What Is a Goal-Date Glide Path and When Should Risk Fall?

Ramesh notices that a bad year would now be different

Ramesh can see his retirement date. His portfolio has done well, and he has held his allocation through two frightening stretches without selling, which he is reasonably proud of.

Something has changed, though, and he has half-noticed it. A bad few years used to be an inconvenience. He would have watched the value fall, kept contributing, and waited — and the waiting worked, because there was a decade of contributions still to come and a decade of recovery available to him. A bad few years now would be a different thing entirely. There is no decade left, and the money is going to start coming out rather than going in.

The risk he can afford has fallen even though his tolerance for it has not. That is the gap a glide path exists to close.

What a glide path is

A glide path is a schedule for lowering the risk in a portfolio as the date the money is needed approaches. Not a rule for responding to markets — a rule for responding to the calendar.

The distinction that matters most is between this and rebalancing, because the two get run together constantly. Rebalancing restores today's target after markets have pushed the portfolio away from it; it assumes the target is still correct. A glide path changes the target, on purpose, because the goal is closer than it was. Ramesh needs both, and needs to know which he is doing at any given moment.

The reason to write it as a schedule rather than to handle it as it comes is that "as it comes" means "when it feels right", and the moments that feel right for reducing risk are systematically the wrong ones. Nobody feels like de-risking after three good years. Everybody feels like it after a bad quarter, which is when the reduction is most expensive.

Why the money's date sets the risk

The logic underneath is short and worth having in Ramesh's head rather than on a page.

An asset that can fall a long way is acceptable when there is time for it to come back and no requirement to sell in the meantime. Both halves matter. Time alone is not enough if a payment falls due in the middle of the fall, because a forced sale converts a temporary decline into a permanent one. That is why the relevant clock is not "when do I retire" but "when does each rupee get spent", which for Ramesh is not a date at all.

His retirement is a stream. The money funding his first year out is needed almost immediately; the money funding his twentieth is thirty years away and should still be growing. Treating the whole portfolio as though it matures on his last day at work would leave him far too conservative for far too long, and would cost him the growth that has to carry a retirement that may run for decades.

So the glide path applies to the near payments, not to everything. This is the same insight the retirement bucket strategy organises more formally, and either framing gets to the same place.

What has to be written down

A glide path that lives in someone's intentions is not a glide path. The test is whether the following could be handed to a spouse and executed without him.

It needs the starting allocation and the date it applies from, and the target allocation at each step along the way, with the dates. It needs to say which assets the money moves into as it comes out of growth, because "less risky" is not an instruction. It needs to say whether each step is implemented by redirecting contributions or by selling, since the first is cheaper and slower and the second is neither. And it needs a rule for what happens if a market fall lands in the middle of a scheduled reduction — which is the part everyone omits and the only part that will actually be tested.

The point of writing it down is not tidiness. It is that the decision gets made now, by the version of Ramesh who is calm, rather than in eighteen months by the version reading the news.

The step everyone gets wrong

The temptation, when a scheduled reduction arrives and markets have just fallen, is to postpone it. Sell later, when things have recovered. It sounds like prudence.

It is a forecast, and a specific one: that the market will be higher at some unnamed future date than it is now, and that Ramesh will correctly identify that date. If he had that ability, the glide path would be the least of what he should be doing with it. The evidence on rules of this kind is covered in why market timing is unreliable, and it is not encouraging.

The honest way to hold this is that a gradual, dated reduction will sometimes happen just before further gains he then misses, and sometimes just after losses he wishes he had avoided. Those are not failures of the path; they are the two halves of what "not forecasting" costs, and they roughly cancel. What does not cancel is the outcome the path exists to prevent — arriving at the date needing the money, with the money down a third, and no time.

What a glide path cannot fix

One warning, because de-risking is often reached for as a solution to the wrong problem.

Lowering risk near a goal protects the amount that has been accumulated. It does nothing whatever about whether that amount is enough. If Ramesh is short, moving to safer assets locks in the shortfall rather than closing it, and the levers available are the ones that were always available: contribute more, spend less later, work longer, or change the goal. A glide path is not one of them.

The right sequence is therefore to check adequacy first and de-risk second. Someone who discovers a gap five years out has genuinely bad options, but they are better than the options of someone who discovers the same gap having de-risked into them.

What to take away

The risk a portfolio should carry depends on when its money will be spent, and that changes even when nothing else does. A glide path is the decision to lower risk on a schedule, made in advance, because the alternative is making it in response to whatever the market has just done.

Write it as dates and allocations rather than intentions. Apply it to the payments that are close rather than to the whole portfolio, since money needed twenty years into retirement still needs to grow. Decide now what happens if a fall coincides with a scheduled step. And check that the goal is funded before making it safe, because safety applied to a shortfall just makes the shortfall certain.