Position Sizing

Risk management starts with position sizing. Every position carries many risks. Some are known, some are unknown. Some can be planned for, some can not be planned for. The typical example that comes to my mind is the CEO (or for funds, the PM) going insane or committing fraud. That's a risk that is (almost) completely independent between different positions and the only way to manage it is to keep position sizes small. Every position also carries the "I'm a fucking idiot and made a stupid bet" risk.

Diversification

A lot of diversification falls naturally out of your position sizing. But you also have to think about how correlated your positions are. A common one nowadays would be someone owning a bunch of stocks that are all highly correlated to the AI factor.

You can:

  • diversify within asset classes (US stocks + International stock)
  • diversify across asset classes (stocks + commodities)
  • diversify across alpha sources (multiple fundamental managers + quant strategies)

You should think about correlations over different time frames and during different regimes. Almost everything (including most commodities) had a correlation of 1 with $SPY over certain timeframes during March 2020. You need to plan for that, so you don't blow up. But as long as you survive, the risk of a poor long-term returns is minimized by focusing on correlations over a longer timeframes (approximately the timeframe that you'd rebalance your portfolio on). Some things will do well in bull markets, some in inflationary regimes, etc. Take some time to think about that and if everything you own will do bad in some regime, then consider adding something that will do well in that regime.

Tail Risk

You can buy 50% out-of-the-money $SPY puts for like 20 basis points a year. If you're chasing large alpha and/or using margin to boost returns, then this should be a negligible cost to CYA in an extreme outlier event. I use it to manage margin call risk and to help ensure I have some liquidity in such an event.

Upside Beta vs Downside Beta

The easiest example here is a covered call ETF. These structurally limit their upside. This more or less guarantees they have higher downside beta than upside beta. For this reason, they make great shorts when viewed through a risk management lens. They do come with a separate risk, which is theta decay (i.e. slowing losing money over time). But again, if you're chasing larger returns then this should be well worth the cost.

Above the portfolio level

Outside of my actively managed IBKR portfolio I also have:

  • a large passive portfolio (my 401k)
  • real estate
  • a couple private investments
  • emergency fund
  • smaller non-margin accounts with other brokerages

Some of these aren't 100% by choice, but they do exist.