Concentrating on Risk
I've written about this before but it's worth revisiting.
Whenever you hear a portfolio manager at a concentrated fund talk about position sizing based on reward rather than risk, you should get a little concerned. And when that conversation happens after a big drawdown, you should get very concerned.
Playing Life on Hard Mode
Managing a concentrated portfolio is playing active management on hard mode. Each decision matters more. The stakes are higher. One is more susceptible to anchoring, confirmation and overconfidence biases.
Imagine a stock down 20% (or 25%). You need +25% (or +33%) just to get back to even. When the stakes are that high it's near impossible not to have an emotional response. If you're reading this, it's probably happened to you too, you tell yourself you were just early, or the market is stupid, or you messed up the entry and it's priced in now. And often the right move is to do nothing.
And sometimes the fund doubles down. The position that was 10% of the portfolio is now 15%. The update calls go on for longer, the upside cases more elaborate. And "risk" often totally disappears from the vocab.
Sizing Based on Risk
Sizing based on how much you can afford to lose not how much you think you can make. I keep coming back to this, it really does save you a lot of pain & difficulty.
Everyone will be wrong at some point in time no matter what, so you try and not let the losers sink you. I've seen many funds blow up by conviction-based sizing where the focus on maximising expected return instead of minimising losses, combined with concentration, eventually catches up with you.
And blowups do something nasty to judgment. It's hard to admit you're wrong when you've lost that much money.
What To Listen For
A few things that have always made me nervous:
Someone tells me "we run concentrated and size on conviction". These two things combined are problematic. People who run concentrated tend to have trouble changing their minds. Blend it with conviction and you get overconfidence.
A PM or analyst spends forty minutes on upside, zero on risk.
"This is now our largest position" (after a drawdown). The position got larger as it fell and they added and is ALWAYS presented as a feature.
Okay this is a tricky one I'll admit, it works sometimes but eventually you will get one wrong. Happened to me more than I'd care to admit, when someone tells you a position which got crushed is a 10-15% weight and here's all the upside... eek. You have a small drawdown window before a stock will end up on the wrong side of momentum factor.
Technical selling pressure is a bitch (3).
This is not a complete set of rules and rather a reflection of stuff I've seen that has worked in the past. It serves me well and hope it serves others well too. You can apply this to an external actively managed portfolio or your own personal one.
Don't blow it up!
Footnotes
I've looked at long-short funds for a long time and tried to understand the point of shorting. Generally, in Aus when you see a LS fund, they're shorting to lever up their longs and usually shorting doesn't make them alpha. Furthermore, most people are geared to think about making money by going long, whereas shorting absolutely cooks brains. It has its place, shorting reduces vol at some points in time and depends on where you are in the market cap, shorting can be quite fruitful.
I can count on one hand Australian LS funds that have generated positive alpha from their short book.
Now, that is a naive way of looking at things, let's be completely honest. Not all managers are fundamental, some are systematic. And their short-book alpha is persistent. I've looked at fundamental managers my whole life and recently gotten into properly understanding systematic (I'm learning machine learning and XGBoost at the moment, fundamental x systematic is the future I tell you).
In fact, the deeper I dive into things, I realise that approx 50% of systematic momentum alpha comes from shorts!
source: See Table 3 here.
There are various different types of windows and measures of momentum factor. Say if you take away the 12 month return of a security from its 2 month return and then map that on a gradient and you can get a structure for how long alpha may persist from there. Awesome study by HIMCO showing that momentum factor returns last around 14-15 months. The signal period (12 mth vs 7mth) doesn't change duration of the factor. Why? God who knows!
Source: HIMCO Insights 2018
Now back to our fundamental concentrated manager, they allocate a whole bunch of capital to a security which falls a little bit and doesn't bounce. If you've ever deeply looked at an order book, day in, day out, you'll sometimes see there is some prick selling into every open and close, someone clearly trying to get out.
And if there is persistent selling, at the 2 or 3 month mark, the stock (if large enough) will filter into the bench at some systematic fund. If the fundamental earnings were remotely mediocre (remember, cockroaches never come in ones!), momentum is a sensational factor for generating returns.
Winners and losers persist. Buy stuff that is going up and sell stuff that is going down.
All of a sudden, mister concentrated fundamental manager over here is trying to work out who keeps selling. While that other seller in market starts watching the stock trade lower and lower as someone (the systematic fund) keeps taking out any volume. Then the seller decides to push the selling harder! And in a few months time you see the "god candle" when mister concentrated fundamental manager admits defeat and sells at the bottom.
Ask your active/fundamental manager about this next time, most don't even have a view on this.
Technical selling pressure is indeed a bitch.