It was ‘convexity week’ on TILS.
After this tweet, a lengthy and insightful debate started on what convexity means.
At the same time, PortfolioCharts posted a link to this Italian blog where the author explains…convexity (there is an English version of the blog but I couldn’t find the link to translate this specific post ¯\_(ツ)_/¯ ).
In another corner of the financial blogosphere, Josh Brown and Monevator were pondering the return of capital protected with a kicker products.
It is time a great time to connect the dots between these two, very important, topics.
Convexity
Looking at the replies to Corey’s tweet, few have a clear understanding of what convexity means within the investing area. I was also surprised that Maurizio Papi, the person behind MP Investit, managed to write a great explanation about convexity without mentioning ‘hockey stick’ or ‘lottery ticket’ ;). Using bonds instead of options to describe convexity is indeed unique.
The lottery ticket example is the most intuitive way to explain a convex investment because its payoff is the most extreme version of a convex curve:
Imagine having the price of an asset on the x-axis and the payoff of your investment on the y-axis. Now, look at what the payoff does when the asset price increases or decreases by the same amount: you make more if the asset moves up/to the right than you lose when the asset moves down, even though the move in price was the same both ways. That’s convexity in a nutshell.
In my badly drawn graph, the orange line represents an investment that is more convex than the yellow one: the investor has a lower payoff below a certain asset price and higher after. The most extreme version is the blue line, the payoff of a lottery ticket: there is only one scenario when the investor gets a payoff and in every other they lose everything.
Why convexity is important
Everyone is exposed to convexity, not only investors. Different careers have different convexity values. Taleb calls them scalable and non-scalable jobs. Entrepreneurship is a scalable job while being a doctor is non-scalable.
In this case, the unit price is an hour of work (and we all have 24 hours in a day). Writing this post takes me a fixed amount of time but the payoff linked to this activity can go from zero to Chiara Ferragni. A great doctor has an hourly rate higher than a bad one but they both have a maximum amount of patients that they can see in a day. Founding Uber is a scalable activity while driving a cab is not.
Scalable and non-scalable jobs have different payoffs but also different risks. Or as Scott Galloway put it: “it’s never been easier to become a billionaire, or harder to become a millionaire”. Scalable jobs are winners-take-all environments: few people at the top earn a boatload of money while everyone else is left with pennies. Jeff Bezos with billions and a multitude of failed (start-up) entrepreneurs with nothing other than a good story.
The difference in outcomes for non-scalable jobs is way lower: a doctor in the 20th percentile is still able to earn a decent salary and the relative distance in earnings between him and someone in the 80th percentile is way lower than Lebron James – random Italian professional player.
If you want to be a millionaire, you have a higher chance to achieve your goal by becoming a doctor than a YouTuber. We are too exposed to the results of other people’s choices and not enough to the probability distribution that leads to those results; we see the YouTubers that made it (and how much they make it) without realizing how many tried or are still trying. We also have low insight into the role of luck in those success stories. If you work hard to become a doctor, you will become a doctor and probably your hard work will manifest in some degree of success compared to other doctors that did not work as hard as you. If you work hard to become an actor, you still probably end up spending all your life as a waiter…and this outcome might not even depend on you. Tinder says that 80% of women want to see in their ‘pool’ only men 6’0” high or taller (but shorter than 6’9”): your dating chances are more linked to something you cannot do anything about, genetics, than you, who you are, what you do and what you like.
Understanding the difference between linear and exponential, in every aspect of life and not only in finance, is one of the key elements to positioning yourself to have higher chances to achieve your goals.
Risk or Reward
Maurizio Papi was prompted to write his blog post by an article published by PortfolioCharts. I remember that article because I instantly forwarded it to a guy I was working with at the time after I read it.
We were managing a portfolio that was mainly floating income instruments (like Floating Rate Notes): that portfolio returns were more linked to the rate decisions of the Bank of England than our decisions. We did not invest in Government Bonds because we saw them as ‘income’ instruments: with rates so low, what was the point of buying them? What we completely missed, and PortfolioCharts opened my eyes, was the hedging role Govies could provide to that portfolio, especially when rates were (are?) so low. The main risk for that portfolio was rates going negative (do you remember? I spent six months designing contingency plans for that event…and here we are) and Govies would have delivered in that scenario. The fact that we are in the opposite reality, where rates are going up and Govies down, does not mean that our decision to not invest was the right one. The higher income generated by the floating part of the portfolio would have more than compensated for the losses on those high duration positions anyway; our real risk was rates going in the opposite direction.
The context, and in financial terms your overall portfolio, is important.
Considered alone, 10 years Government Bonds that pay 0% are indeed a bad investment. Held to maturity, your investment will return exactly that, 0%. Pushing the rate lower, even in negative territory, offers diminishing returns in terms of help to the overall economy. No one knows where Central Banks would stop, -1%? -2%?, but there should be a floor where the banker (or the economy) says enough. At that point, your long-duration Govies are a convex investment but on the negative side: at best, you get your 0 coupon and at worst, you see a massive mark-to-market loss when rates go up.
Understanding your risks
In Italian, we refer to “fare la punta al cazzo” when someone is unnecessarily picky and strict. You might think the following part would be exactly that and you might be correct.
At a certain point in his post, Maurizio compares a Govie ETF with a High Yield one. My understanding is that he wants to show the reader how duration is not only linked to a bond’s maturity but to its coupon as well. Higher paying coupon bonds have comparatively lower duration, true.
What he forgot to say is that he’s comparing an apple with an orange.
SDUE price (the red line) is influenced by interest rate risk.
IHYG price (the blue line) is influenced by interest rate AND credit spread risk.
A credit spread is the yield difference between a bond issued by a Corporate with a certain credit rating and a Govie with a similar maturity. If you want, it is a proxy of the probability of default of that corporate bond (assuming govies never defaults…eheheh). Until Jan 22, the market was pricing the default probability of the bonds in the HY basket as relatively constant; the ETF price was only driven by its interest rate risk. After Jan 22, the market started to price an increased probability of default, or conversely, it demanded a higher yield to hold those lower-rated bonds. Here you can see the credit risk component of that ETF isolated:
The comparison between the two ETFs was legit…until it was not. In another sense, if you were holding the IHYG ETF only because of its low duration (and thought that it was the only risk you were bearing), after Jan 22 you might have woken up to a different world. A Gov ETF can indeed be more volatile than an HY one; here I think there is a bit of ‘cherry picking’ about the period under analysis. If we are going into a recession, SDUE volatility will be dwarfed by IHYG one. We are just in the early innings.
I wanted to highlight this part only because this mistake is quite common in the investing world. Do you remember when investors were buying Russian assets as a proxy for oil? Or when Greek Govies were like German ones but with some added yield? Opsie…
Risks do not matter until they do…which bring us to:
Structured Products
According to Wikipedia:
A structured product, also known as a market-linked investment, is a pre-packaged structured finance investment strategy based on a single security, a basket of securities, options, indices, commodities, debt issuance or foreign currencies, and to a lesser extent, derivatives. Structured products are not homogeneous — there are numerous varieties of derivatives and underlying assets — but they can be classified under the aside categories. Typically, a desk will employ a specialized “structurer” to design and manage its structured-product offering.
Too complicated? In its simplest form, a structured product is two securities combined into one instrument. An ETF is kind of a structured product: you do not see the price variation of each instrument inside the ETF, you only see their combined variation. In a nutshell, that is the raison d’etre of structured products (from now on SP because I am tired of typing).
SPs are a great marketing tool as well. They prey on investors’ lack of understanding of convexity, risks and probabilities. SPs allow banks to sell investment ideas with an enticing return proposition at a seemingly low level of risk. We can mainly divide them into two categories:
- The Zero Risk Investment: these SPs are usually sold after market crashes, when scarred and scared investors do not want to hear about stocks no more. The typical example is a note with 5 year maturity that returns the higher of a tenth of the S&P500 return over the same period or 0; in other words (you see why I do not work in marketing?) the investor gets some exposure to the stock market ‘for free’, because in the worst case they get their invested money back. Behind the note there is simply a Government bond and a call option on a stock index: the note issuer takes the bond coupon and use part of it to pay for the option premium, the rest goes in its pocket. These products are out of fashion not only because of the recent stock bull market but also because Govie yields are so low that their coupon cannot pay for any option premium. These SPs are sold as zero-risk but there is still a way for the investor to lose money: if the Note issuer goes bankrupt, the investor might lose part or all of their investment.
- The High Return / Low Risk Investment: these are the SPs en vague. The investor gets a Note that is linked to four wildly loved stocks, like Tesla or Netflix. The Note offers a high coupon, say 10%, as long as any of the stocks loses 40% or more. Here is basically where the marketing material stops and the ‘small print’ begins; personally I cannot understand why there are investors looking to buy this trash: if you like the stock…why do not buy the stock? Looks like people are happy to receive a fixed ‘high’ return for what they think is a small risk, any of the stock in the basket loses more than 40%. The 10% return is possible because the stocks in the basket experienced high volatility, i.e. they recently went to the moon. Under the Note, there is a sold put option on the stocks basket with a strike 40% Out of The Money; the Note yield is the option premium and if at the Note’s maturity any of the stocks has lost more than 40%, the investor loses an equal amount of principal. It is like those stupid naked option strategies you find on YT but for people so lazy that they do not even want to understand how to sell an option by themselves.
The second type of SP is particularly dangerous because it has negative convexity, meaning the investor has no risk until he has all the risk. It is like buying a knife and holding it by the blade: that’s not how convexity is meant to be used.
Not all SPs are trash. Imagine a meal, it can be good or bad based on the quality of the ingredients or your specific circumstances; Michael Phelps was consuming 12.000 calories/day at a certain point in his life, it lead him to some gold but it will probably lead you to the grave. SP instruments should offer a solution to protect the downside while sacrificing some of the upside, conscious that the sacrifice will generally be bigger than the protection offered. How big is the sacrifice? That depends on the SP issuer’s ‘honesty’, the cost hidden within the product.
The main problems with SPs are two:
- market timing: investors tend to buy insurance after the fact, possibly at the worst possible moment. They also cancel the insurance, i.e. change investment strategy, not because the insured risk disappeared but because they got tired of paying the premium. Again, with lousy timing.
- lack of product understanding: SPs are complex investments. Understanding a product outcomes and payoffs is not easy, especially if you consider risks that might hidden somewhere, like the Note issuer default risk or a seller conflict of interest.
If you want to have an idea of how I approach an SP analysis, in December 2020 I did a review of Innovator Triple Stacker ETF. Simplify ETF has also interesting products.
The ideal scenario for a multitude of SPs is a market that trades sideways. Markets DO trade sideways from time to time but it is really hard to predict when they will do so. I think their best use is to calibrate risk for each investor’s appetite, provided that the added complexity really pays off in terms of enhanced peace of mind.
What I am reading now:
Follow me on Twitter @nprotasoni