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. Welcome back to the latest edition of Top Traders Unplugged, where each week we take the pulse of the markets from the perspective of a real space investor. It's Alan Dunne here, sitting in for Niels, who's away on his travels.
I'll be joining him in the US soon, but in the meantime, I'm here speaking to Mark today. Mark, how are you doing? Not too bad. How about yourself? Good. Yeah, well, I'm in Dublin. We're braced for a big storm here tonight and tomorrow, but hopefully we come through that okay. How'severything in the Boston area? Yeah, we're having a little bit of a polar vortex. So, it's been a little bit chilly out here.
So that doesn't mean that the… While the outdoor temperature is low, I think some of the markets have been heating up or at least with the uncertainty that we're seeing. Certainly, interesting times, and obviously you guys had the inauguration this week. So, what’s been on your radar? I don't need a radar. I guess the big issue, for any market analyst, is what is the elephant in the room? Andthe elephant in a room is the regime change that we see.
And I think that might be the best way to put it is the regime change from Biden to Trump. You know, there has been such a flurry of activities, even in the first 24 hours. It's hard for any one person to keep up. You know, what has changed from last week to this week? Yeah, it's certainly a flurry of activity, is absolutely right in terms of all the executive orders and new policy announcements, and yeah, everything that goes with it.
I mean, you brought a lot of topics with you this week, a lot of very interesting perspectives and topics, a lot around this kind of regime shift and the spike in uncertainty. So, maybe that's a good kind of jumping off point. I mean, in terms of that it is a regime shift. Ithinkone of the things is people always say, oh, there's uncertainty and what comes with that is risk. But I think you have more of a nuanced perspective on that at the moment, isn't that right?
Well, I think the problem comes in that everyone, you can listen to any talking head talking about, okay, now we've had a change with Trump versus Biden. Now the question comes in, especially if you're more quantitative, how do you measure that? How do you sort of put some numbers around that? Or can we put this in the context of a sum number so that'll tell us the amount of, we'll say, change that we're seeing?
Andthe one that I use, which is (I think) a pretty good indicator is that there are a number of uncertainty indices that have been available. This has been developed over the last, you know, say dozen years where they have economic policy uncertainty indices for countries. This came out of some professors out of Stanford, and one of them was one of my friends from grad school.
Butwhat they do is that they measure the word uncertainty under certain newspapers, whether it could be monetary uncertainty, general policy uncertainty. If we look at these numbers, especially in January, this is even before Trump came in because they're collected on seven day rolling basis or a one month basis, we have the world uncertainty and indice has spiked to levels that we haven't seen since 2020.
Sothat puts us, at least in context, to say, look, we're facing a lot of uncertainty right now both in the world and in the United States. And that's interesting. I mean, it always feels like the world is uncertain. But I suppose, to say that we're at an elevated level is quite interesting. It's literally just by the number of times the word uncertain is mentioned in newspaper articles. Is that it? There are other key words that they're looking at.
So, there are different measures, and we'll sort of say that you could have arguments on how they measure the uncertainty because what they're doing is they're pulling words from newspapers and different approaches, and then aggregating it up, and then coming up with an index. So, it's sort of LLM, to some degree, on an uncertainty basis.
Butthe important thing is that when they've done some of the research on this, they look at what would be an uncertainty shock that has occurred across countries. What they find out is that when uncertainty has a shock, you're going to get lower investment, you're going to have lower consumption. I think in general, you're going to have slower action by financial participants.
So,when you think about it, if you have an investment project, if you think that uncertainty is going to change or global uncertainty is going to be higher, well, then you're going to sort of delay any kind of investments you're going to make. So,we see that this actually creates trends because everything is slowed down. And what happens is it also creates more expectation dispersion. So,if you have more uncertainty, then the different views that people have will actually get more dispersed.
Now the interesting part about this is that if you look at the uncertainty index, you say, look, this has just flown off the charts. We've got US uncertainty much higher than European uncertainty, which is usually unusual. Usually, European uncertainty is at a higher level. So, it's a separate topic, but it's usually what we find now.
Butwhen we look at the VIX index, which most people look at as the best forward expectation for risk, when, in reality, it's the best forward expectation for volatility, what we find is that that's not at elevated levels. That's still below 20. So, we haven't really seen… We have policy uncertainty, we have global uncertainty, but we don't have VIX uncertainty.
Andif you look at the V VIX, which is the volatility of the VIX, the last six months have been elevated versus the first six months of 2024. But we're still sort of saying we haven't seen a spike. We saw a spike in December when we had, you know, the sort of Fed action. But generally the V VIX has been fairly stable over the last six months in terms of a range. Yeah, it's just interesting.
I mean, one of the things, I guess when you get a lot of uncertainty, is the same markets and a lot of kind of tail outcomes, but how do you price them? And it sounds like the market isn't really pricing them in terms of at least people not paying up for volatility. So,I mean, from your perspective, are you saying that what is wrong, or is it just reflecting the fact that the VIX obviously doesn't capture the full amount of uncertainty out there in the market?
Well, that's a really interesting question and the way you posed it. If I see volatility high and VIX low, that's one question. You could say, well, the market has not priced this uncertainty, so the market must be wrong because this is going to increase. Okay, and that's a valuation statement. Theother thing you have to say is what is it? What are the two actually measuring? And then let's go back to some fundamental ideas of what risk is.
When you think about it, we generally use volatility as our proxy for risk. But risk is actually probably two parts. Onecould be what is measurable, and we'll call it measurable risk would be volatility. So okay, I could measure the dispersion in prices in some markets and that's a proxy for risk. But then there's also the non-measurable component which is uncertainty.
So,when you think about it, if you want to put this in context, we have relatively stable measurable volatility right now, but we have very high non-measurable uncertainty. So, I would sort of say that when you combine the two, we have higher risk even though it's not showing up in the measurable numbers of volatility.
Okay, so I mean, it’s all very interesting but if, as market participants, as other systematic investors or systematic traders, there are allocators to systematic strategies, what could we do with all of that? Or what are the obvious implications or takeaways, or is there something to exploit in this respect? One, first you have to be aware of the fact that there's higher uncertainty, and then higher uncertainty actually leads to differences in behavior.
So, if there are differences in behavior then we're going to see a potential for changes in trends and changes in relationships across markets. So,the way I look at it is that, in some senses, a quant loves everything that's measurable. So, they hate the fact that there is high uncertainty or the fact that this uncertainty index is high because they don't know how to deal with it.
Nowa discretionary trader, he actually embraces uncertainty because if there's more uncertainty, then there's more opportunity for his unique subjective view to actually have an opportunity to be exploited and lead to gains. Butthe important part from any kind of trend follower’s perspective or for portfolio structure asset allocation is that if there is regime change, then there's going to be a change in volatility, changes in correlation, then, potentially, changes in trend.
And so, what I try to do is to constantly try to figure out ways to measure regime change. Becauseif you get the regime change “wrong”, or you don't see it coming, that's probably where you're going to have the biggest potential for loss because you maybe either have too much leverage, you've under diversified, you've miscalculated correlations, and, ultimately, maybe you've missed trends.
So,what you want to try to do is constantly look for regime changes because that might tell you how you should rebalance the portfolio risk. It may not ensure that you're going to make money, but it may stop you from losing money if, let's say, you look and see that there's a change in correlations, or a change in the volatility regime. Nowwhat's a perfect example is, now we've probably beat the 60/40 asset allocation horse to death.
But when you think about it, if we have a regime change and that means that we're going to have higher volatility in both equity and bonds, well then that's going to have an impact on your overall volatility and your 60/40 allocation. If,given this regime change, that'll have an impact on the potential for growth and then the potential for inflation. That'll have an impact on whether the stock/bond correlation is going to be positive or negative.
And so, when we had higher inflation, we saw that there was an increase in the stock/bond correlation, it went positive. Now if we have a situation that we're in a different regime, that correlation may either reverse or it may sort of move in a different direction. I mean, we're talking about a regime shift here and we're saying uncertainty is high. I mean, what do you think the components of that regime shift are? Well, let's take a look at sort of, yeah, the simplest level is that trade.
Because we know that there's been a lot of talk about trade changes with tariffs. So, some have been bantering around the possibility of a 60% tariff and on all goods, you know, tariffs of 25% for Mexico or Canada. Well, what you find out is that that change is the focus of globalization. Okay, nice comment, but how do I trade that? Well,what we find out is that trade is actually based on capital and actual trade flows.
And if those trade flows change because the prices change because of tariffs, well then that's going to have a breakdown or have an impact on correlations. So, generally, we expect that there is going to be a breakdown in correlation across countries and markets. So,because instead of being more in the globalized world where there's going to be more connectedness, there's going to be more disconnectedness.
I think BCG had a nice piece about how trade flows are going to change over the next 10 years. It really shows that well, there's going to be a decline in trade flows with the US but we're going to see changes in trade flows between China and the US and Europe and it's going to probably be more intra-Asia.
There'sgoing to be more trade flows between other parts of Asia and the United States and EU, which was actually a place where there was a lot of exports into China, given the certain environment and because they're looking at their competitive space, they're starting to. Andwe've already seen some trade deals with the EU, with the rest of the world, which are going to change financial flows. If that's the case, then we're going to see a change in the correlation across markets.
Now,to get even more specific, a perfect idea is that if there's an increase in tariffs in homogenized markets like we saw in the case of steel, well, steel prices went up. Okay, so, what happens is that delivery of steel in the US was, whatever the size of the tariff that was the size of the increase.
So,when you think about markets where the US imports raw materials, let's say copper, zinc, lead, to make some examples where there's futures contracts, well that's going to be reflected in the futures contract prices if, let's say, it’s a US contract versus a London contract. So, what we should sort of see is arbitrage pricing, if there's going to be the actual impact of a higher tariff. So, I mean a lot of potential shifts here from kind of a correlation perspective.
For the perspective of, say, trend followers or quant traders, most managers would be kind of sampling and updating their correlation estimates over time anyway. So, I mean, do you think, based on the most recent data, does this lead to any kind of structural problem, do you think? Or what's the implication of that kind of regime shift and possible shift in correlations?
Well, what happens is it depends on… So, when you look at some managers, some managers will adjust their correlations on a calendar basis. So, they'll, say, let’s take the simplest case, every six months what I'm going to do is I'm going to recalibrate my correlation and then I'm going to make adjustments based on that. Others are going to sort of say, I'm going to look at something that's more shorter time series focused.
Butgenerally, what happens is that that you could say, my position sizes could be affected by shorter term volatility, and I might look at some rolling correlation over shorter horizons. But my overall asset allocation is really done on a more periodic basis. Now,if you do it on a periodic basis, you sort of say, well, I may want to shorten up, or I may make that periodic adjustment in my correlation matrix sooner because it's going to start to play out more quickly with a regime change.
You'renot going to see it right away because in the first week we haven't seen the spike in VIX, but we should sort of see, over the next three months, that there's going to start to be impacting actual market correlations across the markets that most people might trade. Maybe take a pause here before we move on in the topics. I want to get to do our usual kind of quick update on trend following. And it does tie into one of the topics I want to get to now, which is bonds.
Fromthe trend following managed features performance perspective, this week you certainly saw a bit of a give-back month-to-date and year-to-date. SocGen CTA up 0.26% SocGen trend now is down 0.69% on the month and year. AndI suppose, the story of the month and year so far in terms of the major markets is obviously a rise in the dollar early in the month and a fall in bond markets which has kind of reversed - so, kind of that choppiness.
Butalso, on the topic that you're talking about with respect to uncertainty, and uncertainty has started become embedded in the bond market to a larger extent in the last number of months with the rising term premia. But obviously, in the last week that reversed a little bit and yields fell. So,I mean taking that lens that you've had around uncertainty and risk, et cetera, and applying that to the bond market and term premia, what would your observations be?
Well, I think that we've gone for a long period of time in the, we'll call it, QE period of no term premium. What's there are bond premium. Basically,if bonds serve as a strong hedge against equities, which is probably the basis of a lot of asset allocation, then you're not going to see some kind of premium for holding risk. Risk has been relatively low. You had quantitative easing, so the government was buying bonds. So, there's sort of protection for a lot of bond buyers.
Nowwe're in a different environment. We had the former Treasury Secretary, Yellen, tell us now, after she's leaving office, that she has some concerns about the size of the deficit. So, the possibility of crowding out. It wasn't a concern while she was Treasury Secretary, but now, all of a sudden, it's arisen as a possibility.
So,what we see is that if there's more volatility in bonds and more uncertainty overall, then you should see that there should be a term premium or a risk premium associated with holding longer duration assets. So,now what happens, when you look at bond valuations, you say, well, we want to look at the neutral rate of interest but you're really looking at what's the growth rate plus expected inflation. And that's your quick proxy for what is the fair value for bonds.
Now you have to add in the risk premium. Andso, if you just sort of look at growth plus expected inflation, you're probably going to miss it because you're saying, well, I think I should see a big bond rally and it may not occur because, at the same time, we have an increase in term premium. Interesting. I mean it's just occurring to me listening to you, I mean, we're talking about uncertainty, and I wonder.
I'm not sure if the indices you look at differentiate between different types of uncertainty. I mean you could have growth uncertainty, or inflation uncertainty, or some other types of uncertainty, maybe policy uncertainty. Imeanit would seem like if you had more growth uncertainty there might be a bias towards owning bonds, whereas kind of inflation uncertainty, maybe less so. Would you agree, or any thoughts on the nature of uncertainty and how that impacts? Oh, absolutely.
I think that that's a really good point that you make. Now, we're not having a call on just uncertainty indices, but they've broken it up into just uncertainty, then they look at policy uncertainty. There'salso an uncertainty index on trade uncertainty which also has gone through the roof. And then there's like monetary policy uncertainty. And so, if we say that the risk premium in bonds is associated with uncertainty, where is that uncertainty coming from?
Well,one, is to say that there's overall growth because I think that, if we go back a year, most people thought that we would be in an inflation - we would have lower inflation, but they weren't sure on what is the level. On the growth side, most people were expecting that there was going to be a tepid recession, but they were expecting that there might have been some recession a year ago. Now, you know, people are thinking that there's going to be a new horizon.
Interesting,as just a side topic, you know, there's been some interesting work where they look at what is the economic environment when the president takes office? Andwhile there's been a lot of euphoria about the new administration coming in and that this is going to be a great environment for equities and bonds in the economy, it's actually not a bad environment right now.
We look at, in the range of what statistics look like for a new presidential administration, we're at the low end of unemployment. We're probably in the mid to slightly lower end of real GDP. So, there's more room for growth. Economic utilization is at the high levels. Core inflation, you know, we'll sort of say it's a little bit elevated, but it's not that far off. And debt to GDP, we're sort of in sort of a bad, bad situation. So, you'd say, well, how much more improvement can we get?
And there might be a limit to how much improvement we could get. Butgetting back to the term premium issues in volatility, given that we might have uncertainty about what the growth is because we just talked about that, if you have high uncertainty, it has an impact on both consumption and investment.
If you also have high monetary policy uncertainty, and I think that there's a view that maybe September was overly aggressive, I think that people aren't really sure what the Fed might do over the next year. Yeah, interesting. I mean, one thing, on the uncertainty topic, in one sense, it's not surprising. You know, we have a new administration, but at the same time, prior to the election, the argument was that uncertainty is high because we don't know which way we're going to go in the US.
Is it going to be the Republicans or Democrats? Now we’ve got that certainty. Buteven with that certainty and even with Republicans in the president, the highest in the Senate, we still have this kind of sense of uncertainty, which is maybe somewhat surprising, but maybe just reflects the personalities at play. Is that it? Donald Trump wrote a book about deal making back in the 80s.
And part of the idea was that, for his idea of making a deal, you start out with some outrageous proposition or sort of outrageous point of view, and then what happens, if you negotiate away from that extreme, you probably will get close to what you wanted anyway. So, you start by exaggerating and then you come back to some place in reality. Andso, we had the sort of campaign exaggeration, which is normal for any presidential candidates.
Then there is the post-election sort of saying, well, what is he actually going to do? Andso, most of these uncertainty indices started to pop, obviously, before he got into office. Now we're finding out, so, this is what he campaigned, this is what he said post-election but before he was inaugurated. Now we sort of see what is he actually doing. So, each one of those is a different part of uncertainty. It resolves some uncertainty but then creates new uncertainty.
So,there's the resolving, when he said this is what he's going to do, and to some degree is that's resolved because now he's actually doing it. Now the question comes in, well, how much is he actually going to do since he started to do what he said he was going to do? Sonow, this is not unique to Trump. It can be very apolitical. It’s the fact that there always is a disconnect between what is promised versus what is actually delivered.
And part of what you do is you sort of handicap candidates on what you think that they'll actually deliver versus what their rhetoric will be. I mean, one of the things on my radar in the last couple of weeks has been a couple of articles I've written. I think there are, I'm not sure, a couple of people have published studies on the US labor market and immigration flows.
And what they've highlighted is that the immigration flows into the US were much higher in the last couple of years than in the preceding period. I think it was like maybe 2 million per year in the last 1 to 2 years versus maybe a couple hundred thousand was the kind of running rate prior to that. Andobviously you can debate how feasible the deportations are, but certainly if you shut the border and stop that flow of migrants in, that would presumably have a fairly immediate impact.
And I think Biden had already taken steps on that front already. And the argument is that we should start to see this in the labor market data soon. So,it is interesting. I do think this is one of the topics that economists were saying that essentially would be stagflationary in the sense that, if you don't have as many workers, you can't hire, so you won't have that kind of run rate of 150,000 payrolls every month, and the economy can't grow as high.
But, but at the same time you would expect to see kind of upward pressure on wages from a lack of labor supply. But,but that's certainly one area, I guess. But I mean hard to just zero in on that. You also have the tariffs and what we're going to see with tax, et cetera. ButI mean, overall, if you were to take it all together, is that the type of scenario you see?
Do you have a view, are we on the cusp of what many people are betting in terms of a new age or a golden age, as has been suggested this week? Well, without getting into the politics of immigration, it's clear, and this applies to both the US, Europe, and China. So, there are some universal truths that we know. Whatis long term growth going to be determined on, for any economy? It's going to be the growth in population and your productivity.
So, productivity, in the short run, may not be everything we know. In the long run, it's almost everything in terms of growth. So,let's look at Europe or the United States. If you have a declining demographics workforce, then that's going to put a drag on overall growth. Right? Because, if you have the same per capita GDP, the only way that then you're going to be able to grow, if let's say the population is declining, is that you have to have an increase in productivity.
Andwhat causes increases in productivity? Well, there's marginal increase in productivity. There are also other innovations, but oftentimes innovations take a while for them to play out. AI might be a perfect example. So,in some sense, if you want to grow the US economy, then what happens is that you're either going have to Increase productivity or you're going to have to increase the number of workers. And this is sort of a long-term view.
Now,the important part here, and what is missing sometimes, is that there is a school of thought which is called the new institutional economics. And basically what they say is that economic growth, you know, the economy is determined by the set of institutions and the interactions between agents within the economy. So, institutions matter. Ifyou think about most of your macroeconomic classes, most of your classes in finance, we skip over the idea that institutions matter.
So, we'll talk about this in finance and trading in just a second. But what is the composition of your labor force as an institution (it will matter)? Is it young versus older? Are they educated versus less educated? So, in some sense just saying that I need more workers is not an answer. It has to be what are the type of workers you're getting and then what is the flow through to the economy?
Now,when you think about this institutional economics, or new institutional economics, when you think of markets, well, the institutions in which we navigate in to trade will matter a lot. And I think that, while this is different than the immigration issue, there is a recent paper on long-term impact of trend following on equities and long only. So, it’s an exhaustive study. Thenwhat they did is they looked at what are the transaction costs.
So, once they add in transactions costs, what was positive alpha strategies, depending on the size of the portfolio, almost went negative. It just eats up all of your returns. So,in some sense, institutional structure is, what's your cost of trading, what's the liquidity in markets, these will make a huge difference and what is the performance that you have?
So that even though you might say, as a finance person, I'm only looking at prices to build models, it's incumbent on you to also think about what is the institutional structure that you have to navigate to generate those prices. So, in some sense, without transaction costs are the perfect world that you live in, but you don't live in the perfect world. Itwould be like, we don't do physics without friction. We don't do physics without gravity.
So, we have to deal with gravity, friction and other issues. And so that's why it's important to do this also from a trading perspective. Well, that's probably maybe a good segue into another topic which you wanted to touch on. And this is really, I suppose, bringing together what we've been talking about.
We'vetalked about kind of a potential regime shift characterized by higher uncertainty, possible breakdowns in correlations, also shifting term premia and whether that's going to be a feature of the landscape. And you're also kind of alluding to the fact that the institutional arrangements underpinning trading conditions could shift as well at some point.
So,from the perspective of a systematic trader, investor, building models, reevaluating their models for this environment, for somebody running trend following, it's built to be robust. Are changes required to other types of models given what you're saying about a regime shift and a changed market environment? We could do a better job as risk managers and as model builders in looking at different scenarios or simulate how models will do under different conditions.
So, we always want to do our backtests. And then, generally, once you get your business up and running, people say, well, let's throw out those backtests. We don't want to see them anymore. Butthe perfect example is that since we do have a lot of information on spikes in uncertainty, we can go back and take any model and be able to sort of say, well, what was the performance like when there was a spike in uncertainty and how did it behave?
And then, can that be able to tell us, as a risk manager, what might be the possibility of performance during other periods? Andcan it tell us something about how our models are constructed and why they may work or not work? You know, a perfect example would be, one of the big spikes that we had for uncertainty was surrounding the pandemic.
Whatyou found out is that, depending on your time frame for your model, you either actually sort of came out okay, so, you actually made money over that period of time, we'll say from February to May of 2020, or you got your clock cleaned, and a lot of it had to do with your time horizon.
Thatdoesn't mean that you should change your model, but then you should be aware that if I have certain types of spikes in uncertainty and my model is either short term or long term, that's going to have an impact on what my expected returns could be. Now, that doesn't mean that I'm going to say, oh, I've got to switch my model.
You'renot arguing that, albeit it is important to understand those differences, but when you think about it, a lot of these businesses are, one, do you know what you can expect from your existing model? Andtwo, if you have a firm, you have to be client facing and say, what can you tell your client about what you might see over the next three to six months in performance?
And if you're an allocator, what you really want to do, and I think I would appreciate, when being on the allocation side of times, if a manager tells me, look, we see this unusual environment, we see high levels in uncertainty. We went back over the periods of high uncertainty and here's how our model performed.
We'renot saying that we're going to have a repeat of history, but by doing this, we could say, here's a better way we could at least bracket what our performance would be so then you can understand how we're going to do under these different scenarios. I guess, yeah, it's interesting. One thing you can do obviously, is, as you say, stress test and look at models in different periods. And that's all right.
Obviously,what we're saying as well, here, and in other podcasts is kind of trying to envisage different states of the world which we haven't seen as well. So, I mean, is that just via simulation and thinking about tails, or is it a matter of going back further into history books to maybe periods where we don't have a lot of data? Like, I mean, what are the analogs to the current period?
Youknow, I suppose the 1930s was the period when, you know, protectionism and trade wars really ramped up to a level that they became very disruptive and obviously contributed to the Depression. Is there a merit in kind of trying to construct hypothetical data from those kind of periods too far back in past, or is that not robust enough in your perspective, would you say? Well, it's hard to get robust information on there and that's where the institutional differences have an impact.
But let's look at the idea that anything is possible. And that might seem trite, but you have to at least be aware of the possibility that anything is possible. A perfect example, let me put this way, is that while there was a lot of talk about breaking Bretton Woods in 1970, ‘71, yet somehow no one said that was possible. Yet in some sense, okay, we went off the gold standard. Who would have thought that that would have happened? And then that changed the regime, you know, significantly.
And you look at behavior of currencies pre or post and then the volatility we saw was amazing. Welook at who would have thought that we would have taken interest rates up to more than double digits in the short run, at the end of the Carter administration? So, who would have thought that, well, you had the big Plaza Accord in FX, you know, and we actually had the EMS failure and then and the breaking of the pound in the early ‘90s.
Well, if you look at some of the environment in the UK you could sort of say, well, we have a sort of interesting situation there that could be extreme. Ifyou went back in 1990 and said, well, we're going to have a single currency in Europe, I think that, you know, some academics would have said so, but you never would have expected so. We know that there's been a tremendous amount of change which has caused a tremendous amount of change in market behavior.
Atthe same time, one could also say that there's been a tremendous amount of resilience. So, we've gone through these events that we thought that it was impossible to handicap. We could not conceive. We sort of said that the world is going to come to an end if they occur. And somehow we sort of came out at the other end and we survived.
Now,specifically, when you look at trend following, when we think about regime changes, it may be talking my long-term book, but I think that the trend following is actually a good strategy to increase exposure under regime change because one is that it's not really making a lot of assumptions or forecasts other than prices will move in a certain direction and we'll try to exploit it.
Whenyou think about a carry strategy or long short equity, a lot of it is based under the assumption that there's going to be some mean reverting behavior where the linkages between markets are going to have a certain, you know, behavior. Trend followers don't make that assumption. Now, we talk about that being divergent strategies. Ifwe go back to Nassim Talab, you know, we could say that there are fragile and antifragile strategies. We could think of trend following as more antifragile.
Because if, let's say, there are regime changes, that means market behavior changes quickly. Trend followers will say, I'm not going to try to assess what this means. I'm just going to say that the prices are telling me some markets are going up and some are going down. Therefore, I should be a buyer that wants going up and sellers going down. And I'm not going to make any assumptions of interpreting what that means.
So,in a high uncertainty environment, we should expect that trend followers do better. And, you know, the concept I talked about a long time ago is something called betaflex. So, when you think about it, when you have high uncertainty your exposure to beta, market beta, is going to vary quickly. So, what you want to have is a strategy that has a flexible overall beta relative to the stock market or relative to some other benchmark.
So,in a perfect world, if the stock market is going to go up for 2025, even though it may be overvalued, we want to say, well, my flex beta would say I need to have more beta exposure now. If the stock market goes down or the bond market has a certain behavior, I want to be able to be flexible enough to take that beta down and get it to, in the best-case scenario, negative.
So,we talk about that in terms of convexity, unfortunately, convexity is a hard concept for a lot of people to think about. You should say like, well, I've got a convex portfolio. Okay, is that where I've sculpted in, or is it sculpted out? Butif you say like, well, I want to have flexible beta or betaflex, I want to say that if the market is going up, I want to have more beta exposure or I want to have positive beta exposure.
If the market's going down, I want to be able to have a portfolio that's adaptable and adjusts quickly so that the beta will get negative. Okay, that seems to make a little bit more sense and that should be antifragile. Okay. I mean, speaking more broadly, we talked about how to, as a modeler, as a model builder, how to react to a regime shift in markets and a changed macro environment.
I mean, from an asset allocation perspective, the implication seems to be suggesting more to divergent, less to convergent strategies. Is that fair to say? Just for some context, I was at a presentation this week from one of the banks.
Theywere doing their hedge fund review and outlook and they were talking about last year's performance, the last five years performance, and also looking ahead, they had surveyed investors to see what was the appetite, which types of strategies that were being favored looking ahead. Interestingly, the types of strategies that people were increasing allocations for were the multi-manager, multi-strats, stat arb and quant multi-strat.
Whereasthe ones who were reducing allocations were credit, long/short, and lots of different types of credit strategies there seem to be concerns about. So, I mean, I guess that doesn't sound like a huge ramp up in allocations to divergent strategies. I'mnot sure, stat arb would obviously be more of a convergence strategy. Quant, multi-strat, probably the same. But how would you interpret that?
Well, this is fundamentally when we talk about convergence and divergence, we're talking about what are the underlying assumptions of market behavior at a given point in time. And are we going to have mean reversion, and that there's going to be some equilibrium level that we're going to move back to. If we have a regime change, what we're saying is that we don't really know what the equilibrium level is. So, consequently it's hard to move back to something that we can't measure.
So, let's try to be more specific. Ifyou're a bond guy and you have more of a mean reverting strategy, you say like, well, let's look at the, not R Star, but let's look at that simple valuation, a model of long term growth with expected inflation. We look at the nominal rates right now and then we'll say, well, it's going to converge back there. So,that would be convergent assumptions, mean reverting assumptions.
What we're saying is that, well, what happens if there's a term premium that's added in there that we didn't think about before? Okay, well then that convergence is not going to occur. Whathappens if, let's say that, well, we're expecting that inflation is going to stay at a low and then we'll sort of say that they're still going to be able to hit the 2% target. In reality, what we're finding out is that that last mile problem is a lot harder than what we think.
So, it's quite possible that maybe, if we have an increase in tariffs, you might have a situation that inflation could maybe be stickier at current levels or it might actually go higher all of a sudden. Now the valuation is a lot harder to make. So,what we would argue is, let's go back to what is the fundamental strategies and then think about whether there are assumptions, or another way to put this is, what are the causal inferences that the manager is making on how they would make money?
Andif you look at multi-strat, which you know, may be taking over the world, what they're doing is, for the most part, you're building a huge portfolio of managers all of which are probably sort of, in some sense, mean reverting whatsoever. Because, in some sense, what they're looking for is the highest Sharpe ratio trades. So,you want to have high return trades relative to risk and then lever up. And I sort of increased the leverage to be able to make it all work.
So,by definition, you are not going to have a lot of divergent strategies in there because, by definition, the divergent strategy is currently going to have a higher volatility. You can't have a divergent strategy and have a low volatility approach.
Similarly,if you have a divergent strategy, you're having a high return to risk strategy which is also fairly hard to get because you might have periods where you're going to have poor performance waiting under the fact that there might be then periods of high uncertainty, high risk, or dislocations that then are going to give you a lot of concentrated return. By definition that's not going to fit into a multi-strat strategy, regardless of whether it's quant or discretionary.
So, what happens is, the more I put into multi-strat, the more I'm actually making a convergent mean reversion play. Yeah. Yeah. Okay. But more broadly, I mean, do you think there is an obvious asset allocation or strategy? Obviously, you're advocating for more divergent trend following exposure for the current environment. Anything else obviously, or does this create any obvious risks that you think people should worry about?
Well, the one issue that has really been sort of at the forefront that impinges on all quant strategies is that, generally, well, let's take out multi-strat. You look at the last couple of years, the big sucking sound in hedge fund universe has been all the grab for private equity investing. So,in some sense, it is hard for a quant strategy.
It's hard for any liquid strategy where you have monthly volatility, that might actually sort of look pretty high, compete against the private equity market. Yet if we start looking at private equity markets in general, and applies to credit too, is that we're seeing a situation where their performances underperforms versus public equities. So,I've seen some recent numbers where they look at rolling 12 and 6 month periods of time.
The idea that I'm getting paid a liquidity premium for private equity, in fact I’ve got to get excess return versus public markets, has not been the case. We know that volatility has been dampened. But that's because they don't mark to market everything on a daily basis. Andfinally, you have the idea, well, what's the exit strategy for a lot of these private equity strategies?
And what we're finding out is that if we don't have a robust IPO market, then it's hard for them to get out of the strategies. So, then it's hard for you to get your money back. So, now that you have conversion vehicles that will actually take off deals that have been around for a long time, then all you're doing is kicking the can down the road.
Oneof the themes that I would say is not so much a revenge of the liquid strategies, but we'll sort of say, I think it requires a rethinking of liquid versus illiquid strategies, and that there might be even large managers who should say, I'll take my mark to market process, I'll take those marks, I'll take the higher volatility. But at the same time what I'm gaining is liquidity and I may not be diminishing my overall returns. Fair enough.
One of the other topics that came up at this event that I was at, this hedge fund event where people were talking about the multi-manager multi-strat pods, and possible risks and concerns that you sometimes hear from those shops is around leverage and the ability to access leverage. And obviously, the whole model of that kind of multi-manager approach is to bind, as you say, a lot of managers with uncredited positive Sharpes and then lever up.
And obviously they're not always just trading futures. They do rely on bank balance sheets and funding, particularly in trades like the basis trade, et cetera, to do that. Imeanwe've been in a relatively benign banking world. Okay, we had SVB as a kind of a wobble that was very US, very kind of the small, medium sized, bank centric.
Could you see any scenarios where bank liquidity or leverage provision could come under stress, and the obvious trigger for why that could be the case in the coming one to two years? Absolutely, when you think about it, Wall Street and banks are very good at… Once they find something that can be very profitable and easy to administer, they'll throw as much money as they can until they can't. Andthere's nothing wrong with that as a business model of course.
If you have to do your due diligence on, let's say, a leverage loan or if you have to do retail lending, or business lending, it's a lot of work. You know, you have to gather all the data. You have to have your credit analyst. Yougo to a multi-strat, say like okay, I get your prime broker with me, I get all your positions, and then I sort of come up with some risk model to be able to say what kind of risk you can sustain.
And then we'll say we'll give you as much leverage as you can take, given the fact that we know all of your positions. It's a great model. Butwhat we found out in, let's say, August of last year when we had sort of like the carry debacle, where we look at the long/short problem that occurred in June of the prior year. We've also seen that, if you look at the Mag 7, this is in periods when they have been the biggest losers around.
And so, if you have a big exposure to that, that could be an existential risk. So, there are a tremendous number of existential risks that could occur. Andwe have to put that in the context of, earlier in our conversation, we said that risk is two components. It’s the measurable volatility, which Wall Street prime broker analysts are very good at sort of measuring what is countable. That would be the volatility of positions.
Unfortunately,they also have to make assumptions about what are the correlations and what are the linkages across this, in what we call a complex system. But what they're not very good at, and no one is, and so I'm certainly don't say that I'm good at this, is what we'll call the non-measurable risk, which is the uncertainty.
Andso, if we go back to square one, where we started the conversation, when uncertainty is high, which is non measurable, this is something that could be able to sort of come back and haunt you in a way that you didn't think because linkages could change, volatility could change, the dislocations could change, divergences could change. And so, in some sense, if I let the lever up on that, I'd be a little bit concerned.
Andso, if anything, while you want to continue to press money into, you know, multi-strat hands, you say, would this be the best time to increase leverage lending to these institutions? You might be a little bit… You want to give this a little bit more thought .
Just being conscious of time, and we're getting up close to the hour, but one thing we hadn't mentioned, and I know it is one of the topics you had brought, and it definitely relates to all of this, and it is the idea of how the markets and the economy are complex systems. You often hear the term ‘complex adaptive system’. Soobviously adaptive is relevant in the sense of how market participants react to changed regimes, as we've been talking about.
But complex as well also touches on the non-linearities. So, I mean we've had these kind of mini blow-ups, flar-ups in the last few years. You know, we had the yen carry unwind kind of July/August last year, and 2023 - I think it was SVB bank, and then obviously 2022 - no huge vol breakup, but the steady decline in equities, and back in 2020 Covid. Froma complex system kind of framework, obviously Covid was kind of an example of a transmission across a system.
But the other kind of episodes were somewhat contained and didn't spill over into more broader crises. Isit just random when you get these kind of localized flare-ups versus, I suppose, vol flare-ups that then lead to more entrenched or bigger crises for financial markets? Does the complex literature have anything to say on that? Well, I think that I'm a big believer that the world is a complex system and we have to look at the network connections across markets.
And I think that if you go back to the Santa Fe Institute, which has done a lot of work on the complex systems, they've done a tremendous amount of work. On some level they've never been able to be embraced by a lot of traders or modelers in a very systematic fashion. So, we'll sort of say that there's a disconnect between the theory and what is reality. So,this is at the same time I'm saying I'm a big believer in this. So, I don't know what that says about me.
But what we have to think about is, what are the characteristics of a complex system? And then we have to look at the models we build and say, how are those two related and am I missing something by making simplifying assumptions? Okay,so what does that mean? Complex systems usually have nonlinear responses to different shocks. Generally, you build models in a linear world. So, you say, how do you address or how does your model incorporate non linearities?
Youknow, if I was doing a due diligence on a quant model, that is one of questions I would ask. Now there are a lot of other questions, but these are more nuanced questions. So,we know that shock effects have unintended consequences in a complex system, because again, there's not linear connections. So, you'd say, tell me how you think shock propagations will occur. So, tell me about how you think you'll deal with contagion.
Okay,we've talked about dispersion of correlation, but the question that comes up is, what happens if correlations do go to one? What do you do in your model, or how does that change your behavior? Thirdis, how do you incorporate changes in institutions which will impact transactions, costs, or linkages? How do you incorporate that in your model asset allocation, in your model building? So,for example, all we always sort of say is that when you build a model I want as much history as I can.
But if you just look at all the history, in some sense, you're giving the same weight from 15 years ago, if you have 20 years worth of data, as you do today. So, the question is that the world 15 years ago is not the same as it is today. And we'll sort of say that even the world of 10 years ago is different than what it is today. So how do you incorporate or how do you weight the past versus the present in your model?
Andthen I think you also want to sort of say, what are the changes in transactions, cost and incentives in the world in which you live? And one phrase from one of the transaction school economists, Alchian, said, “Tell me the rules and I will tell you the outcomes.” Nowthat was on a macro basis, but in some sense, tell me the rules of the system and I'll give you an idea of exactly what performance would look like.
Looking at a complex system, it could be similar to how do I reweight some market exposures? Aperfectexample is something that I'm looking at. We do know, in a regime change, if there's less pages in the Federal Register, which is the amount of regulations that occur, that actually small cap stocks will do better. And this goes back to, it's an obscure book from Laffer and Canto, on asset allocation. So, Art Laffer, we know, is from the old Laffer curve.
He did a lot of asset allocation work a couple decades ago. They actually looked and said, yes, we can sort of quantify the fact that what the small cap effect is related to regulation. So, there's a change in the rules that might have an impact on overall performance. Theidea would be, okay, cost of regulation if, let's say, you're a smaller company, that's going to have a higher fixed cost of impact. If you can reduce regulation, that's going to reduce their fixed costs. That's a good thing.
So,in some sense the complex world is sort of saying, how does the market behave? And then how is that behavior manifested in your market? And then how do you think in terms of shock effects and non-linearities? I mean the other thing you have to consider, I'm just thinking, is policy responses as well, which I guess also add into the mix and are part of the reaction function. But maybe that's a topic for another day. I'm just conscious that we're probably over on time.
Unless you want to comment on that briefly. I think that a perfect example is, let’s take the classic one, we did a lot of quantitative easing because we said that it’s going to have a big impact on interest rates, that we're going to be able to bend the curve and in effect to try to affect financial decisions. Then what we said is, well, we're going to reverse that, but we're saying that there aren't going to be any implications when we do some reversal.
So, it would seem as though that… Andthen, all of a sudden, depending on what the amount of reverse repo is that you do at the Fed, all of a sudden that has a big impact on money markets, which has a big impact on the flow of cash around the world. Those are the kind of things that, okay, you don't incorporate in a model, but you have got to at least think through those things.
Andthe great part of doing quantitative modeling is that it should allow you to have time to think about these big, deeper issues that could potentially make you money. Good stuff. I think we'll probably come back to it again at a later date, but thanks very much for your insights and thoughts today, Mark. We're over on time, I believe. Nextweek Niels is back, so stay tuned for that. And from all of us here at Top Traders Unplugged, thanks for tuning in and we'll be back soon.
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.
Andremember, 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.