Retirement is still pictured as a cliff: decades of full-time work ending at a stroke of midnight, cake in the meeting room, then sixty empty Mondays a year. Research on wellbeing keeps recording what many retirees discover the hard way — the cliff is psychologically brutal. Identity, structure, status and most weekday relationships all terminate simultaneously, and the honeymoon phase gives way, for a significant minority, to genuine disorientation and health decline.
The emerging alternative is the ramp: phased retirement. Four days, then three; a project role; consulting for the old employer; seasonal or encore work chosen for meaning. The destination is the same. The descent is survivable.
The financial mathematics of slowing down
Phasing is not just gentler — it is arithmetically powerful. Every year of even part-time earnings is a year the portfolio is not being drawn down, which matters most in exactly those first years, when the sequence of returns poses its greatest danger: portfolios damaged by withdrawals during an early-retirement crash often never recover their trajectory. Part-time income through a downturn is the cheapest sequence-risk insurance that exists.
It can also allow pensions and state benefits to be deferred, and deferral almost everywhere buys a permanently higher payment. Model a three-year half-time ramp against a cliff exit and the required portfolio at day one is often startlingly smaller — which means the ramp can begin startlingly earlier.
Designing the ramp
The strong negotiating moment is before resignation: employers facing the loss of decades of institutional knowledge are more open to three-day weeks, mentoring roles and defined projects than standard policy suggests — but almost nobody asks. Elsewhere, the ramp is self-built: consulting in the old profession, teaching it, or the encore move into paid work whose point is purpose with income as a bonus.
Two cautions keep the ramp honest. Check the pension small print — in some schemes, part-time final years can affect benefit calculations, and timing matters. And phase the calendar, not just the workload: the point is to practise retirement while still partly employed — building the routines, communities and projects that will hold the eventual full weight of free time.
Practising for the last job
The deeper insight from those who ramp well: retirement is a skill, and skills need rehearsal. The part-time years are a laboratory — discovering whether the imagined life of travel, grandchildren and gardens actually fills a Tuesday, while there is still structure to return to and income to adjust with. Cliff retirees run that experiment live, with no control group and their whole savings on the table.
None of this obligates anyone to keep working; rest, fully taken, is a legitimate reward. The argument is only for deciding the shape of the exit as deliberately as the savings that funded it. You will spend perhaps a quarter of your life retired. It deserves better than one Friday's notice.
Why part-time income does disproportionate work
The arithmetic of continued earning is more favourable than most people expect, and the reason is that it operates on both sides of the equation simultaneously. Income earned adds to resources while also removing an equivalent withdrawal, so a modest amount of work reduces the required portfolio by considerably more than the amount earned.
The effect is largest in the first years after leaving full-time work, which are also the years when sequence risk is at its most dangerous. Money earned during that window means fewer units sold during any early decline, which protects the base that everything afterwards grows from.
The practical implication is that the choice is not binary between working fully and not at all. A modest continuing income can reduce the accumulated target substantially, which for many people converts an unreachable figure into a reachable one and brings the whole decision forward by years.
Designing the ramp before you need it
A phased exit works best when the arrangement is built while you still have leverage, which means while you are still valuable to the organisation and before any announcement of intent. Once an employer knows you are leaving, the negotiating position for a reduced arrangement is considerably weaker.
The elements worth establishing are the pattern — fewer days, fewer months, or project-based — and what happens to the things attached to employment: pension contributions, insurance cover, and any benefits with a service qualification. Several of these frequently reduce or stop below a threshold of hours, and finding that out afterwards is a common and expensive surprise.
It is also worth establishing what the role actually becomes. A reduced-hours arrangement that retains full responsibility is a pay cut rather than a phased exit, and it is the most common way these arrangements fail. The scope needs to reduce with the hours, explicitly and in writing, or the arrangement produces the same work in less time for less money.
The pension complications of working while drawing
Combining employment income with pension income creates interactions that vary substantially between countries and can be costly if not anticipated. In some systems, drawing flexibly from a pension restricts how much can subsequently be contributed to one, which matters enormously if you intend to keep earning.
Tax is the other interaction. Two sources of income are typically taxed together rather than separately, which can push the combined figure into a higher band and produce a marginal rate on the employment income that makes it considerably less attractive than the headline suggests.
There may also be effects on entitlement to means-tested support, on the age at which certain provisions become available, and on the treatment of any remaining pot for inheritance purposes. None of this is a reason to avoid phased working; it is a reason to establish the rules that apply to you before starting rather than discovering them through a tax bill.
The non-financial case, which is the stronger one
The evidence on abrupt retirement is mixed but consistent on one point: the transition is harder than people anticipate, and the difficulties are concentrated in structure, social contact and identity rather than in money. A phased exit addresses all three by design.
Structure persists in a reduced form rather than disappearing overnight, which gives time to build alternatives. Workplace social contact tapers rather than stopping, which matters more than people credit given how much adult friendship is workplace-mediated. And the question of what you do has a gradual answer rather than requiring an immediate one.
There is also a practical rehearsal element. A phased period reveals what a life with more free time is actually like, which is information that no amount of anticipation supplies. People frequently discover that they want more work than they expected, or a different kind, and discovering that while still employed is considerably more useful than discovering it afterwards.
When the employer will not accommodate it
Not every organisation offers this, and some roles genuinely do not reduce well. The alternatives are worth knowing rather than treating a refusal as the end of the option.
Consulting back to the same employer as a contractor is the most common route, and it frequently suits both sides: the organisation retains access to knowledge without a permanent commitment, and you gain control over the amount of work. The considerations are different — no employment protections, different tax treatment, and the need to handle everything discussed in the irregular income article on this site.
Moving to a different employer for the final phase is the other route, and it is more available than people assume in fields where experience is valued. A role at a smaller organisation, at reduced hours and reduced pay, can be a considerably better final phase than a full-time role held on until an arbitrary date.
Deciding when the ramp ends
The risk specific to a gradual exit is that it never completes. Without a defined endpoint, a reduced arrangement can extend indefinitely, and people find themselves still working at an age they had not intended because no particular moment presented itself as the one to stop.
Setting a date, or a condition, in advance addresses this. The condition can be financial — a portfolio level, a state pension starting — or it can be simply a date. What matters is that it exists, because in its absence the default is continuation and the default will win.
It is equally reasonable to decide that the ramp is the destination rather than a transition, and that some work continues indefinitely because it is wanted rather than needed. That is a legitimate outcome and it is different from drifting into it. The distinction is whether the arrangement was chosen, which is a question worth answering deliberately at some point rather than leaving to inertia. None of this is financial advice, and the rules governing all of it vary considerably by country.
The health assumption underneath the whole plan
Every phased arrangement assumes continued capacity to work, and that assumption becomes progressively less safe with age. A substantial proportion of people who intend to work into their late sixties stop earlier than planned, and health is the most common reason, followed by caring responsibilities for a partner or parent.
This matters because a plan that depends on several more years of earning is a plan with a single point of failure that gets more likely each year. The prudent structure treats continued work as an improvement to a plan that would survive without it, rather than as a component the plan requires.
Practically, that means the portfolio should be able to support a reduced but acceptable standard of living from the point of leaving full-time work, with continued earnings raising that rather than making it possible. It is a more demanding standard and it is the difference between a phased retirement that is a choice and one that turns out to have been a necessity.
What partners need to agree about first
Phased retirement in a household is rarely a decision by one person, and the coordination problem is larger than it appears. Two people rarely reach this stage at the same time, and the period where one has stopped and the other has not creates practical tensions that catch couples by surprise.
The financial half is straightforward to discuss: what income each will have, from when, and how shared costs will be met when the balance between the two earners changes substantially. The other half is harder and matters more: expectations about how time is spent, how the household division of labour changes, and what each person imagines the arrangement will look like.
The recurring finding from people who have been through it is that the assumptions each partner held were different and neither had said so. A conversation covering what a typical week is imagined to contain, held a year or two before anything changes, is one of the more useful hours available and one of the least likely to happen without deliberate scheduling.