One thing I’ve talked about a lot in recent years is how sequence of returns risk is the holy grail of investing. In other words, we absolutely do eat risk adjusted returns because a smoother return path means we can more predictably consume. The investor who earns 12% per year with 20% volatility has a much more unpredictable consumption path than the investor who earns 10% per year with 10% volatility because their returns are a rollercoaster. Even though the 12% investor ends up with more money on average their ability to consume is constantly in flux because their portfolio value is so volatile. The investor with the better risk adjusted return consumes more predictably because their balance sheet value is more consistent.
The path, or sequence of your returns is hugely important in your ability to plan for the future. This is the problem I am solving with the Defined Duration strategy. We’re not just diversifying portfolios over asset classes. We are specifically diversifying portfolios over time horizons to create better sequence of returns. More importantly, we’re matching these portfolios to specific financial plans thereby creating a seamless integration between the financial planning process and the portfolio construction process.
I built this tool to help visualize this. It’s fairly raw, but it communicates an important point – good financial planning isn’t just about diversification. It’s about making sure your diversified assets help you understand your ability to meet specific needs across specific time horizons. This is what investors really care about. While the traditional Efficient Frontier tries to pinpoint a theoretically risk optimized portfolio the Defined Duration Frontier adds the most important element in financial planning – TIME. The DD Frontier helps you visualize the portfolio that best mitigates sequence risk for you. Importantly, that point is different for everyone. There isn’t some cookie cutter or theoretically optimal point on the Frontier because it’s different for everyone. A 30 year old with a 40 year time horizon and high income has far less sequence risk than an investor who is 65, retiring and starting to live off their portfolio.
What you’ll see with this tool is that you need to balance assets over many time horizons to create more certainty and reduce sequence risk via optionality. For example, owning 100% AOR (global 60/40) is a theoretically optimal allocation according to the traditional Efficient Frontier – you’ve balanced your risk in the portfolio in what is often viewed as an optimal level of risk per unit of return. But this is a 2D view of risk/reward. You need the temporal variable to see that 60/40 actually has a high level of sequence risk within it at 16 years. This is because the average defined duration of all the instruments is blended together so when it falls you have zero near-term optionality in that instrument. As the tool shows, you are only diversified over one time horizon (the intermediate 7-15y time horizon). You need to disaggregate the diversifiers to reduce the temporal risk in the portfolio. So, for instance, in the tool you can take 70% AOR and 30% TBills and you drop the Defined Duration from 16 to 9.9. You’ve bolstered the certainty by adding more temporal optionality. Or, if you want to keep the same general allocation as 60/40 you could take 60% VT and then add 30% VGIT and 10% TBills. This creates a slightly longer weighted duration, but it vastly improves the optionality because now you have specific short-term options to withdraw from. In other words, you’ve maintained a similar weighted duration, but you’ve improved your sequence risk optionality by disaggregating the 60/40 into its different durations.
The tool also displays a sequence risk score (1 is low risk and 10 is high risk). Like a golf score, lower is better. The sequence risk score isn’t just measuring the overall defined duration of the portfolio, but it’s quantifying the exposure across different time horizons. A portfolio that is 100% VGT will have a high score because it’s all long duration assets whereas a portfolio that is 60% VT, 30% VGIT and 10% TBills will have a score of 2.6. Interestingly, the 100% AOR option has a sequence score of 5 communicating to us that the portfolio lacks sufficient optionality to reduce sequence risk. Again, it’s reinforcing the fact that even though AOR is a diversified portfolio its homogeneous nature makes it an insufficient sequence risk diversifier. Keep in mind though that a low sequence risk score might not mean it’s a better portfolio. Your sequence risk needs to be assessed relative to YOUR financial planning needs. It could be perfectly appropriate to have a high sequence score and a low sequence score might be overly safe relative to your financial plan. As I write in my new book, this all has to be personalized to find YOUR perfect portfolio.
So, we CAN eat risk adjusted returns. But finding the most efficient risk adjusted portfolio is hard. We know that from decades of material on the poor performance of active management. But it depends on what mean by an “efficient” portfolio. If you mean optimizing it for returns per unit of risk in the short-term then that’s very hard. But if an efficient portfolio means having certainty of a needed asset value at a certain time in the future then adjusting a portfolio to match specific liabilities across time is a much simpler endeavor. And quite frankly, one that’s a lot more relevant for practical financial planning processes.
Anyhow, I added the tool to the Discipline Funds Tool Suite. I hope it fills in some gaps for investors who are trying to understand how important the element of time is in portfolio construction.