Cataloging Growth: A Re

Cataloging Growth: A Re-Evaluation of 1900–1990
From Verónica Bäcker-Peral and Benjamin Wittenbrink of MIT:
Measuring real GDP growth requires distinguishing changes in prices from changes in product quality and composition, yet systematic quality adjustment of price indexes is unavailable for much of the twentieth century. We construct a new quality-adjusted price index for U.S. consumer goods using 5.1 million product listings from Sears catalogs, 1900–1990. We use large language models to extract product information and estimate hedonic price schedules from high-dimensional text embeddings, allowing us to infer annual changes in the cost of living. The resulting cost-of-living index implies substantially lower goods inflation than conventional deflators, and consequently implies much faster real economic growth: between 1900 and 1990, real goods consumption grew by a factor of 39 using our index, compared with a factor of 10.3 using standard goods deflators. The gap between our index and canonical ones is largest before World War II, reversing the conventional view that goods consumption growth was slower before 1945 than in the post-war decades.
Here is a useful tweet storm on the paper, important work.

Facts Only

* Researchers constructed a quality-adjusted price index for U.S. consumer goods from 1900–1990.
* The index utilized 5.1 million product listings from Sears catalogs.
* Large language models were used to extract product information and estimate hedonic price schedules from text embeddings.
* The resulting cost-of-living index implies lower goods inflation than conventional deflators.
* Real goods consumption grew by a factor of 39 using the new index between 1900 and 1990.
* Real goods consumption grew by a factor of 10.3 using standard goods deflators between 1900 and 1990.
* The gap between the index and canonical ones was largest before World War II.

Executive Summary

Researchers constructed a quality-adjusted price index for U.S. consumer goods spanning 1900 to 1990, utilizing 5.1 million product listings from Sears catalogs. Large language models were used to extract product information and estimate hedonic price schedules from these text embeddings to infer annual changes in the cost of living. This index suggests lower goods inflation compared to conventional deflators. The index implies significantly faster real economic growth between 1900 and 1990; real goods consumption grew by a factor of 39 using the new index, versus a factor of 10.3 using standard deflators. The divergence between the two indices is greatest before World War II, which reverses the conventional view that goods consumption growth was slower prior to 1945 than in post-war decades.

Full Take

The methodology introduces a fundamental challenge to conventional measures of economic growth by separating price changes from changes in product quality and composition, a factor traditionally ignored in standard deflators. The core implication is that standard metrics underestimate real economic expansion, particularly regarding goods consumption dynamics over the entire 20th century. The finding that the divergence is greatest pre-WWII suggests that historical macroeconomic narratives about growth trajectories are structurally dependent on the chosen inflation adjustment method. This shifts the focus from simple inflation tracking to valuing actual consumption change as a reflection of real economic velocity. The pattern points toward a systematic omission in historical data processing, suggesting that perceived economic reality is shaped by the tools used for measurement, which has profound implications for understanding relative economic phases like the pre- and post-war periods. What if the discrepancy itself is a signal about how societal values surrounding quality versus quantity have evolved?

Sentinel — Likely Human

Confidence

This text reads like a summary or introduction to a research paper, characterized by precise technical language and an embedded sense of the authors' interpretive significance, indicating human authorship.

Signals Detected

Sentence length variance is somewhat controlled; vocabulary is precise but the structure flows like academic exposition.

The argument flows logically from the problem (lack of quality adjustment) to the method (using LLMs on Sears data) to the conclusion (faster growth). It possesses an idiosyncratic emphasis when discussing the specific historical reversal.

The presentation of the quantitative results is direct and cites specific factors (39 vs 10.3), suggesting a human grounding in the data, even if the methodology uses advanced tools.

The core claims about the results and the context of GDP measurement appear specific and plausible within academic discourse, suggesting actual research output rather than pure fabrication.

Human Indicators

The presence of a direct, self-referential call to action ('Here is a useful tweet storm on the paper, important work.') suggests an author engaging directly with an academic audience in a way that mimics human engagement.

The specific juxtaposition of complex econometric concepts (deflators, hedonic schedules) with accessible narrative framing points toward expertise translation.