The retirement wall: the 1997 build boom meets its 30-year exams with replacement running at a fifth of the old pace

Those hulls are now 25 to 30 years old. That is the age band where the internal structural exam decides whether another reskin pencils or the barge goes to scrap.

Replacement is not keeping up. Completions in 2019–2021 averaged 350 dry barges a year, a fifth of the 1997 peak. Steel plate at $1,120 per net ton in 2026 (Leeco Steel price series) has held orders down since.

Tank barges: 82% of the ≥30,000-bbl class postdates 2001; the exam window lands on the 10–20,000-bbl workhorses

The tank fleet tells the same story one boom later. 2,107 of its 5,811 hulls (36%) predate 2002. Its own boom ran 2013–2015 at 328 barges a year, then faded to 135 a year by 2019–2021. Years like 2023, at 26 units, are replacement for a fleet a hundred times that size.

For operators, the wall is pricing power with a date on it. Every exam season from now through roughly 2031 forces a scrap-or-reskin decision on a 1996–2001 hull. Each hull that leaves tightens supply into a market already running 57–71% above seasonal rates (USDA weekly barge rates against our seasonal baseline, week of Jul 14, 2026).

For insurers it is a severity schedule: an aging cohort works harder in low water precisely as its steel thins.

For investors it frames the arithmetic behind used-equipment prices. Fleets cannot be rebuilt at today's steel costs at anything like the pace they will retire.

The caveat cuts one way. If rates hold, more exams end in reskins, and the wall erodes more slowly than the age math implies.

The census, not the calendar, settles it. Watch the 1996–2001 cohort's survival rate in the next USACE vessel census; that count is what turns an age distribution into a supply forecast.