it.

But I mean, one thing that’s interesting now, you talk about yields rising and some of the rise in yields is real rates and some of it is higher inflation expectations. And I mean, I don’t know. Certainly, the Fed are sensitive to higher inflation expectations. So, if the bond market pushed TIPS to such an extent that inflation expectations were rising, then that is something that could actually influence the Fed’s thinking. We normally think of the bond market vigilantes influencing the Treasury or governments in terms of fiscal policy. But actually, if inflation break-evens rise sufficiently, that very much influences the Fed. So, I think certainly, we are back to a period where what happens in the bond market is quite influential.

[00:14:39] Mark

Well, I guess the issue is always in the details, and you say, well, what does it mean to be a vigilante? Because on one hand we say we have higher break-even inflation. So that would say the expectations are higher in the long run.

If you look at Citibank or Bloomberg Surprise indices that’s moving higher. And when the Surprise index moves higher that tells you is that we’re getting announcements that are higher than what we expected. So, we’re having surprise growth on the upside, which I think usually is a good trend signal that’s going higher.

If you look at the financial conditions indices that the Fed puts out and other Wall Street firms go out, those are actually showing that they’re high. So, financial conditions are not tight, they’re relatively loose. That’s another good trend indicator. The Inflation Surprise index is higher.

So, what you have is that you sort of say, we could talk about why you want to be a vigilante. But, at the very least you look at the data that’s coming out and you look at the trend in that data, then it would suggest that yields should be higher. And I guess this is probably sort of a, we’ll call it a macro trend follower.

I would sort of say that economic trends beget price trends. If you see the trends in economics are all moving higher, then that’s going to actually then translate into the prices we see in markets. And that’s exactly what we’re seeing in the bond market right now.

[00:16:32] Alan

Interesting. Yeah. So, you’re thinking that this move up in bond yields can gather momentum, or do you kind of formulate those kind of views as a systematic trader?

[00:16:48] Mark

Well, as a systematic trend follower I can’t judge how far this is going to continue but we’ll sort of say that you’re more wary if the fundamental data is at odds with what the trend that we’re seeing. So, let’s say that we saw that there was a slowing of economic growth. If the data was sort of saying that that the economy is not doing well, financial conditions are looking a little bit tighter, the Surprise Index is negative, then you sort of say like, okay, I understand that bond market participants are saying that rates are going to be higher, but that’s not borne out in the fundamental data. In this particular case you look at sort of now casts. So, look at the Atlantic Fed now cast their GDP. It’s showing up close to 3%.

So, we’re looking as though growth is strong in the US. All of these other short-term indicators on the surprise side are also headed higher. So that’s all reinforcing a trend which you’d say that it could last longer. And the interesting part too is that we’re actually seeing the dollar stronger, so that the dollar is reinforcing the expectations that are in the bond market.

So, if you sort of said like, well the dollar is selling off, then they’re sort of saying that, okay, that we’re probably going to see that interest rates are going to start to fall precipitously, it’s going to come down and growth is going to slow – so, if you see dollar depreciation. We’re seeing dollar appreciation along with higher yields.

So, when you think about it, you’re always worried about, let’s say, a financial crisis. And so, going back to the bond vigilantes, they’ll say, well, we got to worry about a fiscal crisis. But if that’s the case, you would expect that rates should go higher but the dollar would go lower so that people didn’t want to buy the safe assets. That’s not what we’re seeing.

[00:19:01] Alan

Well, we are at an unusual point in the cycle where you know, we’ve had an election and there’s an expectation of significant policy to come but we haven’t got the details yet.

So obviously, you know, markets are trading on what we’re seeing in the high frequency data and as you say that the stronger profile of GDP, but also presumably building in a lot of expectations around what’s to come in terms of, we know what the broad parameters are – tax cuts, deregulation, tariffs and you know, deportations. But precisely how that’s going to play out I guess is not known yet.

So, I mean that presumably is, as you say, with respect to small caps, heavily influencing what’s driving markets at the moment.

[00:19:47] Mark

Now the thing that I look at, and there are different ways to measure this, is the term premium in the bond market. So, when you think about it, when rates were very low and then there is this negative correlation between stocks and bonds, term premium was negative or there was no term premium.

So, when you think about decomposing, what is the yield? There’s going to be the real yield, which will sort of be associated with long term growth. There is expected inflation and on top of that there should be a term premium. The term premium would say what’s the risk of holding a longer term asset?

And what we’re seeing is that the term premium across the board is starting to move from negative to positive. This is not something that just happened overnight. But if you start looking at the trend, all the trends across the curve show that these term premiums are moving higher. Consequently, there’s a potential for rates to be higher.

Now could this be about policy view that we’re expecting that there’s going to be continued large US deficits? I think that’s partially the case. And then, when you think about it, we’re having more of a two way market, which means that people are going to be wanting to be compensated for the risk of holding treasuries. When I say by a two-way market, it’s that we have quantitative tightening. The Fed has actually slowed the amount of selling of treasuries in their balance sheet. But that being said, they’re not buyers, they’re sellers. Okay.

We have deficits that are going to continue. And that seems as though that it’s probably likely under Trump. It was also going to be likely under both. But then you look across the board, from a bond vigilante perspective you have financing demands that are in the EU and you still don’t have a capital market union. You have China that, while they have new fiscal policy, they’re facing what we call a balance sheet recession because they have a lot of companies that have to repair their balance sheet. So, if they generate more cash flow, they’re not using that to invest in the business or spend, they’re using that to pay down debt. And then in the US, we’re still having large deficits even though we have fairly strong growth.

So, you put that all together and what we’re sort of seeing is that it’s not a good environment to be a bond holder in this environment right now.

[00:22:38] Alan

Yeah, I mean there are a lot of factors going on here. I mean you touched on a number of them there in terms of inflation, break evens, real yields, et cetera, the term premium.

Another story I saw this week was around the relationship between bond yields and swap spreads. I don’t know if you’ve been following this. The swap spreads have been trading lower relative to bond yields of late.

I mean, it used to be, back in the day, they used to always be higher because there was that interbank funding, counterparty risk. But now because it’s secured lending, it’s reversed. But I think the point that was being pointed to in the piece, it was an odd lots piece, they were saying there’s growing expectations that treasuries will be treated differently from a capital perspective.

You might remember, I think they brought in a different rule during COVID that if banks held treasuries, they were not a risk weighted asset or something like that, and then that was taken away. But there are expectations that might be brought back. And I guess maybe that’s been driven by a concern around the bond market and they want to make it more attractive for banks to hold more bonds. But is that something you’ve picked, picked up on or seen that story?

[00:23:59] Mark

Well, that’s an ongoing research topic that a lot of people are focusing in on from different directions. And one issue is, do we have a shortage of safe assets?

Okay, now we have treasuries as a safe asset because, in some sense, regulation makes it a safe asset. If you sort of say that you need to hold treasuries for collateral, then you’re going to have to buy the treasuries. And if there’s a shortage of treasuries, then you have a shortage of the safe assets.

But when you look at foreign buyers of treasuries, they still have to take currency risk, which, you know, you sort of say like, well, you know, over the next four years, what’s the currency risk for a dollar? So, a treasury denominated in dollars may not be as safe. And then if we sort of see that instead of rates being very stable or being very low, if they’re now moving within a band at a higher level, we know that generally the higher the level of rates, the higher the volatility of rates. We know that the move index has been extremely volatile, even probably more so than the VIX.

And so, there’s more inherent risk in holding treasuries than we had before. It could still serve as a safe asset. But when you look at this on the balance sheet, it’s a real economic risk for holding treasuries.

A perfect example is, let’s go back in Silicon Valley bank. One of these banks said, well you know, I got all these new deposits, I can’t find places to lend. Oh, so I guess I’ll buy treasuries out on the curve. And what happens is that we have rates go up.

So, all of a sudden now, if you held them in your mark to market account as opposed to your book value account, all of a sudden you find out, wow, this is blowing a hole in my balance sheet. So, gee, I wish I held more, in some sense, regulators could say; and the banks themselves, gee, I wish I held some more capital against that because now this is riskier than what I expected.

[00:26:22] Alan

No, that’s right. But anyway, we’ll see. I don’t know how that’s going to play out, but certainly one thing, another thing to monitor in the whole bond market mix.

We’ve spoken a lot about bond, so let’s maybe pause there and we did have a question this week. So maybe let’s deal with the question before we go into the rest of the topics. And the question is around cryptocurrencies and allocating to crypto and allocating to trend following. So, Bitcoin has just hit an all-time high and the readers and I have seen some new trading firms applying trend following to bitcoin and exclusively to cryptocurrencies.

I would be interested to hear about the appetite amongst allocators for trend following of cryptocurrencies and whether it is coming from investors with experience in trend following or a different type of investor. So, I’ve certainly seen a bit of this. Have you seen kind of dedicated trend following programs in crypto, Mark?

[00:27:25] Mark

Well, I’ve seen some firms trying to sell crypto within the separate asset class building portfolios. I’ve done some analysis on this. I sort of say that trend following models do work in crypto. So, one, if you’re a trend follower and you want to apply it to crypto, it does work. Okay, that makes sense. Now the question at hand is, what do allocators think about this?

Now allocators, they give a lot of leeway for CTAs to trade whatever they want. So, you have some that trade almost a limited basket, others trade, you know, 50 plus all around the world. At some level the traditional view of a trend follower is that, well, I really don’t care what the asset is as long as it trends. You know, you just give me a symbol and I could be able to trade it.

Allocators are probably a little bit more suspect about crypto because it’s still a relatively new asset for many. Some people have strong opinions which actually sort of said, well, I don’t want to trade that from a more structural perspective.

If you look at how you have to trade crypto within a portfolio of other futures, there are some operational logistic issues that make it more difficult. One is that you have to have a prime broker that’s willing to be able to actually clear that too. You don’t get the cross margining that you would like. So, there are operational details that make it more difficult that an allocator or a customer, if they started to go into those details, they would have some concerns.

[00:29:23] Alan

And from the purely investment perspective, I mean generally with trend following, the idea is to apply it in as many markets (well, okay, I know this is debatable), I was going to say as many markets as you can. I know that’s an old debate in itself, but certainly across many markets. So, I mean, does it make sense to just do trend following on a small number of cryptocurrencies?

[00:29:46] Mark

There is not a great demand for allocators for crypto only.

[00:29:47] Alan

With trend following.

[00:29:47] Mark

With trend following, because one, the market surprisingly doesn’t have that much liquidity. So, you really can’t grow that large. And then what you find out is that yes, you can build a portfolio with a number of crypto names, but once you get beyond the top few, liquidity then becomes a problem and then, actually, it becomes harder to trade. You know, the cost of trading becomes a lot higher.

So, I think that you know, generally allocators have taken more of a cautious wait and see approach to looking at crypto only exposure and then crypto within larger CTAs. Even with that they’re probably sort of say, well, we want to at least hear more about it, and they have some concerns from an operational perspective.

[00:30:47] Alan

Yeah, very good. Okay, well, let’s move on.

Now, I know you had a number of topics. We talked a lot about the post-election environment already and I know within that you had identified just some thoughts around the risk cycle. Did you want to touch on that and Minsky before we moved on to the other topics?

[00:31:10] Mark

Well, the one thing I always want to sort of comment on that is always in the back of my mind is the fact of what is the risk cycle which is different than the business cycle? So, the risk cycle is, are we in continued periods of relatively low volatility? And we’ve had some spikes. We’ll say in August of this last year there was a big spike in volatility.

But generally, volatility, if you look at the VIX, has been low. The move has been higher and then I think that has a lot to do with the inflation and pivoting by the Fed. But generally, we’re in the low volatility environment, and especially, I would say, a low downside volatility environment.

And what research tells us is that low volatility actually begets more leverage. It leads to higher growth, but there’s also more risk taking. And because of that then you plant the seeds of potential risk. So, a lot of times what people will say is, oh the VIX is spiked, then stocks will go down. Yes, that’s the case. But you have to look at what happened prior to the spike in the VIX rise.

You sort of say if there’s a period of low volatility then that causes people to take on more risk and that leads to a situation that when the spike does occur there’s going to be greater harm. So, the one thing I always am cautious and very careful of looking at is how is volatility propagating through the economy? What is volatility doing and what’s the amount of leverage or risk being taken by different players like hedge funds or CTAs?

[00:33:10] Alan

Yeah, well for sure you would say that, at the moment, we’re in a risk seeking environment. If you look at things like credit spreads, they’re pretty tight, isn’t that fair to say?

[00:33:19] Mark

Absolutely. So, let me put this way, we had a little bit of a scare earlier in the year where people were talking about that we were going to have a recession. And you look at the credit spreads on junk, you look at the credit spreads, even just on higher quality corporate. They’re extremely tight. And if we look at some of the consumer side of delinquencies and defaults, especially in credit cards and auto loans, we’re starting to see that those tick up.

So, we’re sort of getting some evidence. And let’s say, if you look at this election cycle, there seems to have been this undercurrent that the economy is doing well for some but not doing well for many others. Sort of the average person that they’re being hit with higher inflation, that they have concerns about their real income and how they’re going to be able to pay their bills. So, when you think about that, how should that translate into credit spreads across the board?

And we’re sort of seeing that credit spreads are extremely tight and that even if you look at smaller cap stocks, the number of stocks that are having negative earnings or very poor earnings (not the top firms, but you know, the average firm), that it would suggest that the equilibrium credit spread should be higher.

[00:34:58] Alan

Interesting, because I was looking at that chart there recently, and to your point about the Minsky moment, the point where you tip into the risk averse environment and the buildup of leverage causes the spike in volatility. But it’s not the case that we’ve had a very long period of uninterrupted risk seeking like say the late 90s.

If you look at say, credit spreads between ‘95 and 1999 or 2000, they were at a low level consistently. I mean we had obviously a blip in 2022. We had the same in 20. So, I wonder does the fact that we’ve had a couple of spikes in credit spreads, even though they were relatively short lived, has that kind of blunted that buildup of leverage in the system or not? Any thoughts on that?

[00:35:50] Mark

I think that there’s still a strong buildup in leverage. Okay. And leverage is insidious because it sort of seeps into all parts of the economy in ways that you don’t expect. So, one is, we’ll say, consumers are in pretty good shape because a lot of them all refinanced to fixed rate mortgages in the US at low rates.

And so that’s the reason why we’ve got housing gridlock. If you own a home at, let’s say a 2.78% mortgage, you’re not going to sell it to then go buy a new home where you’re going to have like a mortgage that’s more than double. Okay. The rest of the world, outside the US, has a more floating rate as opposed to fixed rate mortgages. So, that means leverage is going to be affected by, you know, higher rates and higher nominal rates.

We also sort of see that if, let’s say, you have stability, then you’re going to go out and use your assets as collateral to borrow. And that leads to, when we do have sort of a shock, that then the underlying collateral is going to decrease in value, which causes a problem. We can look at, for example, at China. China actually had tremendous growth over the last decade. But you look at what they’re suffering as a balance sheet, you know, problem or recession. It’s similar to what we saw in China in the early 1990s. There was a big buildup of leverage,

So now they’re finding out that they had a shock, and so now you’re going to have to sort of clean up your balance sheet and then that really has an impact on your future growth. So, we know that if you have extreme periods of low volatility that yes, you are going to have spikes. That’s not the issue.

The issue is that when you have low periods of low volatility, that there can be a situation where now there could be embedded excessive credit or excessive leverage, that when a shock occurs, you’re going to be more sensitive to that shock than you were before.

[00:38:16] Alan

Yeah. Okay. Well, let’s move on.

I know you had a couple of pieces and so a couple of topics around the area of, I guess, information and perfect information and rationality. I mean, will we take them both together or do you want to deal with the kind of rationality side first, or how do you want to discuss this?

[00:38:37] Mark

Well, let me talk about some of the research that I found and sort of tie it back to, you know, quantitative models and trend following because people are always interested in how to do this. They say like, well, you know, why are you a trend follower? Why do you behave the way you behave? Certainly, you should use more information than just prices to be able to make your decisions.

I’m probably not a complete purist in saying I only use prices. Although I always sort of say that prices are primal. So, prices are the most important thing to look at first.

But you know, there’s always different information you start to read that gives you pause about, you know, sort of the conventional wisdom that people are rational, markets are efficient, that we use all the information that’s available to us really well. So, what does that mean?

There’s an interesting piece from the Elm Capitalist. It’s from the same author who wrote the book The Missing Billionaires, I think, which is a really interesting book. So, what the title of the research was that The Crystal Ball Is Not Enough to Make You Rich.

And so, they ran an experiment, and they said let’s assume that we could give a group of people perfect information of what will happen and of the news headlines that will happen tomorrow, today. So, now you could run this experiment because you could be able to say like, okay, we’ll tell you that this is what the headline is going to be tomorrow. Okay, so it could tell you what the unemployment rate is going to be. It could tell you some other macro data. It could tell you stuff about an individual stock. So, we’re going to give you the perfect information.

So, Alan, let’s say we could run the experiment and I’ll give you the headlines for the Financial Times for tomorrow, today, do you think you could make money?

[00:40:53] Alan

Well, it’s an interesting ask because I did come across this research. I’m not sure if it’s exactly the same as today. I think it was. And they gave a link where you could play this game. And, as you say, you were presented with the headline for the following day and then you were asked how to allocate your capital between bonds and equities so you could go long or short the S&P or long or short treasury, and you do it about 20 times and then they calculate your return and Sharpe ratio.

So, I think I was positive. My Sharpe ratio wasn’t particularly high. I mean the only thing about it is you could kind of identify the periods and say, well, you know, roughly speaking, equities were rising or falling over that period or bonds were rising. So, that kind of did influence your thinking. But certainly having that information, it didn’t make it easy to make money.

It was still whatever it was, it was pretty marginal either way. So, I think you’re right. It does highlight that having the information doesn’t necessarily, obviously, guarantee any success in trading.

[00:41:58] Mark

So, I’m, I’m surprised, Alan. I thought that you would have been making money hand-over-fist if I gave you the news beforehand. I also took the test. I will… Let me put this way, I didn’t do as well as I expected. So, that could be an issue of overconfidence though.

But we’ll sort of say that what we find out about markets is that you can resolve information uncertainty, I give you the headlines, but you can’t resolve your reaction or behavioral uncertainty. So, you could tell me what the news is but that doesn’t mean I know what the reaction will be. And this is fundamental to how we build models and how we think about it.

So, a trend follower, in some sense, would sort of say that, I don’t know what their reaction is going to be. I think that the link between information and what the markets will do is sort of loose. And because it’s loose, that’s not something I want to use. I want to sort of focus in on the price itself. And so, I’ll just sort of say I’ll discount what the information is telling me. And I’m going to go to the behavior of the markets itself first as a reason for doing this.

And you say like, well, what’s really going on here? And it’s interesting, and I studied this area for a long while. And one of the people that had a big influence on me is a professor from the University of Chicago in decision science. His name was Robin Hogarth. He died this year. So, everybody talks about Tversky and Kahneman, but if you go one down, who were all the people who are great? Robin Hogarth was one of them. And he talked about that there could be a ‘kind’, or ‘wicked’ learning environment. And I think he was sort of Brit. Somebody sort of used the word wicked.

So, a ‘kind’ environment would be that there’s a good link between the action and then the response of some system. So, let’s say that every time unemployment goes up, here’s what should happen to the marketplace. And so, in that way, if there’s a good link between cause and effect, you can learn.

Okay, now a ‘wicked’ environment would be that, well, it seems as though that sometimes when inflation goes up, this is what happens in the bond market. But it’s not exactly a one-for-one correspondence. That’s a ‘wicked’ environment because then it’s harder to learn.

And he actually wrote a book called The Myth of Experience. So, people can have experience in the marketplace, but that doesn’t mean that they draw the right conclusions or make the right judgments from their experience. So, I sort of half joke is that you could have 30 years worth of experience or you could have, you know, one year’s worth of experience 30 times.

So, I could be experienced. But that doesn’t mean that I can make the right judgment of what’s going on. And to put this more scientifically, one of the big issues in finance right now is to better understand causal influence. What are causes and how does it influence other variables? And I think that this is an issue that’s really important for any model builder.

[00:45:57] Alan

Interesting. Well, you mentioned something interesting there, which I just thought of with respect to a kind or wicked learning environment because obviously we’ve heard a lot about machine learning of late and I guess that’s, you know, machine learning AI strategies learn based on the environment.

So, as you say, you can have environments that are better or worse for that and I guess a better one is when the relationships are more consistent. Obviously, when you go to periods where the relationships had that which had been established start to break down, that’s going to be more challenging for those types of learned strategies. Whereas I guess trend following doesn’t rely on that learning to the same, well, to any sense. It’s purely based on the price. So, that’s an interesting one, isn’t it, with respect to the current talk of machine learning and those types of strategies?

[00:46:51] Mark

And when you think about it, what are the causal influences? For trend following the causal influence is very contained and it’s well defined. We’re not saying that prices yesterday caused prices today to move higher. We’re just making the observation that when you see a set of prices that are moving higher, they’ll probably continue to move in the same direction.

When we talked about it like inflation or Powell, and making a comment, there you’re looking at causal inference. You’re saying is that he is the cause that will influence how markets will behave tomorrow. And so, we’ll sort of say that trend following is more observational as opposed to causal and how they build the models.

And so, I don’t want to go too deep but it’s sort of interesting how you have a different approach and to deal with one of the tough problems. Now, a trend follower may not put it this way, but he’s sort of saying, because I think that that causal influence in economics and finance is so difficult, I avoid trying to find those causal inferences.

[00:48:12] Alan

And obviously they can shift as well over time. And, you know, we’ve all read those kind of articles that highlight that yesterday the stock market went up because of weak earnings, raised hopes of a rate cut and then the next day the stock market went up because of positive earnings. So, the causal relationships themselves are not necessarily consistent. And we’ve gone through prolonged periods where bad news is good news and then we shift into an environment where good news is good news. So, yeah, I guess there are all the challenges of modeling more macro related data.

[00:48:52] Mark

Right. The other comment on that piece of research, which is critical for all of the listeners and for everybody thinking about it, is that you had the perfect information. You still went bankrupt in a number of people who ran that experiment.

And so, what you find out is that you can be an average forecaster, but if you’re very good at position sizing and risk management, you can survive and still make money. You could be a very good forecaster but if you don’t position size correctly, you could still go bankrupt.

So, when you think about how you build models, everyone says, well, I’ve got to get my forecast right. And I can tell you with certainty that you will not be right a lot. So, you know, even if you’re making a 55 or 45 correct, it’s pretty darn good in some circles. So, we’ll sort of say that the real issue is how do you position size to stay in the game? And that’s probably as critical as anything you do to try to find out what you think prices will do tomorrow.

[00:50:07] Alan

Yeah, absolutely. And in that study and the test that we both took, there was quite a lot of variation in terms of how big a bet you could make on the S&P and treasury. So, I think I found that, you know, I made some big gains early on and then sized too small later on. So, definitely that had a big, big influence.

I know you wanted to talk about another piece of research around equity analysts, which you touched on earlier. So maybe we’ll move on to that.

[00:50:40] Mark

Yeah, well, this is the one that I think is again, this is that we want to tie it back to trend following a little bit. But you know, I think that this goes back to the idea that we expect, or we think that all of the Wall Street analysts, that these are the best of the best, that they’re constantly looking at trying to forecast earnings forward and then they discount them back. Okay, that’s how you’re going to get evaluation.

The one mantra that you learned in business school is that you take your cash, you know, future cash flows, you discount them back. And that’s your present value of your project.

You said like, well, lo and behold is that we find out this is that FINRA requires that you have to say what the kind of model is that you use for valuation. So, you don’t have to give the exact model but is it used as some kind of multiple? Is it discounting as a discount model?

What we find is that most of these models actually look at trailing PE. And then they’ll goose it up or goose it down, and then they’ll adjust it as they get a new earnings estimate.

So, what happens is this is that a lot of the leading forecasts, when they come up with their valuation models, they’re actually looking backward. And so, lo-and-behold, we find we have post earnings drift. And we also have, trend following some stocks, why should that be surprising at all? If we find out that all the analysts, actually, are backward looking and they’re looking back, they’re using fundamentals to make their evaluation.

So, if you sort of say like, gee, I listen to all of these smart people and they tell me about why I shouldn’t be a trend follower. The real question is that let’s look at how these actual smart people behave. They’re actually looking at trailing PE. So, looking at price is probably not a bad approach. And it actually is, you know, maybe the correct way to look at things given the fact that the analysts themselves are actually also backward looking.

[00:52:46] Alan

And are there FINRA requirements around this? Is that part of it?

[00:52:51] Mark

So, there’s attestations about what type of model you use. You don’t give the actual model, but you could bucket it. So, what these researchers are able to do is actually then characterize all the analysts and say okay, here’s the kind of model approach that they have. So, it’s up on my blog that I listed that piece of research. But it is called Expected Earnings Per Share Times Versus Trailing PE is the research. I think this is very pathbreaking and it’s very interesting. I don’t know if it’s going to get a lot of play because it’s away from conventional thinking, but I think it’s useful for you to take a look at this and I think you’ll find it intriguing.

[00:53:44] Alan

Interesting. From the perspective of stock prices, so, they’re monitoring the past trend and extrapolating that forward. Is that it? As opposed to. Yeah. Okay. Yeah.

[00:53:56] Mark

When you think about it, I should be trying to, if I’m looking at Microsoft stock, what I should be doing is I should try to extrapolate and I should try to estimate what I think the future earnings will be over the next year, two years or going forward. Right. And then, after I sort of extrapolate what I think that, oh, this business unit is going to make more money, this one’s going to be less, here’s what I’m going to do to build this model, and then what I’m going to do is I get to aggregate all those earnings up and then I discount it by a sub rate to get to the valuation today.

We’ll sort of say that their ability to make forecasts about those future earnings, I think that they realize, is so poor or maybe this process is so difficult that they say, well, what I should do is to say, let me take what’s going on in the past. It’s the trailing PE. I can make some adjustment about what I think a proper multiple should be. I could tweak what’s going on in the past and then that’s what I’m going to do is my valuation forecast. And then you’d sort of say that’s what’s really driving it. And then, as a new earnings announcement comes out, I’ll tweak my valuation model as the new information that comes in.

[00:55:15] Alan

Okay, interesting. Well, it’s certainly worth checking out. I’ll have a look at that. But I know we’re as, we’re moving up on time. I know there were, there were a couple of other topics you wanted to touch on. One was around narratives, possibly themes as well. Where did you want to go on that?

[00:55:33] Mark

Well, this is an issue is that there’s been some research that’s been coming out from a number of people. I know State Street bank, their group and has actually come up with a product that looks at narrative to be able to sort of look at narrative within stocks or macro themes and then be able to say, well, does that have an impact on stocks? And so, this started, you know, with, you know, Robert Shiller who is obviously known for irrational exuberance. And so, he started looking at memes in the marketplace that could explain bubbles. So, we could say that there are themes or memes that start to occur that could cause excesses in valuation that lead to bubbles and then crashes.

But now what people said is that with, you know, LLM and with our artificial intelligence, we can be able to start to look at all of the narrative information that comes from different news sources and start to aggregate them up into sentiment.

Or we could sort of look at are there positive news stories, negative news stories that could be done on a macro level, a micro industry level, or even a stock level. And what they’re finding out is that when they look at all this narrative information is that if you look at the trends in narrative, which would be the aggregation of news stories, there’s actually an under reaction in stocks.

So that, in some sense, what we could say is that you could use narrative or the trend in information and news stories to tell us what might happen to future prices. Now, what that is another way of saying is that trends in news lead to trends in prices.

So, I feel, and I’ve spent a fair amount of time looking at this, that this is a very interesting piece of research, but the barriers to entry are really high. You’ve got to spend a lot of work to be able to do this. There’s a lot of data, you have to process. But the takeaway that I came away with is that if I can aggregate, and I can look at the themes within news, I can be able to sort of tell us something about what I should do to my asset allocation and what should happen to the trends in prices. So, you may not be able to do that. It may be a nice adjunct to what you do.

What it still goes back to is that, because of this slow diffusion of information in marketplaces, that there’s ability to see trends that can be captured and be exploited.

[00:58:39] Alan

Interesting. I mean, it ties into a few different areas. I mean, one of the basis for trend following being profitable in the first place is this idea of anchoring, and that investors do under react to news early on in the trend and later on overreact. So, that would seem to be consistent with this research. And you do see that in the market as well.

But also interesting in the context of the price action in the last week or so we have new themes emerging in relation to policy under Trump. It’s hard to say it’s been an underreaction. So, I don’t know. How would you interpret recent market action from the perspective of this narrative lens?

[00:59:27] Mark

Well, you know, we’re looking on average. And so, I think that the most recent is that what you’re sort of saying is that, clearly, as soon as the election results came out, boom, you had this big jump in price. But then, even in that, when you look at the narrative, we’re starting to develop themes that we’re sort of seeing. There are both good and bad themes, you know, and just not getting into the politics where you say, like, well, one theme is that, okay, we’re probably going to have or maintain tax cuts. Okay, we know that’s being disclosed in the marketplace.

We have the theme that there’s probably going to be more efficiency in government and less regulation. Okay, that theme is being played out. We’re also likely to continue to see deficits. Okay, we’re seeing that being played out in the bond market.

So, we’re seeing different themes starting to play out. And what you asked earlier is that do I think that the trends that we see today are going to continue? And the answer is, well, as long as the price action is behaving the same, yes, I’m going to continue to follow that. If the price action changes you might get stopped out or I’m going to have to reverse trends.

But I would sort of say that when I look at the combination of this, we’ll call it narrative economics or narrative information, the fundamental information, when you combine those two, is to say that they’re all moving in a consistent manner with the trends that we’re seeing now. And so that would suggest that trends will continue if there’s reinforcement from other sources or other information that would say that this trend should have legs.

[01:01:19] Alan

Yeah. Well, I guess sometimes you get a trend in the narrative too. As you say, it reinforces over time.

But I’m conscious of time. And you did have a couple of points around trend following that you wanted to touch on. I don’t know if you’ll have time to delve into them in much detail, but curious to get your perspective on trend following being characterized as unpredictable, alpha, and also the whole area of trend beta and replication. What are your thoughts there?

[01:01:49] Mark

The one thing is, as I saw that Aspect came out (and I think there’s a lot of really good research that’s coming out right now from a lot of the CTAs), one, they said that instead of saying that there’s crisis alpha, they want to say that it’s unpredictable alpha. And the argument would be that, well, it’s unpredictable because you could make money as holding CTAs as a defensive allocation and then it could also be as an offensive allocation.

So, I don’t know if I really like the nomenclature of being unpredictable alpha, but I’ve always sort of think of this as more dynamic beta in the sense that you’re making macro bets. What you can do is you can have beta that’s fairly dynamic. It could either be positive, or negative, or on average 0 relative to a long only portfolio.

I think that this shows that people are starting to think about, how do I sort of tell a story and how do I tell narrative and what is trend following actually generating? And I think that the discussion is good.

Similarly, I think I saw some research that came from Quantico, which is doing some really good work in their quarterly piece. And they looked at, interestingly, how does trend following do in high rate environments?

And they show that actually, if you’re from an allocation perspective, you should probably increase your allocation of trend following in a high interest rate environment, which may seem a little counterintuitive. This is because, one, you have more dispersion in returns for a model, slightly lower returns. You also have higher volatility in a high interest rate environment. You also could say like, well, if I have a higher interest rate environment, I could clip coupon bonds of my bonds and make their risk free rate of return.

But what they’re saying is that because the correlation between CTAs and bonds gets lower and the correlation is slightly lower for equities, in a high interest rate environment trend following could be a better diversifier. Again, an interesting piece of research that’s worth thinking about for all the allocators.

And then finally I’ve been looking at this trend beta and this has been some work I’ve been doing with some other firms. And this is some of the group, Corey Hoffstein, and they’ve done some good work to sort of say like, well, can you actually replicate the BTOP50? And does this actually work?

And they’re showing that with certain techniques that you can be able to bundle a set of strategies and come up with a replication strategy that can do well at times relative to the BTOP50, that you don’t have to go out and form a fund of funds, that you could do replication. I think that this area is interesting, but what you find out, overall, from this work is that there’s a lot of dispersion across trend followers.

Not all of them are the same and so, actually replicating is a little bit difficult because of the dispersion or the different approaches that are used in trend following. And so therefore, it’s hard to find or to replicate skill which you see in CTAs. Now, in some sense, if you’re a large allocator, you may say that, well, I just want to get the average because I don’t want to pick the bad manager.

But I think this idea of trend replication or trend beta is something that we’re going to see more research on, and I think that it’s a continued area of interest from actually swap providers and then from large allocators.

[01:05:59] Alan

Absolutely. Well, we’ve obviously heard a lot about this from Andrew on Top Traders Unplugged and Katy, they are involved in the space.

I think it is interesting, and I mean I’m curious to see if we’ll see more analysis of the replicators as well over time and dispersion between the replication approaches. But that’s probably a topic for another day because we’re over on time.

But Mark, great talking to you today. Thanks very much for all of your insights. It’s always a lot of fun and a pleasure to hear your thoughts. So, before we wrap up, Niels is back next week. He’ll be back in in this seat with Cem. So, if you have any questions, send them through. But from all of us here at Top Traders Unplugged, stay tuned and we’ll be back soon with further episodes.

[01:06:51] Ending

Thanks for listening to the Systematic Investor podcast series. If you enjoy this series, go on over to iTunes and leave an honest rating and review. And be sure to listen to all the other episodes from Top Traders Unplugged.

If you have questions about systematic investing, send us an email with the word question in the subject line to info@toptradersunplugged.com and we’ll try to get it on the show.

And remember, all the discussion that we 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. Thanks for spending some of your valuable time with us and we’ll see you. On the next episode of the Systematic Investor.