Jonathan Forsgren:

Welcome to this Asset TV factor investing masterclass. Market events and investing trends over the past few years are driving investors to rethink their investment allocations. Factors are helping investors understand the current market and helping them inform their investment decisions. Joining me to discuss factor investing are Janis Zvingelis, Principal Director and Head of Quantitative Research at Envestnet, Mark Carver, Managing Director and Global Head of Equity Factor Products and Equity Portfolio Management at MSCI, and Melissa Brown, Managing Director of Applied Research at Qontigo. Thank you all for joining us. Janis, could you kick us off by telling us what is a factor?

Janis Zvingelis:

Absolutely. Factors are variables that indicate that certain bad times have arrived in the economy. And the reason that we focus on bad times is because investors generally are very fearful and wary to experience bad returns in their portfolios during these bad times. It's kind of a double whammy, economy is doing badly and their portfolio is doing badly. The variables that we use to indicate bad times have arrived are things like economic growth, real rates, inflation, credit spread and so forth.

However, the factors that we usually talk about are value and momentum and quality and so forth and market. So the question is, how are these style factors related to these macro factors? And the way that they're related is that these two sets of factors, they correlate with each other. In fact, the academically correct way of calling what we call style factors would be factor mimicking portfolios because they mimic the movements of the underlying risk factors.

A good example here is the market factor. So market factor, A, it is tracking the economic activity really well. So when economy is doing badly, when the growth is bad, the market factor tends to underperform, actually it's a leading indicator. On the other hand, market factor is a zero cost portfolio, meaning that you can enter into this portfolio risklessly if we abstract from trading costs and so forth.

Finally, market factor tends to have persistent and significant risk premium. And so this is where things get a little interesting because a question might be, well, it's easy to understand these macro factors that identify bad times, but how did the finance profession come about finding these other factors, value factor, momentum factor, quality factor betting against the data factor and so forth. And here the finance profession kind of went about it by putting the cart before the horse.

The way that researchers identify these file factors is that they construct long short portfolios, so zero cost portfolios and then they observe whether these portfolios have systematic and persistent risk premia. And if they do, then we postulate that there must be some underlying risk factor that this portfolio is tracking or mimicking. There's certain problems with this approach, data mining and everything, but that's a question for a different conversation.

I wanted to leave with this thought though, that factors the way we talk about factors, style factors, value factors, momentum factor so forth are not sources of alpha. The reason that risk premium exists on factors is because these factors tend to underperform exactly at a kind of the worst moment. So when times are painful already because of economic activity, the value factor for example tends to underperform and so forth. There's other examples.

So there must be a group of people that say, "No thank you, this is a little too much excitement for me. I don't want to have bad performing portfolio during bad economic times." And that opens opportunity up for the risk loving investors that say, "Yes, I'd like to invest in this portfolio given that there is some extra reward for it," and that extra reward will is what we call risk premium. In that sense, the investors in factor portfolios act like insurance companies whereby they take on the risk that the other side is unwilling to bear and they realize the risk premium for it.

Jonathan Forsgren:

Are factors only used when the economy's doing poorly or could there be some factors that look at when things are bullish?

Janis Zvingelis:

No, absolutely. The factors can be used at any given point in time. It's just the reason for there being a risk premium on the factor is because they experience bad prolonged returns when everything around is already performing poorly.

Jonathan Forsgren:

Are there different types of factors?

Janis Zvingelis:

Yes, there are. I already touched upon macro factors or systematic undiversifiable sources of risk. So things like economic growth, real rates, inflation, credit risk, illiquidity risk is actually an important one. You can think about these macro factors as affecting securities across asset classes, whether it's equity or whether it's fixed income, whether it's options, whether it's commodities, what have you. They're all affected by these underlying economic macro factors.

And then there's style factors, and these are the factors that we're kind of used to talking about. So market size, value, momentum and so forth. And these are factors that affect securities within an asset asset class. But I'd like to also point out that mostly the focus has been in the industry and also in academia and in equity, equity style factors, but there's also style factors in fixed income like term and spread, value, momentum carry is a very important one and other asset classes for example, options and volatility risk, and we at QRG, we construct factor strategies in all these asset classes, equities, fixed income and options.

Jonathan Forsgren:

Thank you. Mark, how are factors doing in the current market environment?

Mark Carver:

Well, I mean certainly Janis painted a very interesting picture with that description. So when we think about factors, I think most of the people in our audience are thinking about the style dimensions that he mentioned rather than the macro dimensions. From that standpoint, it's important to say how are equities doing more generally, and so we know that equity markets feel like they've been very volatile this year, but equities have been particularly resilient. If we look at the US equity market through mid-May, we're up roughly 10%.

When we look at factors, there's two ways to think about it. When we think about pure factors, we can look at our model factors. The best performing factors have been things like beta, book to price, which is a value dimension. On the other hand, the worst performing factors have been residual volatility. And so what's sort of interesting about that is there's this divergence that we're seeing take place.

Systematic risk, i.e. beta is doing well, non-systematic or stock specific risk, residual volatility is doing relatively poorly. We can unpack some reasons for that during the session. When we think about indexes and we can use our factor indexes for instance as a proxy. Many of the listeners are familiar with those, for sure what we've seen this year is this flight to quality, but more generally, effectively what we're seeing in 2023 is that last year's winners are this year's losers. So last year was an incredible year for value strategies outperforming growth by historic levels. This year growth is performing well.

Quality has been the standout factor index, so we're seeing this rotation among investors from things that they were potentially exposed to last year more towards this flight to quality. There's even evidence of that in terms of flows. The best-selling ETF this year, for instance in the US market has been a quality ETF, so the performance has been strong, but then there's also been this flight to quality.

I mentioned this notion of divergence. Some of the reason for that is the sentiment or the appetite among investors seems to be very focused on security level fundamentals and particularly the sustainability of earnings, and stocks that have strong steady earnings that are profitable are doing well, they're being rewarded. Stocks that miss earnings or maybe have an earnings revision, an estimate that's to the negative side are getting severely punished. Sentiment overall we're seeing skewing to the negative analyst revisions are skewing to the negative, so overall markets are doing okay. Quality's been the standout factor from an index standpoint. Beta or systematic risk from a pure factor has been the standout.

Jonathan Forsgren:

Have you seen an evolution of factors to reflect the changing landscape?

Mark Carver:

Well, I think there's an evolution in the way investors think about factors. Janis mentioned a lot about academic factors. The truth of the matter is if you talk to conventional investors, probably most of the people listening, they don't think in risk premia terms only. They've moved well beyond that. If you don't believe that's true, just think of the reality that there's $3 trillion of assets in growth investments in the US market. Academics would say growth is not a rewarded factor, it's not a risk premia. Obviously mainstream investors vote otherwise. So I think there's always been this sort of friction between this definitional notion of risk premias and the practical application of tilting your portfolio to characteristics that you want exposure to.

As we talked to clients, we did a survey very recently and what we heard was a lot of appetite for this notion of style dimensions to evolve. We heard appetite for new signals to capture familiar factors. That could be things like adjusting book to price for intangibles. So moving beyond the ways things were done in the '70s, '80s and '90s and even the 2000s and then adjusting your valuations on companies.

We also heard appetite for more what I might call advanced techniques where you can use things like machine learning to identify new signals that might be rewarded, potentially machine learning in and of itself could be something that you can capture systematically. I know that for some that sounds like the dark arts, it's not. We heard appetite for various forms of sentiment. We heard appetite for things like crowding, so that's moving well beyond the traditional value, momentum, volatility, considerations to much more unique and isolated. Even innovation came up a lot in our client survey.

The second thing that I think is evolving is the way people actually implement and use factors where in that same survey and in the practical reality, if you look at flows to public funds, public ETFs, you're seeing this rotation among markets where investors might be overallocated to certain factors in one period and overallocated to another group of factors. This is not factor timing based on the conversations we're having with clients. What it reflects is that more and more investors consider factors as a key part of their asset allocation strategies and they're up and down weighting certain characteristics based on some of those macro variables that Janis talked about.

This is fairly unique in that some factor purist thought, you buy a factor, it's going to be rewarded in the long term and hold it. The truth is people's horizons have shrunk. The cyclicality of factors is recognized, and the result of that is more rotation really rebalancing towards certain exposures and away from certain other exposures.

Just to put a fine point on this, in that same client survey, when we asked clients which factors they liked in today's market and we let them define today's market rather than sort of biasing the jury, the things that came up, nearly half of investors said they thought quality was an appropriate factor for this macro regime. Almost 35%, 36% said value. Interestingly, more than a quarter said they would reduce their allocation to growth. So again, this more evidence that there's evolution in the way we define these concepts and importantly the way we implement inside of portfolios.

Jonathan Forsgren:

Melissa, what is the best way to evaluate the factor strategy?

Melissa Brown:

Well, I mean this really goes to what both Janis and Mark have just said, that there is this academic definition of a factor that Janis called the factor mimicking portfolio. We would call it the same thing, but it's an unrealistic way to invest. Investors can't be fully long short, they're not going to have thousands of positions and so I think if you're evaluating a factor strategy, you really need to look at it in terms of how are you going to use this factor strategy. So many investors don't want to short stocks, so you're going to have a long only portfolio.

If the factor that you're looking at really only works well on the short side, it's not really a great factor for your particular needs. The same thing is, Jan mentioned thousands of long and short positions, many investors you don't want to have that many stocks in your portfolio. You want to limit the number of names that you're investing in. And so you want to make sure that ... let's say you just want to limit yourself to a large cap universe, want to make sure that these factors work in a large cap real life context.

And so I think this process of evaluating factors, it may start with this factor mimicking portfolio, which is basically saying in the best of all possible worlds, what would I expect this factor to do? But you really want to bring it down to a level that's going to more closely reflect what it is, what your investment strategy is going to be. Then I think another aspect of this is I think many of these academic studies will look at a factor and say, "Oh, over the past 30 years," or 20 years, whatever, "This factor produced a sharp ratio," or, "Return per unit of risk of whatever, this is a great factor," but you really I think need to dig down a little deeper and say, "Well, did that just happen? You had a few great years and everything else was terrible?"

We know that most asset owners have basically a three-year investment cycle and so they'll look at a three-year cycle. You don't want to be stuck that I've gotten in at the top and then I'm going to get fired at the bottom. So you need to look at cycles and then some of the other macro relationships and so on that are also important. You may find a factor works better. In fact, you will find that a factor works better in some kinds of macro environments than others. That doesn't necessarily mean you wouldn't want to use it in your strategy, but you just want to be aware of maybe when you will be doing better and when you may be not doing better.

Jonathan Forsgren:

Speaking of digging in a little bit further, why do some strategies that sound so similar have such different performance outcomes? For example, value indices, value ETFs, and then value as a factor.

Melissa Brown:

And this again kind of comes down to this idea of a factor definition. When we talk about a factor mimicking portfolio, let's say for value, we say that is a long short portfolio that has exposure to the factor and no exposure to anything else. So you don't have any industry biases in it, you don't have any size biases in it. It's a pure factor. When we talk about a value index where many of the indices maybe bifurcate the world into value and growth based on some factor like book to price. And so that is going to give you a lot of biases. It may give you a small cap bias, it almost certainly will give you a big industry bias. So is it really that value is doing well or is it that energy stocks which are way overweight in your value index doing well? And so that kind of drives this discrepancy between what one person calls value and what another person calls value.

Mark Carver:

Yeah, I'd just maybe build on what Melissa's saying. I think for many of the people listening, what they're really thinking about is, "I'm looking at these various sets of investments, how do I determine what makes them different?" So what framework can they use? And often what we hear from clients is they start with ... if we're looking at value just to stay on this, I'm looking at all these different value managers. Let me first start with how it's defined. Is it defined with a single dimension like book to price? Are there multiple dimensions, and why? Why does one manager or one solution use a single dimension? Why do others use multiple dimensions? So you start with what's the definition of the characteristics of value?

Then you think about the notion of actually constructing the portfolio. So there's the selection, single or multiple dimensions. Then there's the construction effect. Are you looking at the names relative to their industry peers or across all names? Are you thinking about the constraints of the portfolio? For instance, you mentioned value portfolios might have a big bias toward a particular, not only sector but industry within a sector. So are there constraints put on that? How do you weight the names, how frequently do you turn over the portfolio?

Is it a monthly turnover, is it idiosyncratic with an active fund? Is it rules based with an index? So you should always be thinking about the selection. That's fairly straightforward. Then the actual construction of the portfolio in terms of dealing with all the constraints, dealing with the weighting and ultimately the turnover. It becomes a framework. It used to be you would evaluate active managers with the piece, people, process, performance, et cetera. You could do the same thing with factor-based strategies. You'd just adapt the framework.

Jonathan Forsgren:

Speaking of implementation and putting things together, Janis, can there be variations or different flavors to the implementation of a particular factor? Building off what Mark was just saying.

Janis Zvingelis:

Absolutely, and I think Melissa and Mark highlighted the main areas where we can customize a factor and these are dimensions you used. As Mark mentioned, if we stay on the value factor, do we use multiple dimensions? There's actually reasons and benefits from using multiple dimensions. Or do we use just a single dimension? The original paper by Eugene Fama and Kenneth French, they used one dimension. The other areas where you can differ in construction of the factor, as Mark mentioned, rebalance and frequency. How often do you want to rebalance? Do you want to rebalance once a year? Do you want to rebalance quarterly, monthly, even maybe daily? Then cutoffs. If you form a universe for the underlying factor, which stocks do you include? Is it the top quintile, the top quartile, is it the top half and so forth?

And another one, weighing scheme is a very, very important one. Do you use the market cap? Which is the original way of defining factors. So all of these areas will greatly affect how the value performs even though the broad definition of the factor is the same. I'll just mention there's a couple of examples. I'll mention one example that might be known to our listeners, which is a paper by Cliff Asness and Andrea Frazzini. It's called The Devil in HML's Details. So devil in value construction details. It's a very, very famous paper that came out in 2013 and basically what they do is they analyze the construction methodology of the original Fama-French value factor. And the way that Fama and French originally constructed the value factor in 1993 is they said we're going to rebalance it once a year at the end of June, June 30th, but to rank the stocks that go in that portfolio on June 30th, we'll use the data, price the book data from end of previous December, December 31st.

And the reason that they use that is to make sure that the latest version of the December 31st data is available on June 30th. And so on June 30th using the book value from December 31st makes sense, making sure that data is available there. But what Asness and Frazzini basically say, well why not use current price data? So you have a price data from June 30th from the time period when you rebalanced and use the book value from December 31st.

They did that and they compare the performance of the new value factor. And it is very interesting because the original Fama-French value factor actually can be broken down into this new value factor plus some momentum performance, which also makes sense because you're using a price. So there is some price signal that is six months old. And so I think the bottom line is the individual details of how you construct the factors are do you really impact the performance of the factor itself?

Jonathan Forsgren:

Well that leads me to my next question. Mark, how are you able to tell if a factor is the key driver of performance behind a strategy's performance?

Mark Carver:

I think there's two related topics that are sort of surfacing here. The first is for the factor itself. So typically the way a researcher would determine is the factor the variable that determines the performance difference is you would typically sort names on the characteristic and you build decile portfolios, and you're measuring the return gap between the top and bottom deciles or quintiles if you choose. You're seeing are there differences in the performance. And so the way we would typically describe that is do factors or does the variable distinguish returns? So you look at that, but I think what you are really driving at is if I'm an investor and I'm looking at a portfolio, it could be an index, it could be an active fund, whatever I want to decompose the behavior and the return drivers of that portfolios and factors become a very good way to do this.

Classically people have used Brinson attribution, they've looked at effectively stock and sector effects. By going beyond that, you're able to look at the style dimensions and how much did that influence. We've looked at this issue a number of times at MSCI. There was a paper by a couple of my research colleagues, Raman Subramanian and a guy named Leon Roisenberg back in 2018. They wrote a paper called The Anatomy of Active Portfolios. And effectively at that time what we learned was that roughly 55% of the return of active portfolios, the active return comes from factors, 45% from stocks. What everybody wants is an active fund, all the returns come from the stock dynamics. The truth is much of it comes from the factors.

We more recently did a study specifically isolating just 2022 because it was a very unique year. We looked at first 1,600 global managers, then any manager that was called value or growth. We eliminated those, we were left with 1,300 names. We took those 1,300 names, split them in deciles based on the performance of those managers. So top decile, second, all the way down to the bottom decile. What we found unequivocally was the biggest determinant in the top to bottom deciles was your exposure to factors. And in fact just the style dimensions that we keep hinting at value, momentum, quality that determined 31% of the active return of the top performing managers last year. The bottom performing managers, more than 50% of their active returns were coming from styles.

Now why does this matter? It matters for a few reasons. One is you want to make sure that the things you are hiring that portfolio to do are the things that are influencing their returns. So if you bought a growth manager, are they getting all their returns from value? Just to use a dramatic and extreme example. And that allows you to have that transparency. Obviously you want to make sure that the managers do what they say and those things are paying off or not. It allows you to be able to then compare managers across peer groups to see what's the level of exposure, how are these ideas implemented, what's the return in the risk payoff?

More and more, both wealth advisors and for sure asset owners are using factor attribution to get better insight into their portfolios and allows them to have a dialogue with their asset managers that you wouldn't have otherwise. And so that level of transparency, that level of confidence leads to better dialogues and ultimately leads to a better asset allocation across an equity or even a multi-asset portfolio.

Jonathan Forsgren:

Melissa, kind of failing off that, evaluating in the rear view, what do you say to skeptics who say factor investing is really investing looking into the rear view?

Melissa Brown:

That is a question that I have gotten a lot from clients, from consultants and so on. I think, yes, we are looking at reported data. So is every manager, by the way. What the advantage I think of the factor investors is that you're looking at variables that if we go through my evaluation process that I talked about that we know have had a historical relationship to future returns. So it's not really, I'm not just saying, "Oh, we would've done this last year." I'm saying if this characteristic was the case last year, here's how we would've expected this portfolio to do this year.

And so very similar I think to what any kind of manager is doing, they look at the information they know and they make assumptions about what that's going to mean for the future. In the case of quantitative investing or factor investing, what you're getting is a known quantity and if I look at this set of factors today, I'm always going to come to the same conclusion tomorrow. So I think that's a ... I don't like that statement of, well, you're looking in the rear view mirror. So are all investors. If we knew what was going to happen, yes, of course we would invest that way, but also it would be illegal. So we probably don't want to do that. So yeah, so I think that it's just using history. History repeats itself.

Mark Carver:

One of the things to maybe think about is the turnover of factor portfolios. And if you think about certain factors you might think of them as slow moving. So if Melissa and I build a value portfolio, quality portfolio, a low volatility portfolio, the turnover of those is relatively muted. And so what it's telling you is that the names that often show up as attractive on that dimension are fairly consistent. And so that's a big contrast to something like momentum, which is a much faster moving signal. And so this criticism is true, you are looking back, but all research is done by looking at data that is not tomorrow, it's yesterday. And so it's sort of inevitable.

The question is, is that data predictive? This is why those of us in the industry, the academics spend so much time really saying do these areas of interest, these dimensions that we care about, whether it's macro or style, do they help us distinguish returns? And you can see obviously there's evidence of that, there's bodies of research, there's papers we're citing in this session. But just the simple stuff like look at the turnover of those portfolios and it tells you that there's some stickiness to the names that look attractive to these dimensions.

Jonathan Forsgren:

Janis, can risk factor performance disappear, especially with all the attention to them?

Janis Zvingelis:

To answer this question, we really have to think about the reason for the existence of the factor risk premia. And I alluded to this earlier and there's a couple different ways that you can explain the existence of factor premia. The most prevalent one, it's a reward for undiversifiable or systematic risk. And then there's also some behavioral angles to explaining this. So I'll talk about the risk angle to this and we can touch upon the behavioral explanation later on in the conversation.

As I mentioned earlier, the reason that the factor premia exists is because factors tend to perform poorly when times are bad. So it's kind of a double whammy on the whole or a factor. And this is a very important point to highlight because factors are not sources of alpha, right? They're not a potential risk-free return due to mispricing or riskless arbitrage or new information or anything like that.

No, they are risky and some instances of when factors underperformed are value underperformance during '99, 2000 period, very well-known episode, underperformance of value during 2007 through 2009, and even the most recent underperformance, 2016 through 2020, very, very prolonged performance when a lot of investors kind of gave up on the value factor and said, "Well maybe value factors have been arbitraged away." In 20/20 hindsight, that obviously is not true. And also momentum, momentum crashed 2009, very, very famous episode. So unless practically everyone decides that value investing is for them, it's unlikely that the value or factor risk premia will disappear.

Mark actually mentioned a very good point that a lot of people are moving into growth investing. So long as there's the other side to the trade, so if I want to invest in value so long as there's other people who say, "No, I don't like value because it has these risk properties I don't like, I like to invest in growth," the risk premia that are associated with the factors will remain. I mean, in some sense we can think about the underperformance of the factors or the periods of underperformance of the factors as the opportunity or it sows the seeds for long term performance of the factors.

One last thing I will mention is that the current pricing of the factors, and there's various ways you can price factors whether factors are priced cheaply or richly at any given point in time. And so we at QRG track these measures and we can say that currently quality and momentum factor are cheaper than they have ever been since the beginning of 2018. And the value factor, except for the mid 2022 episode, before the great performance of the value in the 2022 episode, except for that time period, the value factor is also cheaper than it's ever been since late 2017, early 2018. So combining the observational data with this kind of a little bit more theoretical arguments on why the risk premia exist, I think we will have factor premia around for a long time.

Jonathan Forsgren:

Before I move on to my next question, you mentioned that between 2016 and 2020 factor investing might have lost a little bit of favor among investors. What's been the drive behind the renewed interest would you say?

Janis Zvingelis:

Right, so as somebody else mentioned earlier, the factors performed really, really well starting from late 2021. Value factor brought in some of the best performance that we've probably ever seen and it basically returned with a bang. A portfolio, let's say that tilted towards value have done really, really well. Mark also mentioned that quality has done really, really well and that is true. Quality usually does really, really well during a market under performance, specifically a short side portfolio does really, really well. And we put out a white paper on how factors perform in different market cycles and that's one of the things that stood out. So I think the reason that there's this renewed interest is because you have these factor performances that have done well recently.

Jonathan Forsgren:

Mark, Melissa, would you like to add? [inaudible], yeah.

Melissa Brown:

There's an old saying that financial products are sold, not bought and so I think for the intermediaries they are seeing that performance, and maybe you have a better perspective on whether the popularity is being pulled in or it's being pushed out. But I suspect that it's at least partly because I've got something easy to sell, this did better than the market last year, that makes it a lot easier for me to sell.

Mark Carver:

Yeah, I think that maybe I have a more positive view on our industry. I think there's a practical reality that we lived in a period of almost monetary charity from the end of the financial crisis until the Fed and other central banks around the world started aggressively raising rates. The impact of that was inflated asset prices. In the US market we had extraordinary concentration in effectively the technology sector and industries within technology that dominated market returns. That started to unwind certainly in 2021. Obviously it unwound really quickly in 2022. The result of that is cross-sectional volatility, or to put it in plain terms more market opportunity as there's a return to fundamentals. And we're seeing these dynamics of not only rising inflation and rising rates but the likelihood of rates and inflation staying elevated and they may not rise, but we're not going to be likely in a period of zero interest rates in the near term.

We're seeing effectively a change in global relationships. We lived in a period of 50 plus years of relative peace, global economic cooperation. That's starting to unwind. This de-globalization does mean that the way investors will allocate capital is shifting. Again, this means that there's opportunities for investors to diversify in ways we haven't seen over the previous 10 or 12 years. The result of that is clients are reallocating capital. When it comes to factors specifically, the big difference we see in the intermediary market versus the asset owner market is the way they implement. So both are super interested in factors. In the wealth space, it's much more around single factors and many of those organizations, the wealth firms themselves, they build model portfolios where they're implementing to specific single factors. This is not new. For the last 20 plus years they did this except they only did growth and value. Today there's more choice.

In the asset owner space there's more interest in multifactor portfolios, and even those are evolving where the factors that are included there are relatively new, today, many of those asset owners want climate principles brought into that portfolio. There's this big move toward net-zero. So the result is you're looking at multifactor portfolios with an inclusion of ESG and climate principles. This is changing asset pricing, it's changing the opportunity set. So it's not a necessarily bought versus sold. There's a different investment opportunity set today than what we've seen over much of the last decade plus. Is that good or bad? Maybe I'm just not quite so-

Melissa Brown:

As cynical as I am? [inaudible]

Mark Carver:

So cynical as Melissa, yes.

Jonathan Forsgren:

As you're saying that it just occurs to me that we have so much more data available now. So you can parse apart and look at different factors more easily.

Mark Carver:

Of course, and you have more investments available to you through ETFs, where ETFs have countries that are now investible to you that years ago would not have been. So the ability to actually implement investment views today is so much more mature and effective than what it has ever been partly in the intermediary world because of the ETFs and that opportunity set does mean that there is a rotation across factors and some of it is performance chasing.

Some of it is what Janis said, is people have a view of the macro environment. There are characteristics that are obviously cyclical, they perform better in certain environments and investors are trying to study that and then position the portfolios accordingly. That's the growth of model portfolios in the intermediary space. Asset owners have been doing this for decades. They create a policy portfolio and they implement it not with ACWI, to our chagrin at MSCI. It'd be great, just buy ACWI, but no, they decompose that, they break it up, they buy parts, they over and underweight. This is in some ways natural asset allocation.

Jonathan Forsgren:

Melissa, what is the relationship between dispersion and factor performance? For example, if the spread between the best and worst performing stock is high, does that mean the factor performance will also be high?

Melissa Brown:

Well I think that's a common notion that if there's a big spread between the best and worst performing stocks, the reward to being right is bigger. Also, the penalty to being wrong is also bigger. But let's assume that in times that these factors are doing well that you're going to see that dispersion drive the performance. But we actually have found that that's not really true. That most factors, their performance is related. I think as Janis was saying, to economic factors or to other, it might be Fed policy or it might be other things, but it's not necessarily that you would have to be in the right place at the right time to get that benefit from the dispersion. And I'm not sure that that is always happening. So it's an interesting concept that we would hope would be true but at least with our data that we've looked at, does not seem to be true.

Jonathan Forsgren:

All right, Mark, I'm going to bring it back to you. You made a comment earlier about factor rotation. So what are the factor rotation trends you're seeing? For example, growth versus value, any others out there?

Mark Carver:

The one that is obvious to everybody is the growth to value rotation. What's really curious, let's just put a magnitude on that. If we look at just the most basic definitions, looking at our style indexes last year, MSCI USA value outperformed USA growth by let's call it 27%. The only year going back in our long-dated history that was better was the year after the tech bubble burst back in 2000. So it's literally unprecedented outperformance for value, and the result of that has been a lot of people trying to argue, why? Is it a duration play, an equity duration as a concept? The fact is we did see that that performance gap, the result was that a lot of people rotated into value strategies this year. We know that the performance of growth versus value, it doesn't tend to be a 12-month cycle, it tends to be longer dated cycles.

I mentioned earlier in the client survey over I think it was 35%, 36% of clients said they intended to over-allocate value in 2023. Not surprising giving that rotation, but the fact is markets are moving really quickly right now, and this year growth has outperformed value by a very significant percent, something like 18% in the first quarter. Not quite that level as we sit here today, but it's a large rotation back. Why? Something Melissa brought up earlier, the sector effects. Technology has come back, and particularly big cap tech stocks have done extremely well.

The fact is factors do tend to be cyclical. The speed of that rotation is a bit surprising and notable, but we have seen that really since the bottom of the pandemic, it was just the extraordinary returns to some of the volatility dimensions. We've rotated across that, but the impact of that is investors are being more dynamic in their allocations for good and bad probably.

But we are seeing that rotation. The thing that I think Melissa said that is really important to highlight is this dispersion we're seeing in terms of certain signals, and we looked for instance at just one aspect, what happens when an analyst revises earnings? So we have this library called Factor Lab, we looked at just literally short-term forecasts on earnings per share growth and if we create decile portfolios, we're seeing those names that have the highest positive revision in EPS growth versus those that have the highest negative revision, that spread in global markets is 1,000 basis points in the last 12 months. It's extraordinary. In the US market it's 1,600 basis points.

What does that tell you? It tells you that investors are intently focused on those names that will deliver earnings and those names that don't are getting punished. This is why it's really important when you're selecting an investment, really understand the methodology. That thing that you believe has a common name, how is it actually built? What does that mean in terms of the names that it'd be exposed to and how do I verify that exposure through looking at things like factor modeling, looking at exposure, risk, attribution, et cetera?

Melissa Brown:

If I could just add something to that, Mark, you mentioned that the asset owner world tended to be more multifactor where the individual investor is more single factor. I think that really brings up one of the reasons that a multifactor approach is a really good way of ... well, it's taking advantage or avoiding the pitfalls depending on how you look at it of just choosing the one factor, because it's very rare going back historically to see everything not doing well at the same time. You typically have, well it might be that in this particular year growth is doing better and value's not doing so well.

But if you have them together and can find stocks that ... or build a portfolio, it's not an individual stock basis, it's a portfolio that balances those two things. You may have some of the problems of that big spread mitigated. And so we're big believers in multi-factor approaches. You still want to know what are the factors, what are the decisions going into each of these factors. But I think that is a good way to diversify across some of these added risks.

Jonathan Forsgren:

Janis, I'm going to come to you next. Can you time factors?

Janis Zvingelis:

Actually, my answer will be part of following up to Melissa's point, which is that we are also very big believers of diversifying factors, because answer to your question whether it's possible to time factors, is maybe, and it's a very qualified maybe. There's a couple different ways that we can go about timing the factor performance. One is using the price of factors. So there was a very famous kind of academic brawl if you will, between Research Affiliates and Rob Arnott and Cliff Asness of AQR several years ago. There's multiple papers going back and forth. And so the crux of the argument was, is it possible to use the price of the factor? So basically what you do is you take some kind of a pricing multiple of the longside of the factor and compare that to the pricing multiple of the short side.

And if these pricing multiples are far apart, whether you look at the ratio or distance or subtract them, if they're far apart then you know that the factor is richly priced. And if it's not, then the factor is cheaply priced. And historically what has happened is that when the factors are expensive then their subsequent performance tends to be poor, and vice versa. When factors are cheap then their subsequent performance tends to be good. But the problem really here is that the link between the relative pricing of the factor and subsequent performance is very tenuous. It is true that over long periods of time this performance link transpires, but the timing of it is very, very difficult. And that was kind of the crux of the interchange between the papers of Research Affiliates in AQR.

The other way that one can go about timing the factor is called factor momentum. And that's a very, very interesting idea. It's been around only for a couple of years, and there's a couple of different types of factor momentum. And so what is the idea? The idea is that, similar to individual securities that exhibit momentum, right? So securities that perform poorly tend to do so in the future and vice versa, factors themselves also can exhibit this type of behavior. So factors that have tended to do poorly in the past tend to do so in the future. So 2016 to 2020 value under performance is a very good example, and there's a couple different ways you can go about implementing this. One is the so-called cross-sectional factor momentum where you say I'm going to grade everybody on curve, right? So I'll list all the factors and I'm going to go long the best performing factors and short the worst performing factors.

It's a disadvantage of that approach is that sometimes everybody is doing well. So you might be underweighting some factor that is doing well and vice versa. Now, the other approach is this time series factor momentum approach by [inaudible], also associated with Research Affiliates. And it's a very interesting approach and in fact, we at QRG implemented in our portfolios. Basically what it says is that if a factor ... and this kind of goes back to Mark's point about earlier about rotation in and out of the factors. And the question is actually can you do it in a way that is beneficial to the portfolio? And the closest that we've come as financial profession and researchers to finding a method of doing this is this time series factor momentum.

The idea there and it's very straightforward, you overweight factors that have done well in the past and you underweight factors that have done poorly in the past, and it is one of the most kind of hands on methods that actually so far has worked in the portfolios.

One last thing I will say to me that is very interesting and the co-authors [inaudible] mentioned this in their paper is that we can look at the momentum factor. So basically they argue that there is no distinct momentum factor. Momentum factor is an aggregation of factor momentum in other factors. So if value is performing well that will get reflecting in momentum factor. If quality is performing well, that will get reflected in the momentum factor. So I found that to be a very, very interesting idea. Yeah. So as I mentioned we at QRG do use some of this factor timing in our portfolio.

Jonathan Forsgren:

Janis, are there factor strategies that advisors can implement themselves?

Janis Zvingelis:

Absolutely. So constructing a full-blown tool or strategy is a rather difficult task. It involves all types of data issues and optimization and so forth. But there are some rules of thumb, if you will, that advisors can implement by themselves and in their clients' portfolios. So QRG recently put out a white paper on performance of various factors, and I mentioned it earlier, in different market cycles. And there's a way how you can identify market cycles rather than looking at the [inaudible] data, we looked at market performance and then used some statistical techniques of breaking down what we call a recession and then early recovery and in middle expansion and then late expansion.

I won't list all the examples or all the results that we discovered in the study, but I will mention a couple of them because they were brought up earlier as well. So one of them that I found very useful and very interesting is that, so quality factor performs really well during recessions. Mark brought this up, now's kind of that time when market is not doing so well or it's very volatile, economy is kind of somewhat iffy. We discovered the same thing in our study. Basically most of quality's performance comes during these recession periods and importantly on the short side of the factor. So as an advisor you can basically eliminate low quality stocks in your client's portfolio and that will bring about all the benefits of quality factor.

Another example of how you can use a rule of thumb of improving your client's performance is the result that we discovered in the paper, which is that you can overweight value stocks, as value stocks tend to do really well during recession time periods, late recession especially, late recession time periods as well as early expansion time periods. And final thing I will mention is that another thing that advisors can do is overweight investment factor stocks. So basically these are high dividend paying stocks, as they tend to do well, again, similar to value stocks in late recession or early expansion market cycles. So while the implementations of these signals, how you implement them, how do you implement these rules of thumb in your portfolio is also important. I think this might be some valuable information that our listeners can use.

Melissa Brown:

I think that one of the real benefits and the way you get the most out of a factor-based strategy is by being very disciplined in implementing it. And so for an individual advisor, I think it's very easy to say, "Oh yeah, but I really like that company," or, "I don't want to get rid of it," and I mean it's not just advisors. It's any investor will come to the problem with their own biases. And so it's really important if you're going to really try to implement a factor-based strategy to be as disciplined as possible and not try to second guess what a model is saying.

Jonathan Forsgren:

Staying with you Melissa, and this is at the risk of opening up a whole nother masterclass. Can you look at factors? We've discussed factors mostly looking at equities, but can you talk about factors looking at other asset classes, for example, fixed income?

Melissa Brown:

Yeah, and I think Janis mentioned it early on that it's not just an equity based type of strategy, and there are strategies in fixed income, for example, value. If you look at wider or narrower spreads versus other similar types of credit at least, we do see that that can lead to outperformance. It just so far has not been implemented to anywhere near the degree that we've seen in the equity market, but it certainly is possible and can be profitable.

Jonathan Forsgren:

Mark, we're talking about advisors. So what should advisors know about the convergence of factors with other investment themes like ESG, climate and megatrends that we're seeing?

Mark Carver:

Well, I think what they should know is what they're telling us, right? In that, in their minds, they tend to smush these things together. Smush is not a technical term, but it is in their minds in some ways a convergence across specific characteristics that they want to have in their portfolio that help them express their views. The biggest I think, trend in the industry over the last several years has been the inclusion of ESG and now more specifically climate principles inside of portfolios. We all know the number of asset managers, asset owners, wealth organizations who've met, made some commitment toward net-zero or to reduce their carbon footprint. Now they're investing that way.

It is possible to bring a climate consideration into a factor portfolio. In our new equity models we have a carbon efficiency factor, we have ESG, so you can see the risk and return to those specific dimensions. We also know that there's this view toward building portfolios toward these megatrends. Thematic ETFs have been incredibly popular among probably a lot of the people listening, but they're thinking of those as different characteristics that they want in their portfolio. They're converging on these things.

What it really represents is that they have views, they want to get ways to expose those views. In the institutional space, what we're hearing is more appetite for the index to deliberately include these principles. So you start with a multifactor index, you have a climate target, and then you bring in innovation or some secular trend. I suspect that you're going to see more and more of that and then it'll find its way into public funds that the wealth advisors will get access to.

Jonathan Forsgren:

Maybe we'll leave our viewers with what are the top three factors that are most interesting to each of you right now?

Janis Zvingelis:

Yeah, so for us, and I'll speak from the experience of QRG, the most interesting factor that we're working on right now, and it is volatility, is risk premia or volatility risk. And we are using it in constructing options strategies. So this kind of goes back to Melissa's point earlier that there are factors not only in equity but also in other asset asset classes, fixed income and options. And so for us, for QRG, we're very, very excited to bring out a product that focuses on this small [inaudible] risk premia in options.

Mark Carver:

Yeah, I think the things that are really interesting to me are things that we're diving into from a research standpoint, and it's particularly this notion of crowding. Can you look at the relative bubbliness of an asset? And that asset could be a factor, and Janis talked about the value spread that was the fight between RAFI and AQR. But the truth is that's a single component of what you would want to consider when you're measuring crowding. So we're doing a of work on that and then ultimately creating a factor against it. Super interesting. I think from a practical standpoint, profitability's very interesting today because we know that asset prices are moving a lot based on earnings and ultimately profitability. So I think the combination of crowding and profitability are super interesting.

Melissa Brown:

Yeah, I think just the idea of shorter term factors when we talk about a lot of these factors take a long time to play out, but many managers, particularly hedge funds, but even other managers don't have the luxury of having a one-year horizon. And so just looking at what kind of payoffs can you get from the shorter term I think can be very interesting. And while it might not be an investment strategy, it certainly can be an important risk management strategy, particularly as we see things like we've seen over the last few months with bank failures or other dislocations in the market, having a shorter term view can be helpful in managing the risk.

Jonathan Forsgren:

Well, Janis, Mark, Melissa, thank you all for sharing your insights today.

Mark Carver:

It's great to be here.

Janis Zvingelis:

Thank you for having me.

Melissa Brown:

Thank you.

Jonathan Forsgren:

And to our viewers, thanks for watching. For Asset TV, I'm Jonathan Forsgren. We'll see you next time.