Most retirement portfolios are asked to do two jobs at once, and then judged as if they were doing one.
The first job is to pay for your life: reliable money, available on schedule, that does not care what the market did last quarter. The second job is to grow — to stay ahead of inflation and taxes over a thirty- or forty-year horizon, so the money still means something two decades from now. These are not two settings on the same dial. They are different problems. They have different time horizons, different definitions of success, and they fail in different ways. Asking a single pool of money to do both is how families end up with a portfolio that is too conservative to grow and too jumpy to rely on — the worst of each job, sold to them as balance.
For about forty years, the standard answer was the balanced portfolio: some stocks for growth, some bonds for stability, mixed in one bowl and rebalanced once a year. The stock side did the growing. The bond side was the part you were supposed to be able to count on — the ballast that held steady, or even rose, when stocks fell. The whole arrangement rested on one condition: that high-quality bonds pay enough to matter and hold their value precisely when stocks are dropping.
That condition no longer reliably holds, and 2022 was the plain demonstration. Stocks and bonds fell together, hard, in the same year. The ballast didn’t ballast. This is not a claim that bonds are useless or will never play a role again — it is narrower than that. It is that a single allocation was quietly doing two incompatible jobs, and when the stabilizer stopped stabilizing, both jobs failed at the same moment. The person who needed income that year and the person who needed growth that year were the same person, drawing from the same account, and the account let them down twice.
Certainty and growth are two different jobs.
APW separates the two jobs on purpose. Money that has to produce dependable income is built for dependability, using tools designed to pay a defined amount on a defined schedule. Money that has decades before anyone touches it is built for growth, and is allowed to be volatile — because over thirty years, volatility is the price of the return, not the thing to fear, provided nothing ever forces you to sell at the bottom.
That last clause is the entire point of separating the jobs. The danger in retirement is rarely a bad year in the market. It is being forced to sell good long-term assets in a bad year to pay for this month’s life — locking in a loss to buy groceries. When the income job is handled by something built for income, the growth portfolio is never put in that position. It is left alone to do the one thing it is good at, on the only timeline that thing works on. The two jobs stop interfering with each other.
None of this is free, and precision means naming the cost rather than the brochure version. Building a floor of dependable income usually means giving up some upside and some flexibility on the dollars assigned to that job. And an income tool that rests on a contractual promise is only ever as dependable as the institution making the promise — a guarantee from an insurer is backed by that insurer’s ability to pay, not by a law of nature. Those are real tradeoffs. They are worth making for many families and wrong for some, which is exactly the kind of thing that belongs in a planning conversation rather than on a web page.
So this page won’t tell you what to buy. The specific tools, the allocations, the mechanics — those are decided with a client, against a real situation, not recommended to a reader. What belongs here is the principle underneath all of it: certainty and growth are two different jobs. The mistake is not choosing stocks over bonds, or equities over annuities. The mistake is one bucket, one strategy, asked to do two jobs it cannot do at the same time. Build for each, separately, and each one gets to be good at its job.