---
title: "Does Wright's law trace cleanly to T.P. Wright's 1936 paper, and do the New Things Under the Sun / Our World in Data restatements match the original mechanism?"
type: "question"
status: "answered"
date_raised: "2026-07-12T00:00:00.000Z"
tags: ["economics","wrights-law","learning-curve","sourcing"]
answered_log: ["2026-09-05: Answered by [[claim-wright-1936-paper-verified-real-and-accurately-cited]] and [[claim-wright-1936-defines-cost-as-power-law-of-cumulative-quantity]] — Wright's 1936 paper was located, fetched directly (extract_pdf), and read in full: it is real, correctly cited by Our World in Data down to the page range, and its own formula matches the core mechanism (log-log-straight-line cost decline against cumulative production quantity, not calendar time). Two further findings sharpen rather than reopen the answer: [[claim-wrights-eighty-percent-curve-is-lot-average-cost-not-unit-cost]] (Wright's percentage is a whole-lot average, not a per-unit cost, unlike either restatement's framing) and [[claim-wright-1936-reports-four-distinct-curves-not-flat-eighty-percent]] (Wright reported four curves, not one flat 80%). The Boston Consulting Group's studies, named in this question as a second candidate primary, were not located or fetched this session — a further lead, not a blocking gap now that the Wright 1936 original itself has been read directly."]
progress_log: ["2026-07-15: Partially answered. [[claim-sahals-identity-equates-wrights-law-and-moores-law]] reads Lafond et al. 2017 (arXiv:1703.05979) directly — a Tier 1 primary, quote verified verbatim — which empirically tests the progress-ratio mechanism and cites Sahal (1979) for the Wright's-law/Moore's-law equivalence. Still not the demanded T.P. Wright 1936 original, so the question stays open for that specific paper; the mechanism itself now has a Tier 1 empirical anchor rather than resting only on secondary explainers."]
---


[[claim-wrights-law-cost-falls-per-cumulative-production-doubling]] states the mechanism — unit cost falls a constant fraction per doubling of cumulative production — sourced to two secondary explainers (New Things Under the Sun; Our World in Data's learning-curve page). Neither is the originating economic literature. Per the vault's sourcing floor, a specific technical-mechanism claim needs a Tier 1–2 primary.

**What's needed:** T.P. Wright's 1936 paper "Factors Affecting the Cost of Airplanes" (*Journal of the Aeronautical Sciences*), which is generally credited as the origin of the cost-per-doubling relationship in aircraft manufacturing, and/or the Boston Consulting Group's 1960s–70s experience-curve studies that generalized it to price. Confirm the progress-ratio mechanism as described in the secondary sources matches the original formulation (constant percentage cost reduction per doubling of *cumulative* production, not calendar time).

**Candidate next moves:**
- Locate Wright (1936) directly or via a reliable reprint/citation chain; pull the exact formulation and progress-ratio example.
- Check whether BCG's original experience-curve publications are accessible, or whether a reputable secondary (e.g. a peer-reviewed economics paper citing both) is the best available Tier 1–2 anchor.
- Once resolved, clear the `[unverified-mechanism]` flag on [[claim-wrights-law-cost-falls-per-cumulative-production-doubling]] and reconsider whether it and [[claim-cheaper-extraction-disruptions-fall-monotonically-not-hold-then-collapse]] can move past `seedling`.


## Progress log

- 2026-09-05: Answered by [[claim-wright-1936-paper-verified-real-and-accurately-cited]] and [[claim-wright-1936-defines-cost-as-power-law-of-cumulative-quantity]] — Wright's 1936 paper was located, fetched directly (extract_pdf), and read in full: it is real, correctly cited by Our World in Data down to the page range, and its own formula matches the core mechanism (log-log-straight-line cost decline against cumulative production quantity, not calendar time). Two further findings sharpen rather than reopen the answer: [[claim-wrights-eighty-percent-curve-is-lot-average-cost-not-unit-cost]] (Wright's percentage is a whole-lot average, not a per-unit cost, unlike either restatement's framing) and [[claim-wright-1936-reports-four-distinct-curves-not-flat-eighty-percent]] (Wright reported four curves, not one flat 80%). The Boston Consulting Group's studies, named in this question as a second candidate primary, were not located or fetched this session — a further lead, not a blocking gap now that the Wright 1936 original itself has been read directly.