Retirement Spending Smile

David Blanchett documented the pattern in 2014, finding that real retirement spending typically declines through the middle years before rising late, rather than holding flat as most planning models assume. Discretionary categories, travel and dining in particular, account for most of the early decline.

The late rise is driven by healthcare and long-term care, and it has a different character from the early spending. It is largely non-discretionary and the distribution is highly skewed, so an average understates the risk for the minority of households that need extended care.

The distinction matters for projections. A model assuming flat inflation-adjusted spending for thirty years overstates the middle decades and can understate the last, and the two errors do not cancel because they differ in both size and certainty.