After Spirit: Fare Composition, Incumbent Pricing, and Capacity Replacement

Praxis Labs Economics Working Paper PE-001. Author version 1.0, dated 5 November 2026. This working paper has not been peer reviewed. Spirit Airlines ceased operations on May 2, 2026. Using monthly U.S. ticket data and operated capacity, we distinguish the loss of a cheap seller from incumbent pricing and seat replacement. On 109 exposed domestic airport pairs, the observed mean nonstop one-way ticket amount rose from $203.22 in January-February to $261.88 in June-July. An exact baseline-first decomposition assigns $40.41 of the $58.66 increase to a baseline Spirit-removal benchmark; this is accounting, not causal attribution. The original fixed-cell incumbent-fare association is 11.1%. An exploratory entropy-balanced comparison gives 5.7%, with substantial uncertainty and narrower population coverage. A within-carrier, route, direction and month comparison of short- and long-booking fares gives an imprecise 3.8% premium change. We supply a separate positive-semidefinite airport-score covariance and formal projection intervals allowing bounded counterfactual drift. Under airport dependence, the original fare interval includes zero when monthly post-baseline drift may be one quarter of the largest pre-baseline drift, accounting for uncertainty in both. Rivals' observed seat increases replace 27.2% and 39.9% of baseline Spirit capacity in June and July; a longer-history seasonal model estimates a 12.6-point total-capacity shortfall. The evidence documents fewer cheap options and incomplete short-run capacity replacement, while leaving the precise causal incumbent-fare response unresolved. It does not identify the welfare effect of the blocked merger. DJ Bracken directed the study, evaluated the evidence and results, and edited and approved the final manuscript. As sole author, he takes responsibility for its conclusions. OpenAI ChatGPT research agents provided substantial assistance with study design, source verification, data collection and cleaning, statistical programming and analysis, computational checks, interpretation, drafting, and figures. Funding: No funding was received for this research. Competing interests: The author declares no competing interests. Research data and participants: The research uses public aggregate airline records and no recruited participants or identifiable passenger records.