The Glide Path Nobody Reads

Target-date funds: same retirement year, different fund, different portfolio.

Twelve fund families, fifteen series, 54 holdings filings read from the regulator and 13 prospectuses read for the terms. Every equity figure below is measured from a fund's own filing and looked through to what its underlying funds hold.

The finding

At the year on the label, sixteen funds named 2025 hold between 56.57% and 39.79% in equities. That is 16.78 points of difference between two products a saver would reasonably treat as the same thing. On a $100,000 balance it is $16,780 of stock-market exposure.

Providers agree about the distant future and disagree about retirement itself. Thirty-four years out, eighteen funds named 2060 sit within 8.69 points of each other. At the target date the spread is 1.9 times that.

And it widens after the date. Eleven years past their target year, seven surviving funds named 2015 are 20.00 points apart — 48.60% against 28.60%. The dispersion is smallest where the money is furthest away and largest where it is being drawn on.

Why it happens

A target-date fund either stops de-risking at its target date or keeps going past it. Of fifteen series read from their prospectuses, twelve keep going, for between five and thirty years; two stop at the date; and one retired its dated fund before the year on it arrived — a fund holding $6,683,077,772.04 at its final filing, dissolved about fifteen months early.

That choice governs the twenty or thirty years after retirement, which is most of the time the money is invested. It is disclosed in the prospectus. It is not in the fund's name, and no provider in this set puts it on the fund's summary line.

Where each path stops varies as much as how long it runs. Five providers state a landing equity as a number, and they span 25.00% to 40.00%. The measured landing funds span 22.04% to 45.88%.

The three numbers

What the year in the name actually buys

Not a portfolio. Across fifteen families a mean 94.03% of the 2060 fund sits in exactly the underlying lines the 2025 fund also holds, and ten of the fifteen are above 99%. One family's 2060 portfolio shares all 83 of its lines with its 2025 portfolio.

The far-dated fund is the near-dated fund with the equity dial turned up. What the year selects is the weighting. What it does not select is the shopping list, the de-risking schedule, or the destination.

Two things worth knowing about the data itself

The regulator's own file cannot answer this question. Read at face value, its asset category reports these funds as 97.8% equity at every vintage, because a share of a bond mutual fund is legally an equity security. Every figure above required looking through to what the underlying funds hold.

Two competent parties publish equity weights for the same fund 7.98 points apart, because one files a global real-estate equity fund under "Alternatives". A reader comparing providers on their published allocation pages is not comparing like with like, and nothing on the page says so.

What we cannot tell you

• What these funds will do next. The 2015 reading is evidence about how far apart same-year funds sit a decade past target. It is not a projection of the 2025 funds, and none is made here.
• How much equity a fund holds through futures. Derivatives are carried at market value rather than notional, so a fund taking exposure that way is understated here, and the understatement is not quantified.
• Whether one provider's revised path is already in the filing we measured. A supplement raising equity thirty years past the target date was filed while this research ran; whether the measured filing reflects it was not established, and it is recorded as unverified.

What this report corrected in its own work

The first measurement reported a spread at the target date of 29.55 points. Most of that was a classification defect rather than dispersion: a core bond fund was being counted as equity, which alone moved one fund by 14.15 points. The corrected figure is 16.78 points, and it is in every table. The research packs record eleven further errors found in the first pass and four in the second, plus nine holes found in the verification scripts themselves.

2,022 automated checks pass; 84 planted defects were caught, all of them.

Disclosures. The author holds no position in any fund or security named here. No issuer payment, no advertising. Impersonal research, identical for every purchaser, tailored to no one. No statement about what any reader should hold, no recommended allocation, no forecast. Finnacl Research is not an investment adviser.