This article is authored by Christian Billinger, Investor at Billinger Förvaltning, a Sweden-based, family-held investment company with no external capital.
One piece of advice from Warren Buffett and Charlie Munger is that the secret to a happy marriage is to have low expectations. So it is in investing in our experience.
In times like these, following many years of strong returns in equity markets with a sharp acceleration following the policy response to the pandemic last spring and summer, many investors adjust their expectations upwards. According to this logic, strong share price performance and higher than average valuations would indicate continued strong performance and even higher valuations. One recent survey by Natixis Investment Managers suggests retail investors now expect 14.5% annual returns in real terms over the long term. Be that as it may, our expectations are decidedly more modest.
Our ambition is to generate 8-12% returns in nominal terms over the very long run. I would like to elaborate a little on 1) how we arrive at that expectation and 2) the benefits of having any expectations at all.
A couple of caveats are in place: 1) we are not suggesting this is the only way of setting expectations; 2) we are not even suggesting this is the right way to set expectations; 3) we are saying that this is what seems to work for us, which is the key message; find your own “algorithm” and build your portfolio on that basis.
What to expect (when you’re expecting)
We would be satisfied to grow our capital at around 10% over time and have set our ambition at 8-12% in nominal returns. But why is that our expectation?
First of all, we look at what returns we would need to generate in order to reach important personal financial goals like buying a house, buying a holiday home, paying school fees and given our starting point in terms of assets etc. Over the long run, 10% would do very nicely. Of course, we would be delighted if we could generate 15% or 20% per annum but to paraphrase; the portfolio doesn’t know that we own it and quite frankly it doesn’t care. We cannot force a certain level of returns simply because we want to achieve them and I believe this is in fact a very common confusion that results in the sort of “expectations” formed by retail investors already mentioned. This is attempting to see the world as it really is as opposed to how we would like it to be.
Second, we look at what returns we think we are able to generate given our temperament, skillset, opportunity set etc. For us, as a long-term investor in high quality listed companies, the equation looks something like this (what follows is the technical section):
Total return = Dividend yield + EPS growth
or in the case of our current portfolio:
10% = 2% + 8%
Of course, there are a number of assumptions underlying this simplified interpretation e.g. that there is no significant re- or de-rating of our portfolio (although this impact becomes significantly less important as we extend our time horizon/holding period), that payout ratios remain constant etc. Also, these numbers will change over time as the environment changes, i.e., the “algorithm” needs to be dynamic and respond to market conditions. For instance, in spring of 2009 the expectation would justifiably have looked different from what it does today. As a broad rule it has merit however. It is also worth mentioning that this is broadly in line with the historical performance of our portfolio companies over very long periods of time and there is significant amounts of empirical data suggesting this is not an unreasonable expectation for most of us.
One point to note is that we are talking about the very long term. Even 8-12% seems ambitious to me in light of where we are today unless we really stretch out the timeframe. However, over very long periods of time our return should approximate that generated by the underlying businesses and in this light 10% looks like a reasonable expectation.
Why to expect (if you’re expecting)
Why, you might ask, should we form any expectation or ambition in the first place? Can we not just aim for the highest possible return? You can – but for most people I don’t think it is a very good idea. Here are some of the benefits I see of having a return “algorithm”:
A more focused research process
The really important number for us is the 8% earnings growth assumption (or whatever number you come up with). That really directs our research effort, both in terms of monitoring existing holdings and when it comes to identifying new ideas. I think of it as a searchlight or a lens through which we view the investment universe to identify ideas. For us, the ideal company tends to grow revenues by around 4-8% organically which combined with some margin expansion over time (operating leverage/mix/efficiencies etc) and below-the-line impacts tends to result in an EPS CAGR of around 8-12% (although we own companies that grow both a little less and a little more than this). This is our “sweetspot”.
Of course there are other investors who look for low/no growth traditional “value” plays and there are those who look for high growth or venture capital opportunities. All of those can be valid approaches. For us, if we can combine a number of the types of assets I have mentioned into a portfolio we believe there is a good possibility we can generate 8-12% returns as a whole. We also look at what other investors with a similar temperament and philosophy are looking for in terms of returns and all together this suggests that 8-12% is a good starting point. Hopefully one can add some value by buying or adding at attractive levels and trimming positions that have become clearly expensive — although for us this only takes place when we feel the market or individual holdings are at extremes which is very rare.
For other investors, this equation will look different but is nevertheless a good starting point when it comes to focusing ones research effort etc. It also means we can focus on simplifying our process. I remember the days of being a fresh-faced (or, at least, fresher-faced) analyst using ever more convoluted reasoning in order to justify, or “pitch”, my ideas. These days, my justification would usually be, “I think there’s a decent chance I will get around 10% earnings growth from this business over time at relatively low risk.”
Better risk management
Having a moderate (as we like to think of it) return expectation brings a few benefits in terms of risk management, including the following:
– On the asset side, we can focus on high quality assets with relatively low risk, i.e., we don’t need to “go far out on the yield curve” in terms of risk. However, please note that we are referring to operational and financial risk as opposed to variations in mark-to-market prices. If anything, the latter are to be welcomed.
– On the funding side, the 8-12% expectation means that we see no need for borrowing money or employing any other form of leverage. This brings real durability.
– We can diversify across industries, geographies etc. Of course, the biggest winners in any one period will often be investors who have been heavily invested in one or a few names that have done spectacularly well. However, so will the biggest losers. One needs to distinguish between building capital and managing capital.
– If a position moves against us (operationally and/or in terms of share price performance), it is very helpful to have a “ruler” with which to review our position. If we still believe after our review that the holding can generate returns in line with our ambition we can hold or add to our position. Investing/speculating without such a framework is fine when all is going well but will create major difficulties when things start to go wrong. There is simply nothing to anchor your views on an investment in that situation and you may act in panic.
Longer duration as an investor
Most investors focus almost exclusively on the level of returns to the exclusion of the duration of returns when discussing performance. Of course, in order to win awards and trigger bonus payments that is a perfectly rational thing to do, however for those investing their own capital the two aspects of returns need to be approached together.
I will take the liberty to make a very broad distinction between investors generating truly exceptional returns of around 30% per annum for an extended period of time, e.g., Peter Lynch at Magellan and Stan Druckenmiller at Duquesne; great returns of around 20% per annum for a very long period of time which we have seen from a small group of fund managers including famously Warren Buffett over many decades; and good but unexceptional returns of around 10%, which is what many equity markets and institutional fund managers have produced over time. What we need to consider in addition to the level of returns is for how long they can be sustained.
For most of us, based on our skills, temperament, and opportunity set, 20-30% per annum is not a realistic ambition. In addition, for most investors the physical and emotional toll of the sort of high-touch approach employed by the likes of Lynch and Druckenmiller results in burnout, etc. Buying and holding great businesses for the long run however seems to be a more sustainable pursuit for most of us. (If you need any more convincing, Druckenmiller, already well into his billions, turned down an invitation to play in the Alfred Dunhill Links because he was too busy. That’s no way to live.)
We prefer to model ourselves on the likes of Tom Gayner at Markel who has an unrelenting focus on incremental improvements sustained over very long periods of time. This also involves staying in business for long enough to allow that compounding to manifest itself. However, make no mistake; this approach really could be described by the expression “simple but not easy”. Few are able to exercise the discipline and patience required. It should also be said that Tom’s track record is truly exceptional and amounts to much more than the target we have set ourselves. Another investor in the same vein who we much admire is Tom Russo who describes his approach as that of being a “farmer rather than a hunter”.
Here are some “options” for a relatively young investor looking to grow his capital from $10 million to $100 million (or from $1 million to $10 million):
8% per annum for 30 years
10% per annum for 25 years
12% per annum for 21 years
15% per annum for 17 years
20% per annum for 13 years
30% per annum for 9 years
For any investor under the age of forty (by which point you should ideally be well into your journey of compounding capital), even “options” 1 and 2 will produce a very attractive outcome. For most of us options 5 and 6 are probably not achievable. So what? If you start young enough and truly enjoy the process, steady compounding at around 10% per annum will do very nicely. As Charlie Munger says, it is not a great tragedy if someone else is getting rich a little bit faster than you.
One aspect to consider here is structure, e.g., whether you invest through a closed-end vehicle or an open-end vehicle, what liquidity constraints you have etc. One reason we can operate with a “moderate” return expectation is that we have permanent capital. The formulation of an “algorithm” is of course more difficult if you are open to outside investors and need to market your fund; sensible expectations are unlikely to attract as much capital as more aggressive projections.
It’s easier to beat your expectations when they are low!
This perhaps sounds trite but is an important point and comes back to Charlie Munger’s relationship advice. It is much easier to have a satisfying career if you set your expectations at a reasonable level and you will be delighted if you manage to exceed the “target” return over time.
What I can say about our future performance (and that of almost any money manager) is that the outcome will not match our expectations. It will almost certainly not be 10% per annum It is quite possible that it will also fall outside the 8-12% range. We have no illusions when it comes to our ability to predict anything, including our own performance.
However, I would make a couple of points in this context:
1) Establishing an “algorithm” is still useful in order to direct our research efforts, manage risk, etc.
2) The uncertainty involved when it comes to future performance highlights the importance of downside protection. What I mean by this is that given uncertain outcomes, before we even consider the potential upside of an investment we need to do everything we can to “stay in the game” and preserve our capital. For us, this happens both at the company level (owning quality), at the portfolio level (diversification, balance sheet strength, etc.) and at the personal level.
3) The expectation is set at a level where even if we undershoot, we will be fine. For us, given our payout ratio, operating costs etc we need to generate returns of around 5% per annum in order to preserve our capital over time. Should we fail to exceed that level for an extended period of time we would of course need to consider all options.
While we tend to avoid most forms of predictions in our investment approach, we do formulate ambitions when it comes to our long-term returns. The benefits of this are numerous including a better research focus, improved risk management, increased “life expectancy” as an investor, etc.
Considering both our financial needs and the opportunities available, our ambition is in the 8-12% range over the very long term. With this as a starting point, we break down the “algorithm” to MSD to HSD organic revenue growth, ability for some margin improvement and other impacts below-the-line etc to result in long term earnings growth in line with our ambition. Of course, this won’t be the case for each individual holding in each individual year but for a well constructed portfolio it seems like a reasonable ambition overall. Most of our time is then spent trying to identify and monitor businesses which can deliver on those “targets”. While we would be delighted to generate returns well in excess of our “expectation”, we are not planning for it. It seems then that, just like when it comes to relationships, low expectations are the answer.