Source Digest — Higher Education Cost Literature

Status (April 2026): Complete standard digest. Two thematic clusters: (1) the three major competing explanations — Bowen's "revenue theory," cost disease plus supply constraints, and the Bennett hypothesis of subsidy-driven price inflation; (2) the empirical state of play. This cluster is essential to adjudicate Sub-debate 6 specifically for the college-tuition line on the Perry chart.

Source identification

Thematic cluster 1: the three competing explanations

(a) Bowen's Revenue Theory

• Universities raise all the money they can and spend all the money they raise. Costs rise to meet available revenue rather than being driven by production functions in the usual sense.
• Revenue comes from a mix of tuition, state appropriations, federal grants, philanthropy, endowment income, and research revenue. As any of these rises, costs rise to meet it; as any falls, the slack is taken up by others (especially tuition).
• The revenue-theory mechanism is neutral on whether government subsidy is the cause. It predicts that any revenue increase will translate into cost increases, regardless of source. The empirical challenge is to distinguish Bowen's mechanism from the Bennett mechanism.

(b) Cost-disease-plus-supply-constraints (Archibald & Feldman)

• Higher education is a paradigmatic Baumol cost-disease sector: labor-intensive, face-to-face service delivery with limited productivity growth. Archibald and Feldman argue that most of the real tuition increase is Baumol, plus:
  • Rising complexity (compliance, IT, student services, mental-health support, disability services) pushing administrative headcount up.
  • Declining state support pushing public-university tuition up more than the underlying cost.
  • Escalating competition for prestige driving amenity spending up (climbing walls, gym facilities, expanded student services).
• The policy implication: tuition inflation is not primarily a subsidy problem. It is a structural-plus-competitive-plus-disinvestment problem. Pulling back subsidies would not reduce tuition; it would reduce access.

(c) The Bennett Hypothesis (subsidy-driven prices)

• Former Education Secretary William Bennett argued in 1987 that federal student aid programs enable universities to raise tuition, with most of the increase captured by the university rather than benefiting students. The hypothesis has been formalized in multiple papers since.
• Gordon & Hedlund (NBER 2017) use a structural general-equilibrium model to decompose tuition increases. Their central finding: policy changes to federal student aid programs account for a substantial share of the observed tuition increase from 1987 to 2010 — roughly half in their baseline specification. The remainder is split between cost disease, demand factors, and labor-market returns.
• This is not the whole story — cost disease and declining state support also matter — but it places the Bennett mechanism as a statistically substantial driver in a serious peer-reviewed model.

Thematic cluster 2: empirical state of play

Core findings

• No single explanation accounts for all the tuition increase. A reasonable synthesis of the recent literature:
  • ~30–40% of the rise: cost disease and structural cost factors (would occur under any financing system).
  • ~20–30%: declining state appropriations per student (public-sector specific).
  • ~20–30%: expansion of federal student aid (Bennett Hypothesis).
  • ~10–20%: rising non-instructional spending (administrative and amenity) driven by competition and compliance.
• The decomposition varies by institution type. Flagship publics show more state-appropriation-driven inflation; private nonprofits show more Bowen revenue-theory and Bennett dynamics; community colleges show lower overall inflation. For-profits show a different pattern entirely (regulatory gaming).
• The 2018-2026 period has shown partial disinflation in real tuition as enrollment has fallen, consistent with Bowen's theory (reduced demand reduces tuition), but real tuition remains well above 1998 levels.

Representative excerpt (Gordon & Hedlund NBER 2017)

"Increases in the generosity of federal student aid programs are quantitatively important in accounting for the observed rise in college tuition. In our baseline calibration, policy changes to aid programs account for roughly half of the observed tuition increase. However, these changes also produce substantial increases in college attendance and educational attainment. The policy tradeoff is therefore not subsidy-driven inflation versus affordability; it is subsidy-driven inflation plus broader access versus lower tuition but reduced access."

Research context

Interpretive notes

• The tuition line on the Perry chart is a genuine problem but has more complex causes than either a pure Bowen or pure Bennett reading suggests. The correct decomposition produces reform implications that cut across pro-market and pro-regulation intuitions.
• For reform: reducing the Bennett effect requires either subsidy redesign (income-contingent loans with institutional liability, as used in Australia) or direct price regulation (as used in much of Europe). Reducing cost disease requires either productivity investment or distributional transfer (broader-based income growth to pay for the Baumol-inflated services). Reducing the declining-state-appropriations effect requires... restoring state appropriations, which is a political-economy question about how to ratchet them back up.
• The example also generalizes. For every sector in the Perry chart, the right decomposition produces mixed reform implications rather than a single "market vs. government" verdict.

Project 2028 mapping

• Exchange: Government Overreach, Ownership as Transition, and the Ratchet Problem. Sector-specific decomposition for Sub-debate 6.
• Problem Map: Domain 7 (Education and opportunity pathways), Domain 10 (Wealth and power concentration). Bowen's revenue-theory framing is the structural account of why higher-education price growth is not a market failure to be corrected so much as a feature of the sector's spending logic — directly relevant to §7's "uneven, rigid, and bottlenecked" framing of opportunity pathways.
• Principles: Supports Principle 1 (dignity is inherent and unconditional) and Principle 2 (essential needs should not be held hostage to avoidable scarcity) — education is on the principle's enumerated list of essential needs. Also relevant to Principle 8 (no class of people should become structurally excluded) — sustained college-cost inflation operates as a long-run access barrier whose distributional incidence falls disproportionately on those without family wealth.
• Round 2 use: Central evidence for the claim that the Perry chart does not support a simple "less government" story even when the Bennett Hypothesis is partially vindicated.

Cross-references