Steven Nathan: Right half the time, rich all the time
Picture a group of mathematicians and former code breakers in a quiet office on Long Island in the late 1980s. Jim Simons has hired them to find faint patterns in price data, and the system they build will become the Medallion fund at Renaissance Technologies, the most profitable trading machine in history. Medallion earned roughly 66% a year before fees for three decades, about three times Warren Buffett's long-run 20%. After fees, $1,000 invested in 1988 would have grown to roughly $20 million by 2018.
What people forget is that the machine was wrong almost half the time. Bob Mercer, co-head of the firm, said they were right on about 50.75% of their trades, but 100% right 50.75% of the time. The edge lay in cutting a losing trade quickly and without argument. Simons put it bluntly to a colleague who was slow to exit one: when you smell smoke, get out.
Meanwhile, Long-Term Capital Management (LTCM) had assembled two Nobel laureates and John Meriwether, the most admired bond trader of his generation, to run what was sold as a low-risk strategy. Its models said a serious loss was close to impossible, so the fund never settled how much it could afford to lose. When Russia defaulted in 1998 it lost about $4.6 billion, more than 90% of its capital, in under four months, and had to be rescued in a bailout brokered by the Federal Reserve. The smartest people in finance were right most of the time, and it did not save them, because they could not survive the stretch when they were wrong.
The difference comes down to one discipline: ride the winners and cut the losers before they take you out. For families the lesson is not to trade more, but to decide, before markets become stressful, how much downside is acceptable and over what period.
Now look at the typical high-net-worth portfolio. There is a return expectation and a label such as "balanced" or "high risk", which says nothing about how much it could actually lose. Nobody has answered the one question that defines risk management: how much could this portfolio lose, over what period, before it threatens what the money is for? Without that number, families often carry more risk than they can tolerate. So when a severe drawdown arrives they sell near the bottom and buy back once markets recover, which is the one reliable way to turn a temporary loss into a permanent one.
Centric Family Office's wealth management strategy always starts with risk. Before we talk about returns, we agree a risk constraint in plain numbers. One family might say it can tolerate overall portfolio losses of 5% a year over a five-year window. Another might say that over any two-year period the portfolio may not lose more than 2% a year, and that over five years it must at least break even across the range of outcomes we model. This is risk appetite rather than risk capacity. Capacity is what a family could absorb. Appetite is what it is willing to live through without abandoning its plan, and most of our clients could absorb far more than they want to.
Once the constraint is set, growth assets are sized by what the downside allows, and the family knows in advance what a bad stretch may look like, which is what keeps it invested should one arrive. This is not a guarantee against loss, because markets do not honour forecasts, but it makes risk visible before capital is committed. Renaissance never needed to be right more than half the time. A family needs to be right about one thing: how much it can stand to lose, and for how long. That is what keeps it with its plan through the bad years and achieving long-term goals.
Steven Nathan CA(SA) & CFA Founder, Centric Family Office & SWAN Wealth Management
Former Managing Director, Deutsche Bank, and founder of 10X Investments
Centric Family Office is a juristic representative of Centric Wealth Advisory (Pty) Ltd – FSP 47766.