Apr 14, 2022

Episode 196: Sebastien Betermier: Hedging, Sentiment, and the Cross-Section of Equity Premia

Welcome back to the show all about sensible investing in Canada! Today we have yet another masterclass with a wonderful guest, Sebastien Betermier. Sebastian is an Associate Professor of Finance at Desautels Faculty of Management at McGill University, where he teaches investment management, applied investments, and pension funds retirement systems. We have a deep, thoughtful, and precise conversation with him about his recent research and papers, much of which stands in contrast to our usual fare on the show. In our chat, we dive into the nuts and bolts of asset allocation, hedging risk, and his research into what demographics can teach us about investment behaviours and returns.

We also hear from our guest about interesting topics of expected persistence and tilting towards value stocks, before shifting the conversation towards homeownership and property investment. Sebastien provides some sound advice around when it might be a good idea to purchase property over other asset classes, and we evaluate this position from a number of different investing perspectives. Lastly, we spend some time looking at pension plans, and what we can learn from those available in Canada right now.

 

Key Points From This Episode:

  • Sebastien explains the theoretical relationship between labor income and financial asset allocation. [0:04:30]
  • Findings on hedging labour income risks and the paper that Sebastien published on the subject. [0:06:47]
  • The relationships between risk and age, gender, wealth, and heterogeneity across households. [0:10:05]
  • Unpacking Sebastien’s investigation into value and growth investors. [0:12:07]
  • The effect that the characteristics of labor income have on the rate of progression on the value ladder. [0:18:43]
  • What we can learn about expected persistence in the value premium. [0:22:39]
  • Weighing the possibility of predictive demographics for future value premiums. [0:24:29]
  • Advice for young investors looking to tilt towards value stocks. [0:27:50]
  • Explaining differing returns according to the characteristics of people. [0:29:41]
  • Sebastien explains the factors of markets, wealth, and age, in the pricing model. [0:31:24]
  • Understanding how investors tilt to age and wealth factors, and what these portfolios look like. [0:38:19]
  • The impact of age and wealth factors on wealth inequality, and how younger investors can combat this. [0:42:19]
  • Possible rationales for homeownership and the storage of wealth in housing. [0:44:26]
  • The household characteristics that are predictive of larger allocations to housing. [0:48:49]
  • Economic importance of risk-free benefits of homeownership. [0:52:15]
  • The decade-long rule of thumb for purchasing property; Sebastien weighs in. [0:55:31]
  • Why asset-only performance is not the only correct way to measure the success of the Canadian pension fund model. [0:58:50]
  • Differentiating asset-only performance and liability-hedging performance measurement. [1:02:29]
  • A list and explanation of the assets that Canadian pension funds use for hedging real liabilities. [1:04:03]
  • Lessons from the Canadian Pension Plan for individual investors and firms. [1:12:54]
  • Sebastien’s personal definition of success: making the most of opportunities and a balanced life. [1:16:07]

 

Read The Episode Transcript:

So let’s jump right into it with a great question off the top. So can you talk about the theoretical relationship between labor income and financial asset allocation?

Yes. Think of our wealth, the composition of our total wealth. We have financial assets, we have real assets or home cottage, we also have our wages. The bulk of our wealth comes from our wages. To think of our human capital, the present value of all our wages from now on all the way until retirements, that present value that wealth is frequently for most individuals in order of magnitude greater than the amount we directly invest in stocks. And that wealth is risky. In good times, you may get a bonus, you make get a promotion. That’s great news. In bad times, you might lose your job, you might struggle finding a new job. When you find a new job, you may not have a job of the same quality as you used to have. It is not possible to easily insure against that risk, right? It’s a risk that we bear. We do have social programs, of course, in the event we are retired, we lose our job, but overall, it is a major risk that we face.

And so what this means is when we are thinking about building our investment portfolio, well, it’s a bit of the elephant in the room. You have this major source of wealth, but major source of risks that we’re all facing. Some of us are more exposed to that risk than others, depending on the type of job that we face. And so the riskier our human capital is, then the more conservative we may want to go on our investment portfolio so that when a recession hits, we don’t have a double whammy where we lose our job and we may take a hit on our investment portfolio as well. So we may want to be more conservative in the way that we invest. And then within the stock portfolio, we may want to hedge away from particular industries that we are exposed to. And so this is where the theoretical relationship between labor income and financial asset allocation comes in.

That makes a lot of sense. Theoretically, you have a paper in the Journal of Financial Economics, Hedging Labor Income Risk, and you tested that theory empirically, what did you find?

So we find significant evidence of hedging by individual investors. To provide a little bit of context, this channel is as clear as it is theoretically is difficult to identify empirically because when we’re looking at in a data, there are other confounding factors at play. Think of taking two individuals in different jobs. It may be that the individual in the riskier job may be more risk tolerant to begin with. So that means that when we’re looking at their asset portfolio and we see a greater share invested in stocks, is this because of the job that they’re in or is this because of their innate risk tolerance for stocks? So we need to tease these channels apart.

So what we did in this study is a number of things. First, we work with extremely precise data from Scandinavia, hundreds of thousands of investors. We know the exact sector that individuals work in, we know their composition of their portfolios at the end of each year, we know their family situation, their housing situation, we can control for many factors. So that already helps. And then we go one step further and we say, well, okay. Instead of just comparing the portfolio of individual A and the portfolio of individual B and here again, we might have some latent factors that are hard to control for such as their risk aversion, we’re going to focus on how their portfolio switches over time in response to a job switch. So we can identify those events. And in particular, we’re looking at job switches across industries and then we’ll look at what happens when I switch to a safer industry and what happens when I switch to a riskier industry and how do I rebalance my portfolio?

And here we find significant evidence that individuals do rebalance their portfolios depending on the nature of the switch. So if they tend to switch to a riskier industry, we find that individuals tend to reduce their risky share on the financial side. If they switch to a safer industry, they end up increasing their risky share. Just to give you some numbers. If you switch from the private sector to the public sector, on average, the risky share would go up or down by 2.5%. If you take it more of an extreme switch from the safest industry all the way to the riskiest industry, we find that the portfolio share invested in risky assets would change by about 35%. So that’s a bit of an extreme case, but it’s large to give you a sense of the magnitude of these results.

Did this surprise you?

Not really, because again, when you’re thinking of our human capital, it is the elephant in the room. You find that for most individuals, the financial wealth that they have invested maybe 50,000, 60, 100,000, 200, when you’re looking at the present value of their human capital, we’re talking about 1 million, two, five, 10 million, it’s a big amount, right? If I were to lose my job right away, I would lose the bulk of my income. And so the fact that we find that this risk channel appears in the way people make their investment decisions makes sense in this context.

It’s not surprising because it’s consistent with a theory, but it is surprising because you don’t think of or at least I don’t think of the average household making rational financial decisions. It makes me wonder. So what you just talked about was the average result, right?

Yeah.

Was there a lot of heterogeneity across households?

Yes, there is. Portfolios of households are quite heterogeneous, especially once you’re looking at their direct ownership, which oftentimes are concentrated in few stocks. Here for this particular result, we found that when you add up both the directly held stocks and the mutual funds and look at the total risky share that moves around, again going from private to public in general increases your risky about 2.5%, there’s clearly a lot of patterns at play, which is why it is important to control for a number of characteristics.

But the way I like thinking about these issues is thinking about a notion, right? We see the waves on the surface. It can be quite volatile, but underneath, you have some strong currents. And here by looking through the data, by looking through a wide number of individuals, by sorting through the different sources of heterogeneity, we find those currents that are quite strong, that maybe revealing of strong comfort level toward taking risk or not as a function of everyone’s personal finances.

Did you find other relationships with other characteristics like age or wealth or gender?

We do. And in fact, you have an entire literature in finance called household finance, whose objective is to understand how investor portfolio decisions relate to a number of their characteristics. And so the usual suspects there would be their age, would be the level of debt that they have, perhaps scaled by their income, their wealth, their gender has a key effect, just a key relationship with investor portfolio decisions, the level of education, the job situation, the sector that I’m in, the risk of that sector, the family situation, whether I’m married, single, have children, all of these end up playing a role in driving that heterogeneity in investor portfolio decisions.

You had another fantastic paper in Journal of Finance, Who Are the Value and Growth Investors, where you looked at value tilted households while it’s answering the question on the paper, who are the value and growth investors, what did you find in that sample?

So as the title suggests and as you said, we looked at which types of investors tilt toward value stocks and which types of investors tilt toward growth stocks. And the purpose was to understand more the literature around the value premium and the value premium puzzle, which has been a key topic in finance for many years. What we do find is very significant evidence, again, looking at very detailed Scandinavian data. And in that particular dataset, we are looking at the entire population and not just at the level of risky share in their portfolio, but their exact holdings in every single stock, in every single mutual funds.

And so we construct for each household a tilt, a tilt to the value factor, right? So how exposed their portfolio is to the value factor? And then we identify, well, which of these households tend to have the strongest tilt toward value and which of the households tend to have the strongest tilt toward growth on the other side? And in generally speaking, what we find is individuals with a value tilt, they tend to be older, they tend to be wealthier, they’re less exposed to the macro economy through their labor income and they’re more likely to be female and they tend to have stronger balanced sheets.

Fascinating. And how do these value tilts change over time or do they change over time?

So we observe a very particular pattern over time. When investors are young, they tend to tilt their portfolios toward growth stocks. And then as they age, slowly but surely, what we will find is they migrate toward value stocks. We say they climb what we call the value ladder. And the climb is steep. Over the full life cycle, the climb, approximately the equivalent of half the value premium that we observe in the data. So that’s a big change in the tilts from about age 25-30, all the way to age 65-70.

The data from Sweden allows us to track the holdings of individuals over time. We have about nine years of data so we can track the evolution of portfolios. And what we find is fascinating, what we find is when you take a group of individuals and they’re 30, so you’ll find more of a growth tilt and then you follow those individuals. And what you realize is that five years later, when they’re 35, they end up taking the place of where the 35 used to be five years ago. And the 35 end up continuing to climb the ladder and they take the place of where the 40 year olds used to be five years ago. And it’s a very linear pattern that we find, again, one of those strong undercurrents that you may not necessarily see with the naked eye, but once you look at through those large data sets, it becomes hard to not see. And this linear pattern goes all the way to and through retirement.

Wow. Do you think this is just, it’s a shift in hedging demands over time, that’s what’s causing people to climb the value ladder?

We think it’s a mix of effects. It’s a strong result. And oftentimes when you have a strong results like that, you have multiple drivers coming on. To tease apart, the multiple possible channels that could explain that migration, we create an econometric model where we are going to regress or see how that value tilts corresponds or can be explained by a number of characteristics, take age, the balance sheet strength, gender, human capital, and so forth. What we’ll find is that in general, age explains about 60% of the migration. Here financial theory says that growth stocks, they may be risky in the short run, but they tend to be safer over the long run. So a lot of their risk comes from transitory price shocks, which you may not face as much over the long run. So they make more sense for younger folks who have a long horizon. And so as they age, that benefits diminishes as the horizon gets closer. And so that can explain part of the shift toward the value tilt.

Another driver is human capital. We believe that human capital explains about 20% of the migration. So that’s 60, 20 or 80% here. Here, financial theory says something else as well. It says that value stocks, in addition to the horizon effect we just talked, value stocks end up being more exposed to bad recessions. These are stocks with low market value relative to book, there’s few growth options left in them. They tend to have more leverage, they tend to have more operating leverage, more fixed costs. So they have a harder time adjusting their production model during bad times. So they may be more exposed to default in times of bad recessions.

And so when you’re looking at the cross-section of individuals, you have younger folks, older folks, the younger folks have the bulk of their wealth invested in human capital, little financial wealth, whereas among retirees, it tends to be primarily financial wealth. And so the younger folks may want to hedge away from that risk. And part of that means, well, not tilt as much toward these stocks as some of the other investors.

And then the third driver, which we believe explains about 20% of the migration. So we’re at 60, 20, 20, what? A 100 now. So the remaining 20% is the strength of your balance sheet. Again, going back to the concept of the composition of wealth, when you’re young, we find that the bulk of your wealth is in your human capital. It’s in real estate, but it’s very levered at that point, right? So you have some financial wealth, but your cushion is fairly small. There’s also a lot of liabilities. There’s the mortgage. You may have liabilities through your kids and other types of student loans that might remain. Whereas when you’re more mature, you tend to have accumulated more financial wealth and you don’t have as much liabilities. And by the time you’re 60, 65, typically the mortgage is being paid, right? And so these individuals have more bandwidth to take some of those risks and they’re not as exposed to those bad recessions as the younger folks. And so again, it makes sense to tilt toward value if you have a stronger balance sheet and you can afford to bear these shocks.

Do the characteristics of labor income of human capital, do they affect the rate of progression on the value ladder?

They do. When we go back to the theory that financial labor income affects our investment portfolio decisions and in particular, our value tilt, what the theory predicts is that individuals who are more exposed to these shocks maybe the most eager to tilt away from value stocks in their portfolio. That’s exactly what we find. So what we do is the following. You take two investors, same age, two young investors, two mature investors, but we’re looking at individuals with different exposure to microeconomic risk. So here we’re looking at the sectors that they work for and we’re looking at how cyclical that sector is. Some sectors are more cyclical than others. What we find is individuals in the more cyclical sectors, they tend to tilt away more away from value stocks than individuals in safer sectors, less cyclical.

But what’s interesting is that differential in the tilt is greatest among the young and it shrinks over the life cycle. And by the time those two individuals in different sectors are retired, that differential is gone. And by that time, we know that their exposure to human capital now is more similar because they’re out of the job markets. And so what’s interesting is you find those value ladders in a way that are very unique to the sector that you work for. And again, the results are consistent with the financial theory.

That’s really interesting. So what new information did we learn from all this work about the value premium? Like, how do you articulate what you learned and have added to the research into the premiums?

In the finance literature, there’s a large debate about where the value premium comes from. Just a clarification, what I mean value premium is the fact that value stocks tend to generate higher returns than growth stocks and returns that cannot be explained by their exposure to market risk. So set differently, value stocks tend to be cheaper than what our standard models predict. And we’re not sure why value stocks tend to be cheaper and as a result tend to generate higher returns over the long run. You have some studies who argue that the value premium could be the result of behavioral biases. Value stocks were like fallen angels, they’ve had bad news, investors are overreacting in a way to these bad news. They move away from these stocks, the price goes down and they become abnormally cheap. So this is one theory.

Another set of theories are more risk based, arguing that the value premium could be compensation for risk, additional risks that the standard models do not take into account. And so because they’re riskier, the value stocks, well, then they’re cheaper. They deliver higher returns in compensation for investors to bear the risk. And so we have that debate about where the value premium comes from. It’s a longstanding debate in the literature. And then our strategy as finance researchers was to say, well, okay, one way of learning more about what’s going on, it’s to study the ownership of these stocks. Who owns the value stocks? Who owns the growth stocks? And why? How are these investors different from one another? Is it the case that the value investors are in a better position to withstand possible risks? Or is it the case that value investors are perhaps more educated less prone to behavioral biases? And then by looking through the heterogeneity in that ownership, we can extract information about what’s going on with the value premium.

Does that evidence of a hedging demand? Tell us anything about the expected persistence in the value premium. And just to elaborate on that, we’ve seen in the last 10 or so years that the value premium has not been positive. And so a lot of people have called for the death of the value premium. Does this finding tell us anything about expected persistence?

It does. If the heterogeneity in the ownership that we observe is driven by risk exposures and individuals remain just as exposed today as they used to be yesterday and 10 years ago, then we should expect the value premium to persist. Now to be clear, when we’re talking about the persistence of the value premium, we’re not talking about high realized returns year after year, we’re talking about on average, over the long term, higher returns than standard CAPM market risk can explain, right? So there could be long periods, extended periods where the value premium stocks underperform, especially in case of bad recessions if that is the main risk that they’re exposed to. But the answer is yes, in that sense, there is persistence being predicted by these theories.

One of the recent findings in the literature, which is interesting and it’s worth delving further into is that even though the value premium itself has gone down as measured by value to book market measured in conventional ways, increasingly the value of firms is coming from intangible assets. And so once you control actually for the book value of stocks, including intangibles, and you redo the sorting of growth and value, you find that the value premium is still there, but you have to account for the intangible assets, which have become increasingly present in today’s economy.

So could demographics be predictive of future value premium if there’s an increasing hedging demand from the younger investors as they get older?

It could, but I wouldn’t necessarily view it as an increase in hedging demand from the young investors. It could be, as you said, if exposure to recessions is going up. Where to me the demographics are perhaps more revealing is with the shift in the wealth mass across investors over the life cycle. And by this, what I mean is we have a baby boomer population that is about to retire or has started to retire. The baby boomers own a substantial fraction of total assets. They used to be exposed to economic risk through their jobs, but increasingly they’re moving into retirement. And so if a lot of the wealth is moving gradually migrating to more of a retirement type of status, this may actually imply a decrease in the value premium moving forward because the aggregate investor, including both the retirees and the active workforce in aggregate may care less about these risks if a lot of the wealth is moving more into the hands of retirees. So in that sense, I would say that demographics can predict changes, perhaps a decline in the value premium moving forward.

Interesting. But a decline in the premium would also imply a closing in the valuation gap. So if people own value stocks now, they’d expect a positive realized return as the premium declines. Is that right?

Yes. Because as you said, you oftentimes negative relationships between the realized returns and the long run moving forward expected return that we can have. These demographics shift that we have in mind, they’re very slow moving. So we may not see them in the next one to two years, but we may detect them in 10, 15 years as there’s a slow migration of capital moving away from active employees in the workforce.

Right. And of course, as you mentioned earlier, a change in the value premium is the change in that underlying current, but we’re really only going to see the waves year to year.

Exactly.

So your empirical finding suggest a risk-based theoretical explanation for the value premium. Is there still room in your findings for behavioral explanation?

Yes, we believe so. As I was saying earlier, when you have a premium, like the value premium, which is large on average over the long term five, 6% per year, it’s hard to believe that this would be entirely driven by one or even two factors. My opinion, you have a number of factors at play. We do find patterns in the data that are consistent with behavioral explanations as well. We do find that gender tends to have a strong predicting power on who tends to tilt value and growth and gender. So female investors, women are more likely to tilt toward value stocks. We also know that women tend to be less exposed to behavioral biases than male investors. So this is consistent with behavioral theories. We also see that investors tend to own fairly small numbers of stocks in the data. They may not rebalance as actively as we expect them to. So for these reasons, there is certainly room for behavioral explanations as well.

So knowing all this, what is a young person that may have low financial wealth, what do they need to consider if they want to tilt their portfolio towards value?

What our study suggests is the value premium is partly a compensation for exposure to risk. What this means is that tilting your portfolio toward value stocks only makes sense if you are in a position where you can bear that risk. There’s a saying in financial markets that there’s a home for every risk. And the question is, are you the right home? Are you willing to bear those risks? You may earn higher return, but you need to prepare for the eventuality that your investments may not turn out the way you hope, especially in bad economic times when you might lose your job and those investments might lose the most as well. We saw in 2008 during the Great Recession that value stocks were hit the most.

And I would say that every situation is different. Some individuals have risky jobs with the ability to bounce back fast, some have a partner with another job in a different sector that can mitigate some of the risks. Some have a house or mortgage and a lot on the line where if they lose their job, there might be a chain reaction of events unfolding. So there needs to be a very careful evaluation of an individual’s willingness and ability to bear risk. And whether these premier that we observe in the markets may be the right premier for these investors.

Such an important lesson. Like we talk about differences in expected returns and okay, hey, small cap value stocks have higher expected returns than large cap growth stocks. And a lot of young people will say, “Great, I want to do that. I want to take as much risk as possible.” But I think what you’re saying is so important that there’s a whole other side of the, I guess, of the balance sheet, the human capital side that has to be considered in making that decision.

Exactly. Yes.

Okay. So we’ve heard from you so far that investor characteristics are related to the value premium, that is the characteristics of the investors that hold the stocks. Does that also mean that if we sort stocks by the characteristics of people that own them, does that also produce factors that explain differences in returns?

Potentially yes. When we think of the risk factors that drive stock returns, we think of risk factors that investors are worried about. So in principle, these factors should affect the makeup of our portfolios, especially to those of us who are quite exposed to those risks. And so from there, that tells us there may be a possibility to extract what those underlying risk factors are going to look like by looking directly at the portfolios of different types of investors with strong tilts. But empirically it’s an open question, whether as you say, we could construct pricing factors directly by creating portfolios of stocks sorted by the investor characteristics.

When we’re looking at most of the risk factors that we talk about in the data, most of them are constructed from firm characteristics, not investor characteristics, right? Think of size, think of value, think of profitability investments, these factors work well empirically at explaining the cross-section of stock returns, but they don’t have immediate links to the way investors make portfolios. So it is difficult to understand why they may contribute to higher, low returns, which was the whole motivation of looking for the value growth who owns these stocks and why.

So interesting to think about. So you looked at this in a recent working paper called What Do Portfolios of Individual Investors Reveal About the Cross-Section of Equity Returns? So what are the factors in the pricing model you test?

So first, what we’ll show in the model is that theoretically, the common factors that lead investors to systematically tilt away from the markets, these factors should explain the cross-section of returns across stocks. And the idea is, look, if we hold the markets, we have a market’s risk market factor that comes in. And then if we tilt away, for whatever reason, we tilt away systematically away from the markets, those tilts should predict what the excess demand for particular stocks should be, and as a result, what their pricing should be relative to standard models like CapEx. So they should predict additional factors. And so then empirically what we do and theoretically as well, we find that those two factors that we can construct from investor characteristics, age and wealth, so investor age and investor wealth, we find that these factors do a remarkable job by explaining not only the cross-section of portfolio holdings of investors, but also the cross-section of stock returns.

And again, for that, we’ll look at very detailed data from Norway this time, about 22 years of data, where we see the exact holdings of all the investors along with detailed characteristics about these investors. We have approximately 360,000 investors in a particular year. So that’s a lot of data and a lot of rich information to work with, but we find that if you construct three portfolios, one is the market, one is what we call the long-short portfolio that is long the portfolios of the wealthy investors and short the portfolios of the less wealthy, and a third factor is another long-short portfolio that is long the portfolio of more mature individuals and short the portfolios of the youngest individuals. Those three factors, which we call the markets, the wealth and the age factors, they will explain most of the systematic tilts in individuals and they’ll explain the majority of the cross-section of stock returns from the data.

Unreal. Our listeners are somewhat familiar just from us talking about things like the Fama-French five-factor model, which is based on firm characteristics as opposed to investor characteristics. How does the investor characteristics model perform in terms of explaining the cross-section of returns compared to something like the Fama-French five-factor model?

So they perform remarkably well. The first thing we can say is they explained really well the cross-section of investor portfolios. So first of all, we’ll get rid of some noise because again, there’s a lot of idiosyncratic tilting investors. So we’ll create groups of investors, groups of them sorted by education, sorted by age, by wealth, by where they live, by the occupation, the sector that they work in. So we’ll have about 93 aggregate portfolios of investors. And what we’ll find is that those three portfolios explain more than two thirds of the tilts of these portfolios away from markets. So that’s a lot. We say that portfolios have a strong factor structure that few factors are able to explain the common deviations from the market portfolio, the tilts.

And then we take those factors and we see, okay, how well do they price stocks? And to benchmark the performance, we’ll look at the CAPM model where you have market risk. And then we’ll look at a Fama-French model with the value and size. Then we’ll add momentum and profitability and investments. These are all standard, very well performing firm based factors. We’ll find that on average, the age and wealth factors deliver about 1% per month, which is large. It’s about 12% per year. Most of it is not explained by CAPM actually, it’s virtually all alpha. It is comparable in size to some of the profitability and investment factors from Fama-French, but it is not spanned by these factors, it is not explained by the firm-based factors.

So we run a battery of tests to see how well the firm-based factors explain the performance of the investor factors. And we find that that 1% alpha abnormal return per month shrinks down to about 66 basis points once you’ve controlled for these additional factors, but it’s still there and it’s still quite significant. So about 40% of that is explained by the firm-based factors and 60% of that is not.

Are the factors constructed using ex-post data or is it possible? I mean, I guess what I’m saying is are these portfolios investible? Can they be formed ex-ante?

Oh yeah, they are. And in fact, we make sure in our empirical tests to do what we called out of simple tests. So at time T, we’re looking at all the information from investor portfolios in the year before, we construct, we resort stocks. Instead of looking at let’s say the size of a stock or the book-to-market ratio of a stock, we’ll look at what we call the age of a stock, but it’s not the age of the firm itself. It’s the average age of the investor base of the stock. So if a stock has 60 years, it means on average, investors holding the stocks are 60 years old wealth weighted. Same thing for wealth. So we’ll sort stocks by their investor age, investor wealth, we’ll sort the stocks, we’ll create long-short portfolios, the same you would do for the Fama–French factors. You only look at information before and then you’re looking at performance afterwards. And on average, we find that the performance of the investor factors is either as good or slightly better than what the traditional firm-based models would predict.

That’s wild. I guess the constraint on actually building an investment strategy around this is data, right? Like you have good data in Norway, but in the US, it would presumably be much more difficult.

That’s right. That being said, we look in the paper for relationships between the investor-based factors and the firm-based factors to better understand without that data, what could we do to recreate those factors based on the firm characteristics? And we do find a number of linkages right there, which makes it possible to proxy what these investor-based factors would look like given the information we have about stocks in other countries.

What explains how investors tilt to the age and wealth factors?

We find that there’s a mix of risk-based and behavioral effects at play. Because we have so much information on investor characteristics, we can do a similar analysis as the one that we did in the value and growth paper, where we study how these characteristics predict a tilt that we have investors on these age and wealth factors. So we find similar ladders. We find that over the years, there’s a shift toward the age factor, toward the wealth factor as investors become older and wealthier. We create variables capturing risk exposure. So we look at an investor’s debt-to-income ratio, we’re looking at an investor’s exposure to macroeconomic risk through his or her labor income. We’re also going to look at a number of variables more behavioral, such as investor experience, how many years have you been investing in the stock markets? What is your level of education? What is your gender?

And we find that both factors play a role. We find that investors who are more exposed to macroeconomic risk have more depth to income tend to tilt away from the factors. The same way we saw before that investors more exposed to macroeconomic risk used to tilt away from the value factor, we see a similar pattern here. But we also see that on the behavioral side, the more experience you have, the more education you have, and if you are a female investor, you’re also going to tilt toward the age and the wealth factor, suggesting that there’s a mix of behavioral and risk-based effects at play in the data.

So interesting. Did you also find like equivalent to the value ladder how people are changing their exposure to the over the life cycle?

There is, there’s a big shift over the life cycle where investors progressively migrate toward the age and the wealth factor. Looking at the age factor, there’s about a 30% shift in the portfolio, comparing someone aged 30 to someone aged 65, 70. So it’s a significant shift in the composition of the portfolio. And again, it’s a fairly linear pattern that we see. We did not go as much in depth as in the value growth paper, where we looked at the rebalancing over the years, but the same effects are present here as well.

So from a firm characteristics perspective, what are the age and wealth factor portfolios actually look like?

So the stocks with high exposure to the age and the wealth factors, they tend to be more profitable, they’re larger. They tend to be more value types of stocks, higher book-to-market ratio. They have lower volatility, they have lower market volatility and they tend to have lower share turnover and greater institutional ownership.

Interesting. But do you have any idea why the age and wealth portfolios tilt toward large rather than small?

I think it’s easier to think the other way. Why are the small stocks on the other side on the short leg of these factors in a way? So if I were to flip around the discussion, what we find is that stocks held by the younger and the less wealthy investors, they’re smaller, they tend to have significantly higher volatility, higher share turnover, lower institutional ownership than the stocks held by the more mature and wealthier investors. What we find is the results are consistent with, say, prior work that argues that these stocks perhaps they’re more difficult to arbitrage and so they’re more sensitive to changes in sentiment in the data. And so younger, less wealthy investors may be pushing the demand for these stocks, other investors may have a harder time arbitraging some of these stocks away. And so the result is that sentiment or behavioral effects may be contributing to the pricing of the age and wealth factors through that channel.

Wow. Fascinating. Now when you think through this, we have older and wealthier households earning higher expected returns and younger, less wealthy households earning lower expected returns, do you think there’s an impact on wealth inequality from this effect?

I believe so. The wealthier we are, usually the more able we are to take on risk. What we find here in this paper is this is not just taking more risk across the board. This is loading on more risk factors, in particular, there’s a systematic deviation. So we’re taking on that so away from the market portfolio, taking on these types of stocks. But the result is, well, the more risk we take as more mature wealth investors while over time, the higher the return we’re going to get on our portfolio on average. And so that’s if anything contributes to inequality because those willing to bear the risk over time end up doing better, on average, of course. If you have a recession, they may be hurt the most.

My thought is, does that mean younger investors can combat wealth inequality by taking on more risk? But of course, they’re taking on more risk by doing that. So it’s not like it’s a solution to the problem.

So again, if it makes sense for them to bear the risk depending on their personal situation, it may make sense to load on the risk factors. But if you are that investor and your job on the side is sensitive and you have a mortgage and you have all the kinds of priorities and you may not be able to afford that risk, you would love that higher return, but you may not be able to withstand it in those current conditions.

Right. We spoke with Morgan House, who wrote the book The Psychology of Money. And I think one of his rules in the book is that you have to stay in the game, get knocked out of the game by losing everything, then, well, you don’t get during the higher returns higher expected returns.

Precisely. And the worst that can happen to you is to sell it all at the trough of a crisis, which is oftentimes what happens when you have investors panicking, but this is guaranteeing the loss moving forward and not being able to rebound.

So let’s shift a bit to home ownership and obviously home owners have a huge portion of their wealth in their housing. So from a theoretical standpoint, what are the possible reasons for that?

Some reasons have to do with what I would call the intrinsic value of housing as an investment assets. It’s a good that allows you to borrow, it’s a good collateral if you would like to borrow and invest heavily, it’s a good that provides a commitment device to save because we end up having to save for the house, we have to pay back the mortgage. And so it forces us to be disciplined about saving in a way. And I would say perhaps most, it’s a good that has its own tax benefits. If I own a house I live in that home, I am not taxed on the dividend of the home since I live in that home itself. I implicitly pay the rent to myself. You have interest tax deduction on the mortgage as well. So you have a number of tax incentives.

But I would say perhaps most importantly, it’s an asset that provides a fairly stable set of cash flows that protects you against, I would say, a rise in the future price of rents. If you live in a house, you’re a homeowner, you get to live in that home for as long as you want to. You’ve paid the mortgage. The neighborhood that you live in may become wealthier, prices go up, rents would go up, but you are in that home and it’s years. And so it allows you to stay in that good consume that house for long periods of time, in a way, it looks like a bond that is indexed to your housing consumption that provides cash flows in real terms that are stable over time.

So I would say these are all factors that contribute to the strong intrinsic value of housing. But then on the other side, you also have frictions in the housing market and the frictions can also explain potentially why you have so much of your wealth invested in housing. If you think about the housing markets, you can rent, you can own, you can be a landlord if you want. So you own a house that you rent out to someone else. What you cannot easily do is own one piece of your home and rent out the remaining fraction from someone else, another co-owner. So you cannot go and say, you know what? I love that house. Let me buy 50% of it and will pay the rent to myself on that one, I’m the owner. And then I’ll find another co-investor and I’ll pay rent on half of the total value of the house. And so that allows me to decrease my total investment, yet consume the full house. We cannot do that, right? We have a friction. There’s a housing market preventing us from this partial ownership types of arrangements.

The result being that if you want to be more on the ownership side and the rental side, well, perhaps you just have to own up the full value of the house that you would like to live in. And that could potentially explain why we tend to have two thirds of our wealth, financial wealth, putting human capital aside, two thirds of our wealth invested in one home.

So you looked at this, of course, which is why we’re asking you about it. As with all the other questions, you looked at this in a recent working paper, why do homeowners invest the bulk of their wealth in their home? Which of those theories that you just described is supported by the empirical evidence?

So what we find is even though the frictions in the housing markets, they are large, they do matter quite a bit. The bulk of our investment in housing mainly has to do with the intrinsic value of a home. Said differently, suppose that we immediately could live in a different world where we were allowed to sell a fraction of our home, we may change homes, we may decide to enter those partial ownership arrangements, but in the end, after having done all that, we would still own the majority of our wealth in one good housing. And part of that has to do with the long run safe value of the housing good that allows you to live in that one house over long periods of time and not having to worry about changes in the value of the rental amount in that house moving forward.

Are certain household characteristics predictive of a larger allocation to housing?

Yes. Several predictors will drive our consumption value for housing, right? The size of the house that we would like to purchase, think of a large family, think of having a long investment horizon. I am going to stay there for a long time in that particular neighborhood. This could be the case, for example, if I work in a licensed industry where my job is given in Quebec, but in order to move to another province, I have to pass another license. So I’m perhaps likely going to stay in that one job in the same province for a long time. And so that brings stability to my human capital. And as a result, it lengthens my investment horizon and makes it more appealing to purchase a home today.

Risk aversion also can play a role. We tend to think of housing as a risky asset, and it is in many ways when you see the price of housing going up and down, but ironically, the more risk averse we are, the more likely we’re going to buy a house before we even touch stocks. What we find in general is that more risk averse people will tend to have a higher housing share. What we call is a housing share is the share of housing inside the financial portfolio. And again, that is partly because housing plays a role partly as a risk-free asset as an indexed good on that housing consumption, which is for most of us 25 to 30% of our consumption expenditures in a given year.

Unreal. So it’s like, we’re in a position of giving financial advice to people. The housing decision is not, I mean, it’s kind of like in portfolio management, you can’t just think about mean-variance optimization because that doesn’t make sense for everybody for all the reasons we talked about previously. And it sounds like it’s similar for housing, where you can’t just say, well, no, look, you’re going to have more wealth if you rent and invest in small cap value stocks, it’s got to do as much with the hedging demand.

Yes. If you think of housing in that long term mindset, you can think of it in two terms, you have the component of housing in that mean-variance portfolio that you just mentioned. If you and I imagine that the price in our neighborhood is going to go up, we have the cash, we know something about this neighborhood that may go up in value, yeah, it might make sense to speculate on that one good because it’s a good investment to have in the portfolio. So that comes in and that plays a key role.

But also perhaps even more importantly, especially for the more risk averse individuals, housing provides a hedge, a long term hedge. And in fact, what you can show in very simple long term model of our consumption over the life cycle, as you take an individual who is infinitely risk-averse refuses to take any risk, that portfolio is not a completely 100% bond portfolio. That portfolio is going to be, roughly speaking, 30, 40% housing. The remainder will be some indexed bonds, some tips on the rest of our consumption. And when I’m talking about 30, 40% and 50%, I am not talking about our financial wealth. I’m talking about our total wealth with human capital, right? And so you take that into account, that can explain potentially why housing is such a large role in our portfolio. It gives us access to a long term bonds on that one particular good.

Yeah. So such an interesting perspective. Why are the risk-free benefits of owned housing, like you were just saying, why are those economically important?

Because housing plays a key role in our consumption baskets. Already, it is about anywhere ranging typically from 20 to 30%. So that’s one third of our consumption expenditures. And then it affects many of our other consumption expenditures. It affects whether I need a car, it affects whether I want to use public transportation, it affects which schools, even public, private school I may want to send my kids to, it affects where I may want to even work depending on the set of options available. For all these reasons, it ends up being one of the core consumption goods that we have that influence the rest. And so being able to provide security on that when good tends to be key.

And again, I insist on that because we tend immediately to think when we’re looking at changes in the price of a house to think of it as a risky, and it is partly a risky investment, but so is a bond. I mean, if we looked at the price of treasuries, price would fluctuate too. But we tend to think of a bond in terms of the cash flows that it gives us, giving us safe cash flows in particular federal government bonds. Well, same thing here, you think of a house, you think of a very steady state of cash flows, it allows you to live in a particular home in a particular neighborhood over a long period of time.

I want to keep going on that. So if owning a home is like having a perpetual bond that’s indexed to the housing consumption need. So you’ve got cash flows coming from the house that match your housing consumption needs, that’s the cashflow flow perspective. But like you were just saying, the asset perspective, that the price of the home can fluctuate wildly as a perpetuity or a bond or a long term bond can also fluctuate, what happens if the homeowner needs to sell that bond, the house by moving? Like what if they change jobs or whatever else, have to care for a family member or something, does that make it more risky?

Yes, it does. When I teach finance and I teach bonds investments, we talked about short term, long term bonds, we talk about the fact that local long term 30 year bonds might be safe if your investment horizon is 30 years, because you’re not going to touch it for 30 years. And in 30 years, you’re going to receive the bulk of the bond on the face value, especially a zero-coupon bond. Perhaps a better example would be a consol bonds that give you cash flows every single period, but think of a long term bond for now. Even though a long term bond is very safe over 30 years, it is not safe at all over one year, right? Because when you sell the bond over one year, you’re selling a 29 year bond. And the value of that bond is going to depend on whatever interest rates will take place on that day. And so you might sell the bond at a premium at a discount. There’s quite a bit of volatility over the very short term.

Well, the same logic applies here. If I think of housing as a long term bond index to housing consumption, makes sense if I have a long investment horizon and I cash out, I use the cash flows to live in that house. But if I need to sell the house over the coming year, who knows what the price of that house will be. And at that point, I face what we call price risk. And the price risk could be significant, just like it is for long term bonds.

So a rule of thumb that we like is that if you don’t plan on staying in the home for at least a decade, buying a home is a risky proposition. Does that make sense to you?

It does. Because as we just discussed, we have a price risk that becomes particularly important over a shorter horizon. And then adding on to that the fact that the house is not a very liquid product, it’s less liquid than a bond would be, the result being that every time you trade, you have a heft transaction costs. And for these reasons, unless you have reason to believe that the value of the house may be going up significantly in a very short run, you may want to speculate on that, but as you said, it becomes more of a risky asset in your portfolio.

So interesting. You mentioned people with riskier labor income have more housing consumption, they want to own more housing. You said that, right?

I would say people in a licensed profession would have a greater need for housing in a particular area.

In a licensed profession.

Yeah. I think what you would find is people in, and this is not my study, but I’ve seen other studies in the field by colleagues at UBC actually in Canada, that individuals with riskier labor incomes, if anything, would take on less aggressive positions on housing, because oftentimes they have a mortgage on the line and you want to avoid a situation where at any point in time, you’re not able to pay back the mortgage.

Interesting. Okay. That’s where I was going with that question, because someone with riskier labor income, they might want to hedge their housing consumption, but they’re also taking on a whole lot of risk by taking on debt or potentially having to move geographically or something like that.

That’s right. Now when we are talking about moving geographically, we have to be careful because if I move in the same neighborhood, sure I’m going to face price risk, but I’m going to buy another home that is very likely going to move in synchronization with my own home. And so that mitigates the risk. The risk really is when I’m moving somewhere else, where you have a housing market that is not connected to the one that I’m in right now, at which case the price risk becomes more important.

Yeah. That’s something I’d love to get dead on. There’s an old paper from Robert Shiller, where he looked at the volatility of real estate indexes compared to the volatility of individual real estate properties. And the individual properties were dramatically more volatile in price than the aggregate index.

They are. We estimates in that one study that’s the volatility of a house. If you take the house price year to year, the volatility is about 10% per year. Whereas if you take an index, it’s going to be more in the order of two, 3%.

Oh, I missed that in the paper. I’ll have to go back and find that. That’s awesome to have that data point, because we’ve been thinking about that from a financial planning perspective, like what do we use as the volatility on the owned residents, because often in financial planning, related to what we’ve just been talking about in financial planning oftentimes, people will include the value of their home asset in their financial plan with the intention of downsizing or potentially moving elsewhere. And if we include that with a 3% volatility, it’s going to tend to look very good in a financial plan, but much less so if we assume a 10% volatility.

Yes, I agree entirely

Interesting. Okay. We have one more segment based on one of your papers. I’m quite excited for this one. I was excited for all of them, but this one too. So CPP investments and maybe pension funds in general, I’ll even generalize in Canada, get a bit of a hard time in the media because they’ve got internal costs that are pretty high. I mean, Canada Pension Plan, I think they’re running around 1% on a massive dollar amount, when of course, they could be paying whatever, four bases points to own low cost index funds. Now is asset only performance the right way to measure the success of the Canadian Pension Fund model?

I would say no. It is important determinant of performance, especially compared to benchmarks that may be cheaper to invest in, but you also want to take into account the risk management aspect of the job, which is just as important. When you take a fund like CPP and many of the others, they’re going to invest large amount of their capital, not just in stocks and bonds, but in infrastructure, in real assets, in private markets, private equity, private debt. The portfolios of these funds are much more diversified than the reference portfolios that we’re talking about, which typically only have public equities and bonds.

And so we have to be cautious there because when you’re not comparing the same apples to apples, they’ll be years when public markets do really well, at which point, they’ve may outperform the rest. The same way if I invest my entire wealth in one stock, there’ll be some years where the stock did really well, but there’s also going to be other years where the public market is worse.

And so what’s important is not just the performance of the overall portfolio of a CPP versus the bond stock benchmark. What is important is over the long term, the risk-adjusted performance of the portfolio, how well does it do in terms of overall return generated, but also in terms of the risk, the volatility of these returns, the probability of getting into an underfunding status than the low cost reference portfolio.

And so I’ve done research on this topic, studied the Canadian Pension Fund model in particular, we’re looking at data from CEM Benchmarking, it’s a global benchmarking firm based in Toronto, they collect data on hundreds of pension funds in Canada, around the world. I mean, you have US, Australia and Netherlands, all kinds of UK, other countries, you have multiple countries. And we looked at the performance of the Canadian funds net of fees along multiple metrics, including not only the return performance itself relative to a benchmark, but also the overall return compounded over 20 years, the volatility of that return, the Sharpe ratio, right? So the risk-return trade-off of the entire portfolio.

And what we found was quite remarkable actually, we found that net of fees Canadian funds not only perform better than their global peers, but on top of that, they tend to hedge more of the liability risks than their peers. And that should come almost as a surprise because typically the more you hedge, the more you give up on returns, right? But here we find that they tend to do better, perform better in net of fees and have assets that are more aligned with the structure of their liabilities. And so I would argue that these metrics would be more appropriate to assess the performance of large pension funds like CPP and others.

I realize as you’re talking that my question probably had a term that even some of our listeners who tend to be more technical may not be familiar with. Can you just briefly touch on the difference between asset only performance measurement and liability relative or liability hedging performance measurement?

Yeah, it’s a good question. So when we’re thinking of the portfolio, let’s say you have a pension fund with a $100 billion of assets, right? These are the assets. And so I can look at, okay, what is my expected return? What is the volatility on that? But then we have to think that these funds have to pay liabilities to their pensioners. These liabilities typically are long term streams of pensions. And so in a way, these funds are short. The liability portfolio, they’re responsible for paying out those liabilities. And oftentimes those liabilities look like bonds. When I owe a pension, it tends to a fixed amount that is predetermined year after year. Some of it is indexed to inflation.

And so my job as a pension fund is to make sure that the assets that I generate are sufficient to pay for these liabilities and to make sure that I can do it in a way that is sustainable, because you could imagine that I could invest everything in bonds to match completely the risk profile of my liabilities. But if I invest all of my pension in bonds that yield only 1% or 2%, I’m going to have to invest a lot more today, contribute a lot more in order to get the system to work. And that may not be efficient in the long run given what we know about compound returns from these investments across asset classes.

You briefly mentioned earlier some of the asset classes. So like there’s infrastructure real estate and maybe you can remind us again what the other ones were. Why are those assets that Canadian Pension Funds are using better at hedging real liabilities than stocks?

Or bonds. When you’re looking at many of the European systems or many of the corporate systems with very, I would say, strict regulatory requirements, making sure that the value of the assets is always sufficient to pay for the value of the liabilities, which is itself a bond-like portfolio, many of these funds will invest mostly or the majority of their assets in bonds, long term bonds. Some bonds are nominal, some bonds are real bonds to give you that exposure to inflation. The problem with that strategy, as safe as it is, it’s a very costly strategy because when bonds yield 1% or 50 pips or in some countries, negative yields, you’re going to have to invest a lot of capital upfront to make sure that you can pay for the total amount of pensions that have been promised to the plan members.

So then the question becomes, and this is what the question is, how else can I hedge the liabilities that in particular for many of the Canadian Pension Funds, many of the liabilities are real, they’re index to inflation? What we find is the Canadian funds end up investing a lot more in real assets as their global peers. And by real assets, I mean REITs, real estate investment trusts that are traded on public exchanges, but also private real estate, residential, commercial, office, and so forth, distribution centers, warehouses. And then infrastructure. You find that many of the ports, seaports, airports around the world tend to be owned by these large funds. All of these assets tend to generate cash flows that have similar properties as long term bonds. They tend to generate cash flows that are fairly stable over time and that are indexed to inflation. Some of the private real estate investments are a good example.

And so what we find, and this is fascinating, we find that for the large Canadian Pension Funds, they have about 18% of their assets under management invested in infrastructure private real estate. This is twice as much as what some of their global peers of the same size will invest in that market. And what we find is that a lot of the hedging, the alignments between the assets and the liabilities will come from these types of investments.

And my question to link to that is the well-known Norwegian model, the very large government pension fund there is much more similar to a simple low cost passive index fund. So why have Canadian Pension Funds taken such a different approach than that model?

So you could indeed have more of a passive approach, which will be low on fees, at least in the public markets and give you decent risk return trade-off overall. The Canadian Pension Fund model was developed, I would say, mainly in the 1980s. And the idea was to be more efficient, more long term and create value with these pension assets by running the pension fund more like a private corporation, which might come as a surprise because you wouldn’t think initially of let’s say the Ontario Teachers’ Pension Plan, a very public plan as running its pension plan like a private business. But in fact, what the business model has allowed to do is to save on costs for a number of reasons.

When you’re looking at the business model of many of these funds, a lot of that is in-house. When we’re speaking about these large pension plans managing more than $50 billion of assets under management, about 50 to 70 to 80% of the assets managed are in-house, managed internally by the funds as opposed to being externalized. We find that this reduces fees, costs about one third of the cost by having the management done in-house.

The savings from these investments, we find have been redeployed in two ways. They’ve been redeployed into having more sophisticated investment teams within each asset class, pursuing more sophisticated investment strategies trying to achieve alpha beat the benchmark net of fees. The other redeployment has been to go more into the private asset classes, such as infrastructure, such as real estate, which are hard to do from small investors, but allow the funds to generate value, and again, have assets that are aligned with the liabilities that the funds are facing.

And so the business model that has been developed around the 1980s moving all the way to today has generated return from six to 8% per year and has generated value added relative to the classic benchmarks net of fees, right? So the fees about 1% or 60 pips on average, we find a large pension plans typically cost about 65 basis points scaled by assets actually have generated value net of the benchmarks by about 50 pips to 1% net of the benchmark net of fees, right? So the fees have been worth it in the sense that even afterwards, the return generated by these funds has exceeded what you could have achieved using the classic benchmarks or the classic low cost indices for smaller investors.

I haven’t looked at the Norway financial statement in a while, but I think last time I looked, it was much less expensive than the Canadian model. Do you think Norway, should they be taking a lesson from Canada and saying, hey, maybe we should be increasing our costs and doing more of the private investments?

So we do find that in some domains such as in infrastructure and real estate, we find in Norwegian model has a lot of similarities to the Canadian model. In the study that we run, we ask the question, is there room for other investors to emulate this model? The answer is yes, but you need to have a strong governance structure allowing you to bring your assets in-house and being creative and entrepreneurial with your investments. For many of the very classically run pension funds, the governance structure may not be there. And so without the proper governance structure, it’s going to be hard to achieve.

But if you do achieve, if you do create a robust, independent governance structures that allow these funds to run in a way similar to private business models, you can generate very cost efficient vehicles that may ultimately allow pensioners from the public sector, individuals who may not be savvy with their finances to benefit from professional and high quality talent. Some of my very best students work at these pension funds. And I would personally like to have some of my assets managed by them. It’s actually a huge gain for the population to have talents managing their assets in the current conditions.

Interesting, which is such a different perspective than what you read in the media that the costs are too high and the returns are barely better than the index.

It is indeed, but it is difficult to make sense of these results from what we read in the media, because we need to ask ourselves, what has the performance looked like net of fees over the long term relative to the risk? And how would that have been comparable to what standard benchmarks or low cost investments would’ve given you? And once you do that comparison, you get a different picture.

Yeah. And it’s relative to the risk of the liabilities, because I think people hear relative to risk and think standard deviation, but it’s relative to the risk of the liabilities that the pension fund has, which is you can’t look at their return number and say whether it’s good or bad without knowing what their liabilities.

And you cannot look at the realized return in one year, they will be years when the fund outperforms the benchmark, it’ll be years when the fund outperforms the benchmark. This is part of the risk that we take, but we should be judged over long term performance metrics, not just the realized performance in a given year relative to what 60/40 portfolio in stocks and bonds would give you.

You alluded to this briefly, but I want to ask about it explicitly. Are there any lessons in what the Canadian Pension Fund model looks like for individual investors managing their retirement accounts or maybe even for firms like PWL that are managing retirement accounts on behalf of individuals?

They are. But I think in your case, you already do a lot of them quite well. The model itself, the Canadian Pension Fund model is harder to replicate for small investors. We cannot go and invest in airports, right? I mean, you have to have the funds and the ability to invest those large amounts and take on that liquidity risk that small investors cannot do. So there’s a few things that individuals cannot do. And in that sense, going forward, low cost ETF based portfolio makes a lot of sense.

But I would say there’s a number of lessons to draw from the model. The first one is that patient capital and having a diversified portfolio really, really helps over the long run. We may not appreciate it from one year to the next, but over the long term, about 80% of the pensions of the individual from these plans come from the compound returns, not from the contributions. And that has to do with disciplined investing. We have a very diversified portfolio across asset classes and the ability to be patient, not to overreact when you have a crisis, but going through the ways and really thinking long run.

Another related point is that discipline savings make a huge difference. So one of the reasons these plan do well is you have commitments, right? Employers put in five, 10% of your income into the plan, you put five, 10%. You don’t even see it because it’s withdrawn from the source. That means that there’s an automatic savings going from the earliest stage. This is not something that we see when we’re dealing with defined contribution plans and many investors are left on their own and the savings will typically start a lot later less disciplined. And the impact of compound return over the long term is not extracted to its full value.

The third key component is the ability to hedge against longevity risk. These plans pool the assets, right? That gives them an opportunity to hedge against our own longevity risk, where those pensioners receive a pension forever as long as they live, could be up to 120, 130, the pension will be there was when we’re on our own, we tend to be fully responsible for that longevity risk. The longer we live in good conditions, the better, but we need to be able to afford that long retirement. And we have increasingly through CPP and QPP, we have the ability to extend and to provide that fairly low cost longevity risk. It’s important to maximize that annuity in a way that we can get potentially get more on our own if we need to, not the full portfolio, because that can be quite expensive, but we think carefully about how much we can hedge of our own longevity given the tools available.

That was a great answer. So our final question for you, Sebastien, how do you define success in your life?

Making the most of the opportunities when they come to you. It’s taking risks, good risks and having good opportunities at all levels, not shying away from opportunities at all levels when they’re there and making sure to maintain a balanced lifestyle that allows you to have a successful career, but also a successful family life at the same time because it is so easy to sometimes forget one and regret it later on. And so maintaining that balanced lifestyle, yet the same time being open to taking opportunities whenever they come, not shying away from the risk is two key features.

Great answer. Terrific answer. Sebastien, this has been a real pleasure to meet you and hear your thoughts and all this. So thanks so much for your time.

Thank you for having me. It’s been a pleasure interacting with you both, Cameron and Benjamin.

About The Author
Cameron Passmore
Cameron Passmore

Cameron Passmore has been a leading advocate for evidence-based, systemic investing for over 20 years in the Ottawa area. Today, Cameron and his team serve a broad range of affluent clients across Canada.

Benjamin Felix
Benjamin Felix

Benjamin is co-host of the Rational Reminder Podcast and the host of a popular YouTube series.

Meet With Us

Meet With Us Flyout Form
What are you looking for?