Source
Pierson (1997), LIT-1372, read in full in NOTE-1076 from the EUI working paper that preceded the APSR article: Part III (pp. 23–36) for the claim, Part IV (pp. 36–41) for its implications. Page numbers are the working paper’s. The 2000 article was not read and may state the list of features differently.
What was actually shown
Nothing was tested. The account is argued in two steps.
- Prevalence. Arthur’s conditions for increasing returns (set-up costs, learning, coordination, adaptive expectations) are found in three features of politics: institutional density, collective action, and the complexity and opacity that make “mental maps” self-reinforcing. “Increasing Returns in fact characterize three of the most prominent features of political environments: processes of institutional development, processes of collective action, and processes of social interpretation” (p. 30).
- Intensity. The correctives that Liebowitz and Margolis rely on in markets are weaker in politics: competition and learning, long horizons supplied by firms and capital markets, and flexible authority within firms. Elected officials “employ a high discount rate” because voters’ decisions “are taken in the short-run” (p. 32), so they pick paths with early payoffs and rarely leave one whose exit costs come now. Institutions are made sticky on purpose, to bind successors (Moe) or oneself (credible commitment).
- What could have come out otherwise. Only the illustration: the Clinton plan (1993–94) and the Republican Medicare and Medicaid cuts (1995–96) both failed against interests grown up around existing policy and against veto points (pp. 35–36). Two opposite reforms failing is what the account predicts; one succeeding would have counted against it. It was chosen after the event.
What this does not say
- Not that politics is path dependent everywhere. “Not all aspects of political life are subject to increasing returns” (p. 41).
- Not that persistence shows path dependence. An arrangement that lasts because the interests behind it still hold is not locked in by early events. The account names mechanisms; it gives no test for telling the two apart in a case.
- Not a claim about which actor governs. Short horizons explain why paths are kept, not which component of a polity holds authority over another. That is CLAIM-423’s question, which Pierson names as open (“we know relatively little about the time horizons of different political actors”, p. 32) and does not answer.
- Not efficiency. It says inferior paths are more likely and harder to leave in politics. It does not say any given political path is inferior, and it calls the remediability test too demanding for explanation.
Connections
- THEORY-429 (North). North’s increasing returns to the institutional matrix is this account’s premise; this account adds why politics has more of it and holds it harder.
- THEORY-863 (Arthur, Active). The model behind “lock-in” and the “Potential Path-Inefficiency” Pierson lists. Pierson uses its properties as analogies, not as a model of any political process.
- CLAIM-184, CLAIM-277. The record’s own entrenchment claims (anchors, early influence) are instances of the same feedback in an organization.
- CLAIM-423. The horizon half of that claim has its nearest source here; the authority half does not.
Note of 2026-10-30: what holds it back
No comparison in the record measures how often and how cheaply matched political and market arrangements are reversed. Pierson argues the contrast, with one illustration chosen after the event, and this reading has only his 1997 working paper (NOTE-1076).