Kenneth French once defined risk as the uncertainty of lifetime consumption, which is a clinical way of saying that what you’re actually worried about, when it comes to your money, is whether it will be there when you need it.
That definition reframes the whole purpose of a portfolio. A portfolio isn’t a return-maximizing engine sitting off to the side of your life; it’s a funding mechanism for the things you plan to do, and the measure of whether it’s working is whether your future spending needs can be met.
That reframe is also the starting point for a framework Cullen Roche calls “Defined Duration Investing,” which, once you see it, becomes the cleanest way to think about building a household portfolio.
The old idea underneath it
Banks and pension funds have been running a version of this approach for as long as there have been banks and pension funds. It goes by the unattractive name of asset-liability matching, and the mechanic is simple: match the tenor of what you hold to the tenor of what you owe. A pension fund that owes a retiree fifty thousand dollars ten years from now doesn’t fund that obligation with a two-year CD or a thirty-year equity position. It finds an asset whose cash flows line up with the bill, because mismatches between when money is owed and when assets mature are the thing that breaks institutional portfolios.
Households face the same structural problem and generally don’t treat it the same way. A family portfolio tends to get treated as a single pool that’s supposed to go up over time, even though the obligations it’s funding are anything but uniform. Some of that money is needed next month, some in five years, some in thirty—and if the assets funding those obligations aren’t themselves laddered across those time horizons, the portfolio is carrying a mismatch that will eventually show up as forced selling at less-than-ideal moments.
What Cullen Roche added
Bonds held to maturity have always given investors a form of built-in certainty: a known maturity date, a known principal, and a interest payment in between.
Roche’s contribution is to extend a similar measure of certainty to the broader portfolio (bonds AND stocks) by estimating a “point of indifference” for different permutations—essentially, the holding period long enough that a typical drawdown is likely to have been recovered in real terms. T-Bills land near zero because their outcome is close to certain; the bond aggregate comes in around five years; a 60/40 portfolio runs closer to ten; global equities sit somewhere around seventeen.
Once every asset has a duration, you can line them up against your real life instead of against a benchmark, and something meaningful happens as soon as you do.
That kind of framework is what keeps investors from making consequential mistakes under stress, and it’s a feature every short and medium-term allocation in a portfolio should be delivering.
The Stoneholt translation
Our planning process starts from this premise and works outward.
We plan first (give every dollar a job), and then we choose the asset (or combination of assets) whose duration fits the job in front of it.
Short-term jobs get short-duration assets. Money earmarked for the next two to five years—emergency reserves, a down payment, a planned sabbatical, tuition—lives in T-Bills, short-dated government bonds, or money-market funds, because near-term spending needs predictable principal and the return math beyond that doesn’t meaningfully improve the household’s life.
Medium-term jobs, meaning money that has to bridge the five- to fifteen-year gap between work-optional and full retirement, get a blended portfolio built around that horizon, because a roughly ten-year duration mix is what actually matches a ten-year need.
And long-term jobs, the money that won’t need to be touched for two decades or more, belongs in equities and other long-duration assets, where year-to-year principal variance is part of the design rather than a problem to be managed around.
The payoff for the whole structure shows up in a bad year. When the long-duration bucket is down thirty percent, nothing forces us to sell it, because the near-term buckets are already funded and the thirty-percent drawdown is happening to money that wasn’t supposed to do any work in that window anyway. It becomes noise instead of damage.
Why it’s worth the trouble
The case for duration matching is that it keeps the plan intact through the moments when a standard portfolio turns its owner into a forced seller—and in practice, those moments are where plans actually fail.
That’s what “every dollar has a job” means in operation. It’s the structural reason a family can sit through volatility without panicking, why we don’t chase returns in assets that are supposed to be funding a house purchase, and why the plan continues to work in years when the market doesn’t. The goal, in the end, is to make sure the money is there when you need it.
Nothing in this essay is investment, legal, or tax advice. It’s a general discussion of concepts I find useful in my work and shouldn’t be relied on as a recommendation for your specific situation. Talk to a qualified advisor (ideally one who knows you) before acting on anything here.
