The Aging Operator Ch. 3: Designing a Portfolio to Be Run Badly

Investment Background

Portfolios are designed on the assumption of competent operation. Every backtest, every allocation model, every rebalancing schedule quietly assumes someone performs the required actions correctly and on time.

Chapter 2 established that this assumption weakens over a long retirement, and that you will not be notified when it does. So the design question changes. It is no longer "what does this portfolio return when run well." It is "what does this portfolio do when run badly for five years by someone who does not know they are running it badly."

A portfolio that performs well under competent management and fails under neglect is a fragile portfolio — and it is fragile in the specific direction that a thirty-year retirement is most likely to bend.

The Wall Street Translation

The Design Criterion Nobody Applies

Grade each holding by a single question: what happens if nobody touches this for five years?

The ordering is the point, and it maps almost exactly onto complexity. The structures that survive neglect are the ones requiring no decisions. The structures that fail are the ones whose returns depend on someone continuing to act.

Not the Same Argument as Automation Against Emotion

Three books in this library already argue for automation, and this chapter is making a different case that must not be confused with theirs.

The distinction is not academic and it changes what counts as sufficient. In all three existing cases, the operator is healthy and returns to baseline once the moment passes — automation carries them across a temporary emotional state. Here there is no return to baseline. The relevant test is not "will I be tempted to act badly during a crash" but "will this still work if the operator never again performs a discretionary action correctly."

That is a strictly higher bar, and some arrangements that pass the first test fail the second. A rule you have committed to follow mechanically still requires someone to follow it. A structure that requires nothing requires nothing.

Simplification Has a Real Cost, Stated Honestly

This chapter argues for fewer holdings and less complexity, and that is not free.

Consolidating may realise capital gains. A single target-date fund may be marginally less tax-efficient than a hand-built allocation, and may hold an allocation slightly different from what you would choose. These are genuine costs, usually measured in tenths of a percent.

They are worth paying because the risk on the other side is not measured in tenths of a percent. A portfolio that drifts unmanaged for a decade, or that is liquidated in confusion at the wrong moment, or that becomes unmanageable by a survivor, costs far more than the tax efficiency it was optimised for. The trade is a small certain cost against a large uncertain one, and this is the case where paying the small one is clearly correct.

One caution against over-applying this. Simplifying is not an argument for selling everything and sitting in cash — that is its own failure, and a more expensive one over a thirty-year horizon. The goal is fewer moving parts, not less equity exposure.

Division of Labor With the Rest of the Library

The relationship with index-fund-machine is worth stating. That book teaches you what you own so you are not surprised by it. This chapter adds a criterion that book does not use: prefer the structure that needs the least from you, because the amount you will be able to give it is not constant. A target-date fund is unremarkable to an engaged investor and quietly excellent for an eighty-five-year-old, for reasons that have nothing to do with its expense ratio.

Executable Trading Rules

1. Apply the five-year neglect test to every holding, in writing. If nobody touches this for five years, is the outcome fine, degraded, or catastrophic? Anything in the third category is a structural problem regardless of how good an idea it currently is.

2. Reduce the number of accounts before reducing the number of funds. Multiple accounts at multiple institutions are the single largest source of administrative failure — forgotten balances, missed RMDs, unreachable statements. Consolidating custodians is usually free and does more than any fund substitution.

3. Prefer one fund that de-risks on its own to three that require you to do it. The glide path from human-capital-portfolio ch05 has to execute for thirty years. A structure that executes it without an operator is categorically safer than one that depends on your continuing to act.

4. Set the simplification to a date, not to a feeling. "At seventy-five I consolidate to no more than three funds and two accounts" is a plan. "I will simplify when I start finding this difficult" relies on the signal chapter 2 showed to be unavailable.

5. Write a one-page description of the portfolio in plain language. What is held, where, why, and what the annual routine is. If it cannot be written on one page, it is probably too complex to survive the operator it will eventually have. This page is also the artefact chapter 5 depends on.

Relevance to a Retirement Portfolio

This site's portfolio-simplifier tool exists for exactly the operation this chapter describes — reducing overlapping funds and accounts to the smallest set that preserves the intended exposure.

And the connection to the rest of the retirement machinery is direct. retirement-decumulation-mechanics chapter 6 specifies a nine-step annual review, and chapter 3 of that book builds a bucket buffer whose refill rules require judgment. Both assume an operator. Neither is wrong — but a plan whose annual routine is nine steps is a plan that will, at some point, be run by someone who can comfortably manage three. Designing the routine so that its most important elements survive the loss of the rest is the work of this chapter.

The standard recommendation is unchanged: a core of low-cost, globally diversified index funds, no leverage, a cash buffer covering essential spending, and withdrawal rules containing an adjustment mechanism. What this chapter adds is a tiebreaker — when two arrangements are otherwise comparable, choose the one that asks less of the person running it, because you are choosing on behalf of someone who will have less to give.

Chapter 4 turns to the risk that does not wait for neglect: the people who look for exactly this situation on purpose.