Why Assumptions Matter More Than Precision In Projections

Retirement projections are frequently presented to the dollar across thirty years. The precision is a property of the calculation, not of the knowledge behind it.

A handful of inputs drive nearly everything

Longevity, investment return, inflation and spending account for most of the variation in a projected outcome, and none of the four is known.

Everything else in a typical model, however carefully specified, moves the result far less than a modest change to any of those.

This is why two competent projections for the same household can differ substantially without either containing an error.

Small assumption changes compound over decades

A difference in assumed return that appears trivial in a single year is applied repeatedly, and the gap between two paths widens throughout the projection.

Inflation behaves the same way on the spending side, which is why a projection's later years are almost entirely a product of its assumptions rather than its starting balance.

Because the compounding runs in both directions, the range of defensible outcomes at year thirty is wide enough that a single figure misrepresents it.

Longevity is a distribution, not a date

Planning to an average life expectancy plans to a figure that roughly half of people will exceed, which is the wrong side of the distribution to be on.

Joint longevity in a couple extends further still, since the relevant horizon runs to the later of two lives rather than either one.

Projections that end at a fixed age therefore answer a narrower question than the one households are actually asking.

Sensitivity testing is more informative than refinement

Running a plan under deliberately worse assumptions shows which ones the outcome actually depends on, and that is usable information.

A plan that survives a lower return assumption but fails under higher care costs has told the household where its attention belongs.

This is a different exercise from adding detail, and it is generally more valuable than modeling a household's utility bills accurately.

Projections are reviewed, not solved

Because the inputs are estimates, a projection is a statement about the present rather than a forecast that stands for decades.

Revisiting it periodically with actual experience substituted for assumptions is what keeps it connected to the household's real position.

Building and interpreting such projections against a specific set of accounts, income sources and obligations is work for a qualified financial professional, and the rules affecting the inputs change over time.