Take everything an employer transfers to an employee over a year. Salary, bonus, the health plan, the disability coverage, the savings match, the equity if there is any. Now sort it not by what it is, but by what it is supposed to accomplish.
You get four piles, and I have not been able to find a fifth.
- Consumption. Funds the life being lived now. Rent, groceries, childcare, the commute.
- Protection. Absorbs shocks. Health, disability, life, income continuity.
- Alignment. Ties reward to outcomes over a defined period.
- Preservation. Carries the value of today's labour to a future far enough away that the value has to survive a journey to get there.
Three of those four have a mature science behind them.
Consumption has market pricing, salary structures, job architecture, pay equity analysis, an enormous survey industry. Protection has actuaries, insurers, underwriting, decades of loss data, and pricing models that work. Alignment has incentive theory, a large empirical literature, a specialist consulting profession, and disclosure regimes that force people to think about it carefully.
Preservation had one too. It was the defined benefit pension.
What the pension actually did
It is easy to be sentimental about defined benefit pensions and it is worth resisting, because they had real failures. Benefits in payment were typically fixed in nominal terms, and the inflation of the 1970s found that seam and tore it open. Coverage was never universal. Vesting rules were often brutal to anyone who left early. Plenty of sponsors underfunded them.
But look at what one did for somebody who stayed. It pooled longevity risk across thousands of people, so no individual had to guess how long they would live. It carried investment risk across four decades of markets on their behalf, professionally, without them having to watch. It linked the benefit to final salary, so inflation during the working career could not erode the accrual. And it converted thirty years of labour into an income that could not be outlived.
Four genuinely difficult problems in finance, solved institutionally, on behalf of a person who never had to read the booklet. That is what the fourth job looked like when somebody owned it.
The promise was denominated in outcomes rather than in dollars, and institutions the employee would never meet were paid to keep it.
The job did not go away
Here is the part that took me a long time to say precisely.
When defined benefit coverage receded, the profession did not decide that preservation no longer mattered. Nobody held that meeting. What happened is subtler and worse: the job stayed exactly where it was, and its client disappeared.
The instrument that replaced it is a savings account with a tax wrapper and a menu of funds. I want to be fair to it, because it does things the pension could not. It is portable in the sense that the balance moves. It is transparent. It is the employee's property, which matters enormously and is easy to underrate. It does not fail when the employer does.
What it is not is a preservation architecture. It is a container. It hands back to an individual, one at a time, the four problems the pension solved collectively, and it does so without pooling, without professional stewardship of the decisions, and without any promise about what arrives. The employee is now the actuary, the investment committee and the decumulation strategist, in addition to their actual job.
The Four Jobs. Consumption, Protection, Alignment, Preservation. Three have a mature science. Preservation is the one the profession stopped designing for, because the institution that used to own it stopped being available and nothing inherited the role.
The part we do not say out loud
There is a consequence of this that I think is genuinely under-discussed, and I want to state it carefully because the careless version of it is wrong and easy to attack.
When the real return on the safest accessible claim is at or below zero after tax, an employee who simply wants to maintain the value of what they earned cannot do it by being cautious. Standing still now requires accepting variance. The gap between zero and that real return is what the papers call the preservation burden, and when it is positive, the safe option has stopped being safe in the only sense that matters.
So a portion of the investment risk that ordinary workers carry is not risk taken to get ahead. It is risk taken to stay level. That portion is real, it is not chosen in any meaningful sense, and its size is genuinely unknown. I am not going to put a number on it, and I would be suspicious of anyone who does, because it is defined counterfactually and the counterfactual is not observable.
The concept does not need a number. The observation is enough: we never asked employees to become investors. The design made them one, and then the profession described the result as a matter of personal financial literacy.
What follows, and what does not
The obvious objection arrives about here, and it deserves a direct answer. If preservation is compensation's problem, has the employer just been made responsible for forty years of markets and personal decisions?
No. And the line is drawable.
You are accountable for the architecture of your own increment. Whether what you offer survives a job change. Whether an ordinary person under stress can actually operate it. Whether you told them the truth about what it does and does not do. Whether the thing is denominated and structured for the horizon it is supposed to serve.
You are not accountable for returns, for what someone does with their own property, or for the outcome of a life you do not get to live for them. Architecture, never outcomes.
There is also an ordering rule that matters more than it sounds, because without it this argument becomes a licence to sell long-dated products to people who need cash this month. Preservation instruments become offerable only behind funded adequacy in the present: emergency liquidity in place, high-cost debt addressed, any available match captured. Preservation is what you build once the present is solvent. It is never a substitute for making it solvent.
The question I keep ending on
Every employer I know offers something for the first three jobs. Salary for consumption. Benefits for protection. Some form of incentive for alignment.
Almost none of them offer anything designed for the fourth, and most have never been asked why, because the question does not exist in the frameworks we use.
We built a benefit for every job but one. What would a benefit designed to carry value through time actually look like?
I have a commercial answer to that question, which is exactly why I am not going to give it here. The question is worth more than my answer to it.