I continue to read the 2018 edition of Reducing The Risk Of Black Swans by Swedroe and Grogan, and I'm thoroughly enjoying the intellectual exercise.
Today, I'll share my latest A-ha moment.
When designing a portfolio, we often use the terms risk tolerance or risk aversion. The caveat to using these terms is that investors don't have a terrific way of accurately quantifying such concepts, and attempts to do so often fail miserably when an actual recession or major correction occurs.
Benjamin Graham, a Columbia University business professor and Warren Buffett's mentor, famously endorsed a maximum exposure of 75% stocks based on his experience with investor risk aversion.
I wonder if perhaps the problem is with the term. Investors are not risk averse so much as they are loss averse. The literature of behavioral economics, brilliantly investigated by Kahneman and Tversky, demonstrates that it pains humans more to lose a buck than to gain one. The logical implication is that investors are more likely to stick with an equity allocation as long as they can reduce their down side risk.
Here's where Swedroe and Grogan re-enter the discussion, suggesting you can reduce portfolio volatility (and thus down side risk) without reducing expected returns by making two interesting, counter-intuitive moves:
- Reduce your overall allocation to stocks.
- Alter your stocks to tilt heavily in favor of the Fama and French factors, small size and value, so that the portion of stocks you hold have an increase in expected return.
Okay, your mind is reeling. Mine is too. Let's review some pertinent fundamentals before we flesh this out a little more.
Great Moments In Finance
Following is an abbreviated primer of influential moments in understanding investment theory. I'll favor brevity at the expense of detail, so please forgive the lack of nuance:
- 1952: Harry Markowitz publishes a paper introducing modern portfolio theory, where he shows that an individual stock's risk and expected return is not important; rather, it's the contribution it makes to the portfolio as a whole. He gives birth to the idea that adding uncorrelated assets to a portfolio can optimize return while minimizing risk - also know as the efficient frontier. He later wins a Nobel prize for this concept.
- Capital Asset Pricing Model: defines the relationship between risk and return through a single factor, market beta. Market beta measures volatility relative to the market. Thus stocks have high market beta (their values are more volatile than the market as a whole), while T-bills have low market beta.
- Over a long time horizon your expected return can be explained by the proportion of stocks to bonds (high market beta to low market beta investments) you hold.
- 1992: Fama and French introduce a three factor model, demonstrating that stock returns can be explained not just by a stock's market beta but also by its size (stocks with smaller capitalization, i.e smaller companies) and value (stocks with low prices relative to book value, i.e selling at a discount). Size and value are independent risk factors, and provide exposure to different risk than market beta does. Thus size, value, and market beta are poorly correlated over long time horizons. This is believed to be because small and value companies carry greater risk, and the market rewards that increased risk with greater reward.
- Over a long time horizon your expected return can be explained not only by the proportion of stocks to bonds (high market beta to low market beta investments) you hold, but also by any increased exposure to value or small stocks which account for an independent additional risk premium.
Enter the Larry Portfolio, so dubbed in a 2011 NY Times article based on Larry Swedroe's unorthodox investing strategy: reduce the proportion of stocks you own, but increase the risk (and commensurately increase the expected return) to preserve the expected returns you project will help you meet you investing goals.
To understand what the Larry portfolio accomplishes, think of the bell-shaped curve that encompasses statistical distributions on either side of the arithmetic. When you invest, you are creating your own bell curve around your expected return. The potential range of outcomes dispersed on either side of the mean is represented by the standard deviation, a frequently used measure of investment volatility.
Bonds have a narrow dispersal range, corresponding to a low standard deviation. An all bond bell curve is narrow and tall.
Stocks, in contrast, have a wide dispersal range, corresponding to a high standard deviation. An all stock bell curve is broad and short.
What the Larry portfolio does is modify the shape of the bell curve to specifically narrow the dispersal (reduce the standard deviation) of the overall portfolio. Even though the Larry portfolio includes some highly volatile components (small value stocks), it contains them in small enough quantities that the overall bell curve has the narrow dispersal range more characteristic of bonds while possessing the increased expected returns characteristic of stocks.
To illustrate the principal, let's make some reasonable mathematical assumptions based on historic returns to use in assembling your theoretical portfolio. (It should go without saying that past performance does not guarantee of future results.)
- Total stocks have a 7% expected return
- Small value stocks have an 11% expected return
- Bonds have a 5% expected return
Let's also assume you are an investor who, after factoring in your annual contributions and expected age of retirement, needs to achieve 6.5% returns from your investment portfolio as a whole in order to retire at a spending level that will support your lifestyle.
A 75/25 stock/bond portfolio has an expected return of (75%*7%)+(25%*5%) = 6.5%. The price of an expected return meeting our stated goal is significant volatility from the high stock allocation. The bell curve is broad and short. Considering that stocks may lose 40-50% of their value during a severe bear market, many investors will not tolerate such high volatility and will seek a more conservative allocation.
Compare that to a less risky 50/50 stock/bond portfolio which has an expected return of (50%*7%)+(50%*5%) = 6%. The bell curve is narrow and tall. The reduced volatility of this portfolio means you'll fall short of your goal.
Let's look at what happens if instead you decide on a 50/50 stock/bond portfolio where stocks are split evenly among a total stock market index fund and a small value stock index fund. In this case the expected return is (25%*7%)+(25%*11%)+(50%*5%) = 7%. You've juiced up your expected return while maintaining the lower volatility that comes with a 50/50 portfolio. You now exceed your target, and the bell curve is narrower and taller than it would be with an all stock portfolio, but broader and shorter than it would be with an all bond portfolio. We've created a weird and wonderful hybrid with desirable attributes of both curves.
Finally, let's reduce your risk even further by choosing a 40/60 stock/bond portfolio, again splitting stocks evenly among a total stock market index fund and a small value stock index fund. This yields (20%*7%)+(20%*11%)+(50%*5%) = 6.6%. Still meets our target, with the benefit that volatility is further reduced (this bell curve is narrower and taller than the 50/50 portfolio above).
Reduced volatility in a portfolio means that investors who struggle with loss aversion (i.e., all of us) will feel less pain and thus be less likely to sell stocks in a downturn - the classic investing mistake that undermines even the best thought out investment policy statements.
The gift of reducing volatility while maintaining higher expected returns needed to meet retirement investing goals is the gift of the Larry portfolio. Psychologically, reducing down side risk makes you likelier to stick with your plan.
In the end, the Larry portfolio is about playing a mind game to induce you to act in a manner aligned with your interests. Kind of a neat party trick!
If you are navigating a personal or professional crossroads and seek assistance, I'd be grateful if you'd consider my burnout coaching service. Thank you.
Comments 10
Dork that I am, I still think the “Take on Me” video is pretty cool. I predict this song will be stuck in my head today 😁
Author
Mrs. T,
Dork that I am, I not only share your opinion but can still name many of the forgotten (and forgettable) tracks on A-Ha’s debut album (“Looking for the Whales” – a truly Norwegian song title if ever there was one), and am in the minority of human beings to have liked them enough to have bought their second cassette tape (cassettes were state of the art technology at the time).
I’m glad to give you something fun to hum today, Mrs. T – when I placed that link, I thought to myself that you in particular would appreciate it.
Thanks for being my dependable 80s nostalgia sap. Got a John Cusack reference coming up, keep your eyes peeled…
With affection,
CD
LOL-too funny! Two of my vehicles still have tape decks, which is my defense when asked about my archaic cassette collection. (Lest you consider me a complete Luddite, I do also own one of the first generation iPod’s.) Will be keeping my eyes open for that John Cusack reference ☺️
Pretty much I feel like I’m living in a cartoon.
Good article except the market really doesn’t trade on price anymore. It trades on levered flow. In Graham’s day price was related to value. If I held a property say a company XYZ and I tried to sell it for 300 and it didn’t move. I lowered to 250, it didn’t move. I lowered to 200 and it sold. The value of the property XYZ is 200 because that is where it became liquid. I own the company and buy back shares with cheap leverage. Now my company is liquid at 225 plus some debt. Price went up but value didn’t. Add some more leverage and buy back more shares stock now “sells” for 250. Same value more price + debt . I’m a CEO and get paid in options when the option exercises I make more with a 250 price than a 200 price so I’m induced to lever up the price. The value of the company is masked by the leverage and price no longer discovers when liquidity happens aka value
You don’t own the company directly but through an index funds passively You send SPY $5000 and spy buys some of the above at any price. And $5000 buys one share of XYZ next month the stock is levered up again and 5K buys .99 shares. In 30 years you own shares which are appreciated in price but not in value and your portfolio consists of buy high to start, buy higher each month. What is your portfolio worth if price no longer discovers value but only discovers momentum? It’s worth what you can sell it for except you also own all that non productive deb. So you bought high and you bought high on leverage and didn’t even know it because (here comes the mantra) “investing every month in low cost mutual index funds is the ticket”
So what is your risk? Your risk is you think your stock is worth 200 and plan your future based on 200, but it’s worth 50 because of the leverage and especially if the market doesn’t recover from a downturn with a brisk and sustained upward flow. Once again your fortune is momentum based not value based. Prez Prezzy wants to get re-elected so he floods the market with leverage blowing up the price with no or reduced increase in value. What is your risk? Booooomers retire and instead of buying every month they sell every month forcing the market to lower highs. A virus comes and knocks hell out of employment putting half the restaurants and vaca industries out of business. What’s your risk since price is no longer sensitive to value only leverage and momentum? Previously for 40 years the bond and the stock were uncorrellated so if stocks crash bonds stay stable, but now the uncorrelation has become correlated and if stocks drop, bonds drop as well. What is your risk? The government starts printing money and sending people checks. Inflation catches and a poorly growing economy is coupled with high inflation. What’s your risk? The bogglehead retort is past performance “will” mimic the future because that is the bogglehead bet. The bet is markets always go up. What if they don’t? The Larry portfolio is 70% US bonds. In Switzerland their bond is yielding -0.5% meaning they charge you half a percent for owning the bond. A 1M bond would loose 5K per year PLUS inflation. What could possibly go wrong?
Author
Have a guest post you will appreciate coming up next Monday that supports precisely what you describe. Friend from the hedge fund world feeling that price discovery and value have come completely untethered, and that passive investors are a part of the problem.
Do you know of any companies that have decided to remove the CEO incentive to continue acquiring debt by severing the connection between executive bonus and share price?
I believe the crash had that effect to some extent, also I think of you connected to the fed money firehose you couldn’t go back to the shineup. Part of the reason volatility across asset classes went into outer space (VIX went to 85, OVX went to like 300, even Gold went to 55) was to de-lever. When you’re accelerating into the ground you sell everything. That event has yet to play out in terms of understanding who hit and who bounced. Many companies that still seem alive are dead. In one of Ford’s conference calls in the past couple months it was stated “there is no way forward” or something to that extent. Imagine a time with no Ford. A time with 50M unemployed may be a time that time. Tesla has a bigger market cap than all the other car makers together yet we don’t have anything like the KWH infrastructure needed to support a plugged in car economy. When you fall the ride down is exhilarating. It’s in the microsecond when your face impacts the ground that the party is over. So the answer is nobody even knows what companies are still alive enough to play financial engineering games any more. VIX is a log function meaning when VIX is small variations are quasi linear. It’s like Rt. When cases are small the population hardly feels the impact, but when cases are big compounding drives home the point in a non linear way. People think in linearly not log-rhythmically. Pretty much the jury is out, but you can bet every sucker is grabbing every buck out of the till on the way into the ground.
Great news about your guest post CD. You’re one of the few willing to explore the narrative.
I had the pleasure of meeting Larry and talking with him about this. NN Taleb has also written about his “barbell” version of this. He and I exchanged emails many years ago. I have had more “low risk” investments and “high risk” investments in my portfolio since then.
The gist of what I remember is having a component of higher-risk assets (such as small value (Larry) or private placement/options (NNT)) boosts your returns. It may even have less portfolio volatility.
The challenges are primarily psychological. We tend to use “mental accounting” and put our high-risk investments in a different mental filing area. We don’t feel good when it goes down. We question our judgment. Worse, we may act on that and panic at the worst time. Been there, done that. More than once!
So if you can think of your portfolio as a whole and also not be upset by “tracking error” it is a great investment option. Many of us feel foolish when the “market” is doing well and we are not. The numbers are great if your mind and behavior can comply.
Author
Wealthy Doc,
You have been in the business so long you are part and parcel with a nexus of the finance writers I respect and admire – no small feat. Part of what keeps me going when my well runs a little dry is the opportunity to keep excellent company and use my JV blog as a pretext to meet people who are truly interesting, as your correspondence and conversation demonstrates.
I admire the Larry Portfolio for basically tweaking the portfolio to match the frailty of human psychology – it addresses our weakness and builds our failings into the plan. That’s some impressive insight.
Thanks for dropping by, and for always being a generous supporter of my writing (in spite of its weaknesses).
With affection,
CD
I love Larry Swedroe and I totally agree with this approach. I also have a strong small-cap value and micro/small-cap tilt and with a significant allocation to bonds I can backtest that portfolio and beat a 100% VTSAX portfolio EVEN with the bonds. Lower standard deviation and equal or higher expected future returns is an awesome thing.
Author
Dan,
I’d completely agree that the intellectual basis for the portfolio is sound (and scores extra points for creativity!). I’ll look forward to updates to hear how your portfolio fares over the long run – appreciate your letting the rest of us follow along on your journey.
Fondly,
CD