Industry Reference and the Market Report

---

Two genres of industrial analysis exist. They look similar from the outside — both carry charts, footnotes, cited sources, and conclusions — but they answer different questions, and the distinction shapes what the reader can do with each.

Industry reference describes. How an industry is structured, how operating performance is distributed across its peer set, where its establishments physically sit, how those characteristics have moved over time. It is fundamentally backward- and present-looking, descriptive in voice, and reproducible from primary sources.

Market report predicts. What an industry will look like, what its winners and losers will be, what an investor or executive should think about it and do next. It is fundamentally forward-looking, prescriptive in voice, and built around analyst judgment.

The two genres do different things, and both have their uses. Market reports compress a sector view into a thesis — useful for opportunity identification, scenario framing, and asset allocation. Industry reference grounds a specific firm against its peer set — useful for diligence underwriting, board calibration, and operating self-assessment. The case for reference is not the case against reports; it is the case for the layer below them.

The content market for industrial decision-makers is currently dominated by the predictive genre and underserved by the descriptive one. Three reasons matter.

First, the evidence on long-range forecasting in complex domains is weak — and the market has not adjusted. The work of Philip Tetlock and Dan Gardner on expert prediction and on forecasting accuracy more broadly shows that confident, high-status experts forecasting beyond a short horizon perform barely better than chance, and that the most confident forecasters are systematically the worst-performing. Yet the market continues to reward confident prediction and undervalues descriptive context. The incentive structure favors the prescriptive genre over the descriptive one, even where the evidence base does not.

Second, the empirical literature on within-industry variation is well established but rarely carried into commercial reports. Chad Syverson's survey of productivity dispersion documents that operating outcomes within an industry are systematically more variable than between industries — the 90th-percentile firm in a four-digit NAICS code typically operates at one-and-a-half to two times the productivity of the 10th-percentile firm. Hortaçsu and Syverson, studying ready-mix concrete, show the same dispersion in unit economics. The dispersion is what diligence actually needs to underwrite against; single-point industry averages, which are the default unit of most commercial reports, suppress the distribution that contains the answer.

Third, the separation between industry structure and firm performance — fundamental in industrial economics — gets collapsed in most commercial analysis. Ellison and Glaeser on geographic concentration, Axtell on the Zipf distribution of US firm sizes, and the broader productivity literature all show that what an industry IS — its concentration shape, its fragmentation depth, its geographic dispersion — explains a substantial share of how the firms inside it perform. Yet most commercial reports fold structure into performance and treat both as a single industry-attractiveness question rather than as two distinct empirical layers that each deserve their own description.

A reference publication is what the commercial content market would have if the descriptive layer were as mature as the predictive one. It reports distributions instead of averages. It separates structural condition from operating performance from geographic footprint. It cites primary sources at their published vintages and discloses suppression rather than imputing through it. It does not forecast.

The methodological underpinning is federal statistical infrastructure. The U.S. Census Bureau's Economic Census, the BLS Quarterly Census of Employment and Wages, IRS Statistics of Income, and Census County Business Patterns publish the most durable, methodologically-disclosed, audit-trailed industrial data available in the United States. National Academies reviews of the federal statistical system document that these sources operate on multi-decade horizons under explicit methodological governance — a standard that commercial industry-research subscriptions, built on proprietary panels and analyst estimates, rarely meet.

Commercial market research is built around the analyst's view. Industry reference is built around the source's audit trail. Both have a place; the reference layer is the one currently underprovided.

Buy-and-build private equity is the most visible case. Hammer, Knauer, Pflücke, and Schwetzler (2017) document buy-and-build as a structurally distinct strategy whose success depends on selecting the right industry setting — fragmented at the firm level, with a deep bolt-on pipeline, with operating fundamentals stable enough to compare firms against. The selection question is a reference question: what does this industry look like, structurally, and where do its establishments sit? A market report that forecasts the industry's growth tells the buyer how to feel about a setting; a reference publication tells the buyer what the setting actually is. The first is interesting; the second is underwriteable.

The same logic applies outside private equity. Operators evaluating their own performance against a peer set need to know where in the peer-set distribution they sit, not what their industry average is. Boards reviewing operating budgets need the same. Advisors building case structure need a defensible external baseline rather than a thesis. None of these decisions is well-served by a forecast; all of them are well-served by a description.

A reference publication operates under a discipline of restraint. Interpretations are stated with their boundaries; observations are reported with their distributional context; suppressed data is shown as suppressed, not imputed away; locked editions are never amended silently. No edition forecasts; no edition recommends; no edition values a firm. A reference is durable because it does not try to be a forecast. That is not a marketing position — it is a structural one. The publisher does not gain analytical credibility by being more confident; it gains credibility by being more correct, more often, with a smaller surface of contestable claims.

Industry reference does not replace the market report. The two genres serve different decisions. But where the report dominates and the reference is missing, the decisions sitting on top of the report are being made on a substrate the report cannot actually see. The case for descriptive industry reference is the case for reading the substrate — for the diligence team, the operator, and anyone whose decisions assume the industry below them looks roughly like the analyst said it does.

This is the position behind the Industrial Patterns catalog. The reference modules — Operating Benchmarks (operating performance distributed across the peer set), Industry Structure Reference (structural condition across population, size, concentration, dynamics, and geography), and Add-On Density Atlas (county-level establishment geography and drift) — each apply this approach to a specific empirical question. The current catalog covers US Building Materials and US HVAC & Plumbing, with additional industries added as editions ship. Each edition is descriptive, federally-sourced, distributional, and locked at publication. Each edition can be audited; each observation can be questioned; every figure has a federal source and a stated vintage. The work does not seek to be persuasive. It seeks to be true, durable, and useful.

Industrial Patterns modules are built for the diligence process — see where industry benchmarks and structural references fit in a buy-and-build strategy in mid-market US manufacturing. 

Browse the editions catalog; review the full methodology; more on the publisher's posture on the About page.

Tetlock, P. E. (2005). Expert Political Judgment: How Good Is It? How Can We Know? Princeton University Press.

Tetlock, P. E., & Gardner, D. (2015). Superforecasting: The Art and Science of Prediction. Crown Publishers.

Syverson, C. (2011). What Determines Productivity? Journal of Economic Literature, 49(2), 326–365. https://doi.org/10.1257/jel.49.2.326

Hortaçsu, A., & Syverson, C. (2007). Cementing Relationships: Vertical Integration, Foreclosure, Productivity, and Prices. Journal of Political Economy, 115(2), 250–301. https://doi.org/10.1086/514347

Ellison, G., & Glaeser, E. L. (1997). Geographic Concentration in U.S. Manufacturing Industries: A Dartboard Approach. Journal of Political Economy, 105(5), 889–927. https://doi.org/10.1086/262098

Axtell, R. L. (2001). Zipf Distribution of U.S. Firm Sizes. Science, 293(5536), 1818–1820. https://doi.org/10.1126/science.1062081

Hammer, B., Knauer, A., Pflücke, M., & Schwetzler, B. (2017). Inorganic Growth Strategies and the Evolution of the Private Equity Business Model. Journal of Corporate Finance, 45, 31–63. https://doi.org/10.1016/j.jcorpfin.2017.04.006

U.S. Census Bureau. Economic Census. https://www.census.gov/programs-surveys/economic-census.html

U.S. Census Bureau. County Business Patterns. https://www.census.gov/programs-surveys/cbp.html

U.S. Bureau of Labor Statistics. Quarterly Census of Employment and Wages. https://www.bls.gov/cew/

U.S. Internal Revenue Service. Statistics of Income — Corporation Source Book. https://www.irs.gov/statistics/soi-tax-stats-corporation-source-book

National Academies, Committee on National Statistics. https://www.nationalacademies.org/our-work/committee-on-national-statistics