Research and Development in the Social Sciences and Humanities (NAICS 541720)

A Histometrics industry primer for public- and private-market investors

1. Overview

This industry is the business of producing knowledge about people, institutions, and culture — economics, sociology, psychology, political science, education, public-health behavior, linguistics, history, and related fields. Its output is not a product but evidence: program evaluations, survey datasets, policy analyses, cost-benefit studies, behavioral research, and expert testimony, sold under contracts and grants [4].

Why it matters to an investor: this is a labor-and-knowledge services business, not a factory or a balance sheet of physical assets. Firms make money by hiring credentialed researchers (often PhDs) and keeping them billable on funded projects — the same economics as engineering or management consulting. Demand is unusually policy-driven: the single largest buyer is the U.S. federal government, which makes the sector steady in normal times but sharply exposed to shifts in the federal budget and administration priorities. That exposure is the dominant story right now (Sections 6, 9, 10).

Ways in differ by market. There is no pure-play public company dedicated solely to social-science and humanities research. Listed exposure is indirect, through diversified government-and-commercial consultancies such as ICF International and Charles River Associates (Section 4). The field's marquee names — RAND, NORC, Westat, Mathematica, Abt, the American Institutes for Research, the Urban Institute — are nonprofits or closely held / employee-owned firms that are not directly investable as equities. Private investors reach the industry through buyouts, growth equity, private credit, or partnership, not ordinary shares (Section 10). The underlying case rests on trusted methods, specialized talent, proprietary data, and recurring institutional relationships.

2. What it is and how it's structured

In scope: establishments primarily conducting research and analysis in the social sciences and humanities — cognitive development, sociology, psychology, language and behavior, economics, and related fields [4]. In practice the work clusters into a few business lines: government program evaluation, large-scale survey research and data collection, applied and litigation economics, think-tank policy analysis, and academic-style humanities scholarship. Typical tasks include survey design and fieldwork, longitudinal studies, causal-inference and statistical analysis, evidence synthesis, and data management.

What it excludes (this matters for sizing and for picking comparables) [4]:

• Marketing research and public-opinion polling → NAICS 541910. Commercial opinion polls and consumer market research are a separate, larger industry.
• R&D in the physical, engineering, and life sciences → NAICS 541715 (with 541713 nanotechnology and 541714 biotechnology as their own codes). A nonprofit like RTI International does mostly life- and physical-sciences work and sits largely here, not in 541720.
• Management and marketing consulting (advice rather than research) → NAICS 541611 / 541613.
• University-based research (classified under education, NAICS 611310) and research performed inside government agencies. Both are enormous in the social sciences and fall outside the business statistics for 541720 (Section 3).

Ownership mix is distinctive. Unlike most 541-sector industries, this one is heavy with nonprofits and employee-owned firms: independent research institutes (RAND, NORC, the Urban Institute, MDRC), university-affiliated centers, employee-stock-ownership-plan (ESOP) firms (Mathematica, Abt), and mission-driven contractors, alongside for-profit and private-equity-backed players. The federal statistics do not report a public-versus-private ownership split, but this tax-exempt-heavy structure shapes both the economics (Section 5) and who can invest (Section 10).

3. How big it is

Our ground-truth federal figures for NAICS 541720. County Business Patterns (CBP) data are for 2023; Economic Census receipts and concentration data are for 2022 — so these are not a single-year income statement.

An establishment is a physical operating location; a firm may own several. The average establishment is small — about 22 employees (46,634 ÷ 2,082) — and average receipts per firm are roughly $5.2 million ($9.30B ÷ 1,782). Those averages hide a bimodal structure: a dozen-plus large multi-program institutes plus a long tail of small and solo research shops (Section 8). The federal file provides no industry-wide figures for operating margins, utilization, backlog, contract mix, or client concentration — so we do not state them.

The undercount caveat is large here. The $9.30 billion receipts figure counts only for-profit and nonprofit business establishments classified in 541720. It systematically misses the three biggest venues for this research: (a) universities, where most federally funded behavioral and social research is actually performed (classified under education, not 541720); (b) government agencies doing research in-house (Census, the Bureau of Labor Statistics, USDA's Economic Research Service, agency evaluation offices), which CBP excludes; and (c) the smallest operators — independent scholars and nonemployer businesses captured only in separate Nonemployer Statistics [5]. For scale, federal research obligations for psychology and the social sciences alone have run on the order of $7–8 billion in a single recent year [6] — comparable to the entire measured industry's receipts, and that is just the federal slice of one funding stream. The true footprint of U.S. social-science and humanities research is several times the 541720 business-statistics number.

4. The investable universe

Public companies — indirect exposure only. No listed company is a pure social-science / humanities research play. The closest proxies are diversified consultancies and research-data firms where such work is one line among many:

Treat ICFI and CRAI as the primary listed reference points; both derive most revenue outside 541720, and Ipsos/YouGov are mostly adjacent polling and market research.

Private / nonprofit leaders — the real center of gravity (not directly investable):

Nine of these policy-research leaders together employ more than 20,000 people [16] — a workforce roughly half the size of the entire measured 541720 industry, underscoring the classification/undercount issues in Section 3.

5. How the money works

Revenue is almost entirely project-based, won as contracts and grants rather than sold off a shelf — from federal agencies, state and local governments, foundations, universities, and corporate research budgets. Three contract types drive the economics:

• Cost-reimbursement (e.g., cost-plus-fixed-fee): the client pays actual costs plus a set fee. Common for open-ended research; margins are thin and predictable.
• Fixed-price: the firm bears cost overruns and keeps savings — higher risk, higher potential margin.
• Time-and-materials: billed at negotiated hourly rates.

The core mechanic is billing direct labor and recovering overhead. A researcher's hours bill directly to a project; the firm then adds negotiated indirect cost rates to recover expenses not tied to one project — grouped into fringe-benefit, overhead, and general-and-administrative (G&A) pools [20]. For nonprofits and universities the equivalent is the facilities-and-administrative (F&A) rate. Profit (for-profits) or operating surplus (nonprofits) is the fee or margin layered on top. The main non-labor costs are survey panels, subcontracted fieldwork, cloud infrastructure, cybersecurity, compliance, and data licensing.

The levers owners actually watch:

• Billable utilization — the share of staff time on funded projects. Idle PhDs are the main way this business loses money.
• Indirect-rate competitiveness — leaner overhead wins more competitive bids and protects margin on cost-reimbursement work. This is why the federal push to cap indirect rates (Section 9) is existential for some firms.
• Backlog and book-to-bill — new awards versus revenue recognized. ICF reported a book-to-bill of 1.24 in 2024 ($1.24 of new awards per $1 of revenue), a growth signal [9].
• Win rate, renewal rate, and realized bill rates — reputation and past-performance record drive re-competes.
• Contract/client concentration, fixed-price exposure, and government payment timing — the working-capital and risk profile.

Margins tell the split. ICF's diversified government/commercial model runs roughly an 11% adjusted-EBITDA margin [9], while CRA's litigation-and-economics work — more commercial, less cost-reimbursement — carries higher margins and grew revenue to a record $687M in 2024, with a 2025 guide of $715–735M [10]. The pure-government, cost-reimbursement end of the field is a low-margin, steady, relationship-driven business; the commercial/litigation end is higher-margin and cyclical with the economy rather than the federal budget.

6. What drives demand

Demand tracks who is buying evidence and how much they will pay for it:

• Federal research budgets. The National Science Foundation's (NSF) Directorate for Social, Behavioral and Economic Sciences (SBE) — the largest single funder of U.S. university-based behavioral-science research — was funded at about $290 million in fiscal 2024 [7]. Add the Institute of Education Sciences (IES) for education research [29], the National Institutes of Health's behavioral and health-services research, Department of Health and Human Services evaluation offices, and the National Endowment for the Humanities (NEH) for humanities scholarship.
• Evidence-based-policy mandates. The Foundations for Evidence-Based Policymaking Act (the "Evidence Act") requires agencies to build evaluation capacity and evidence-building plans, creating baseline, recurring demand for independent studies [24].
• State and local governments commissioning program evaluations and accountability studies.
• Private foundations and international donors — large health, education, and poverty foundations, plus multilateral institutions funding measurable social outcomes.
• Commercial and litigation demand — antitrust, damages, and regulatory economics — which tracks corporate deal and dispute activity rather than the federal budget. This is the resilient, counter-cyclical slice (and the one CRAI is levered to).

Because government dominates, aggregate demand is politically cyclical: it rises with discretionary research appropriations and evidence mandates, and falls fast when an administration deprioritizes a field. That cyclicality turned into an outright shock in 2025 (Section 9).

Forward-looking judgment: artificial intelligence (AI) will automate some literature reviews, coding, transcription, and routine reporting. That should raise the premium on the things AI cannot fake — high-quality samples, credible causal inference, human validation, data provenance, and defensible methods — while commoditizing generic analysis (a two-sided force revisited in Section 9).

7. Regulation

There is no product regulator here. The binding rules govern contracting, research ethics, and data:

• Federal contracting/cost rules. The Federal Acquisition Regulation (FAR) Part 31 governs allowable costs; the Office of Management and Budget (OMB) Uniform Guidance (2 CFR 200) governs indirect-cost rates on grants; cognizant agencies and the Defense Contract Audit Agency audit those rates [20]. How a firm's indirect rate is negotiated and audited directly sets its recoverable overhead.
• Human-subjects protection. Research on people is governed by the Common Rule (45 CFR 46), requiring Institutional Review Board (IRB) review, informed consent, and added protections for vulnerable populations [22].
• Data privacy and confidentiality. FERPA (education records), HIPAA (health data), and CIPSEA (the Confidential Information Protection and Statistical Efficiency Act, for federal statistical data), plus Paperwork Reduction Act (PRA) clearance through OMB for federal surveys [23].

These rules are both a barrier to entry (compliance infrastructure, secure facilities, IRB standing, security clearances) and a source of ongoing policy risk — most acutely the indirect-rate fight (Section 9).

8. Competitive dynamics and consolidation

By our federal concentration data, this is a fragmented, unconcentrated industry [2]:

An HHI of 390 is well below the 1,500 threshold antitrust agencies treat as "unconcentrated," and a CR4 of just 30% confirms it: no handful of firms controls the market. The structure is bimodal — roughly a dozen large multi-program institutes at the top, plus a long tail of mid-sized specialists and single-researcher boutiques, many under the $28 million SBA small-business threshold [3] and eligible for small-business contract set-asides.

Competitive moats are reputational and relational: past-performance record, methodological credibility, representative samples and proprietary panels, secure-data facilities, federal contract vehicles, and incumbent relationships with contracting officers. Barriers are modest for basic analysis but high for national fieldwork and sensitive-data work.

Consolidation is real but gradual: private-equity roll-ups of government-services firms, rebrandings (Abt Associates → Abt Global), university-affiliated spin-outs (NORC), and capability acquisitions by diversified public consultancies. Most visible PE activity is in adjacent government consulting and commercial research — Bain Capital's ownership of Guidehouse (acquired from Veritas Capital in 2023) [27] and Kantar [28] illustrates the appeal of scale, recurring contracts, and proprietary data — and should not be read as direct measures of consolidation within 541720. The tax-exempt segment structurally limits classic M&A, because nonprofit research institutes cannot simply be bought.

9. Risks

• Federal budget and political cyclicality — acute now. In 2025, government-efficiency contract reviews terminated large volumes of research work: Mathematica and AIR each had 15 education-research (IES) agreements cancelled — among the most of any organizations — prompting layoffs at research firms [26]. The NEH cancelled more than 1,200 grants and moved to cut roughly two-thirds of its staff, with the administration proposing to eliminate the agency entirely [25]. The NSF SBE directorate faced a fiscal-2026 request slashing it to about $94 million from $290 million [7], with the Senate moving to protect it and the House to cut deeper — an unresolved appropriations fight [7][8].
• Indirect-rate cap risk. In February 2025 NIH tried to cap indirect (F&A) reimbursement at 15%, versus negotiated rates that had averaged around 58% [21]. Courts blocked it (a permanent injunction, upheld on appeal in January 2026), but the episode signals durable pressure to compress overhead recovery — a direct hit to the economics in Section 5.
• Buyer concentration. Heavy reliance on the federal government as the dominant customer amplifies every budget swing; dependence on a few large contracts or sponsoring agencies compounds it.
• Reputational / methodological risk. Credibility is the product. A flawed, non-reproducible, or contested study — or the perception of partisanship (several 2025 terminations targeted diversity-, equity-, and inclusion-related work) — can cost re-competes.
• AI-driven commoditization. The flip side of Section 6: routine research tasks that AI can do cheaply may compress prices for generic analysis and undifferentiated labor suppliers.
• Talent, data access, and execution. The business depends on a scarce PhD labor pool, on continued access to government administrative and survey data, and on clean fixed-price delivery; senior-researcher turnover and subcontractor failures hit quality and margin.
• Cybersecurity. Custody of sensitive personal and administrative data makes a breach an existential, contract-eligibility risk.

10. How to invest and the outlook

Public-market routes. There is no pure play. The practical way to get listed exposure is through diversified consultancies with meaningful social/economic-research practices — ICF International (ICFI) for government program evaluation and survey work, and Charles River Associates (CRAI) for litigation and regulatory economics [9][10]. Both trade as ordinary equities with the valuation and (for ICF) modest-dividend characteristics of professional-services firms. Because most of their revenue sits outside 541720, this is thematic, not pure, exposure — compare valuation multiples only after adjusting for the unrelated businesses. CRAI's commercial, non-federal mix makes it the more insulated of the two from the 2025 federal-budget shock; ICFI carries more direct federal-contract exposure.

Private-market routes. The heart of the industry — RAND, NORC, Westat, Mathematica, Abt, AIR, the Urban Institute — is nonprofit or closely held / employee-owned and not directly investable as equity; nonprofits and university affiliates are reached through grants, contracts, partnerships, or private credit, not shares. Private-capital equity exposure comes indirectly, via PE-backed government-services platforms that acquire evaluation and analytics capabilities. Underwriting there should center on recurring-contract and renewal behavior, proprietary panels/datasets/software, researcher retention and owner succession, fixed-price execution history, customer diversification, working-capital needs and government payment timing, and (for ESOP firms) employee-repurchase obligations.

Near-term outlook (forward-looking judgment). The dominant variable is the trajectory of federal research funding and the unresolved NSF appropriations split between House and Senate [7][8]. Expect the federal social-program-evaluation segment to stay under pressure through the current budget cycle, with continued contract-termination and re-compete risk, while the commercial and litigation-economics segment (CRAI-style) and state/local and foundation-funded work look more resilient — and are where the listed names have been growing. Over a longer horizon the structural demand for evidence — mandated program evaluation, data-driven policy, and dispute economics — remains intact; the swing factor is how much of it Washington chooses to fund in any given year. The strongest firms pair trusted methods, specialized talent, and recurring data assets with disciplined project economics; the weakest are generic labor suppliers whose broad-consultancy growth is mistaken for direct 541720 exposure.

---

Sources

1. U.S. Census Bureau, County Business Patterns (CBP), 2023 — establishments, employment, payroll, and Q1 payroll for NAICS 541720. https://www.census.gov/data/datasets/2023/econ/cbp/2023-cbp.html
2. U.S. Census Bureau, 2022 Economic Census — Concentration of Largest Firms — receipts, firm counts, and concentration ratios (CR4/CR8/CR20/CR50, HHI) for NAICS 541720. https://data.census.gov/table/ECNSIZE2022.EC2200SIZECONCEN
3. U.S. Small Business Administration, Table of Small Business Size Standards, 2023 ($28 million receipts threshold, NAICS 541720). https://www.sba.gov/document/support-table-size-standards
4. U.S. Census Bureau, 2022 NAICS Manual / NAICS definition, "541720 – Research and Development in the Social Sciences and Humanities" (scope and adjacent-code exclusions). https://www.census.gov/naics/reference_files_tools/2022_NAICS_Manual.pdf
5. U.S. Census Bureau, County Business Patterns: Coverage and Methodology and Nonemployer Statistics (employer-establishment coverage; exclusion of government employees and nonemployer businesses). https://www.census.gov/programs-surveys/cbp/technical-documentation/methodology.html
6. National Center for Science and Engineering Statistics (NCSES), NSF, "Federal Support for U.S. R&D," Science & Engineering Indicators (federal research obligations for psychology and social sciences). https://ncses.nsf.gov/pubs/nsb20257/federal-support-for-u-s-r-d
7. Federation of Associations in Behavioral & Brain Sciences (FABBS), "NSF Budget Protected in Senate, Slashed in House," 2025 (NSF SBE ~$290M FY2024; FY2026 request ~$94M). https://fabbs.org/news/2025/07/nsf-budget-protected-in-senate-slashed-in-house/
8. Congressional Research Service, "The National Science Foundation (NSF): FY2026 Appropriations and Funding History," report R48783, 2025. https://www.congress.gov/crs-product/R48783
9. ICF International, "ICF Reports Fourth Quarter and Full Year 2024 Results," Feb. 27, 2025 (FY2024 revenue ~$2.0B; adj. EBITDA margin ~11%; book-to-bill 1.24). https://icf.mediaroom.com/2025-02-27-ICF-Reports-Fourth-Quarter-and-Full-Year-2024-Results
10. Charles River Associates, "CRA Reports Fourth-Quarter and Full-Year 2024 Financial Results," Feb. 20, 2025 (FY2024 revenue $687.4M; FY2025 guide $715–735M). https://www.businesswire.com/news/home/20250220132389/en/Charles-River-Associates-CRA-Reports-Fourth-Quarter-and-Full-Year-2024-Financial-Results
11. ProPublica Nonprofit Explorer, "RAND Corporation," IRS Form 990 (~$462M revenue; ~1,800 staff). https://projects.propublica.org/nonprofits/organizations/951958142
12. Westat, "About Westat" / ZoomInfo company overview (employee-owned; revenue estimated ~$829M). https://www.westat.com/about-westat/
13. Zippia, "Abt Associates Revenue," 2023 (estimated ~$531M revenue). https://www.zippia.com/abt-associates-careers-13371/revenue/
14. Zippia, "American Institutes for Research Revenue" (estimated ~$459M revenue). https://www.zippia.com/american-institutes-for-research-careers-1202717/revenue/
15. NORC at the University of Chicago, "NORC and the University of Chicago" (independent 501(c)(3) UChicago affiliate; revenue ~$500M–$1B). https://www.norc.org/about/who-we-are/uchicago-affiliation.html
16. Urban Institute, "Introducing the Evidence and Equity Collaborative," 2025 (nine policy-research leaders; >20,000 employees). https://www.urban.org/press-releases/introducing-evidence-and-equity-collaborative
17. Mathematica, "Celebrating Employee Ownership at Mathematica" (fully employee-owned via ESOP). https://www.mathematica.org/news/celebrating-50-years-of-employee-ownership-at-mathematica-anniversary-video-highlights
18. RTI International, "Our History" (independent nonprofit research institute; center of gravity in life/physical sciences). https://www.rti.org/our-history
19. Booz Allen Hamilton, Form 10-K, 2025 (diversified federal analytics and program support). https://investors.boozallen.com/sec-filings/sec-filing/10-k/0001443646-25-000076
20. U.S. General Services Administration, Acquisition.gov, "FAR Part 31 – Contract Cost Principles and Procedures," and OMB Uniform Guidance (2 CFR 200) on indirect-cost rates. https://www.acquisition.gov/far/part-31
21. Congressional Research Service, "NIH Indirect Costs Policy for Research Grants: Recent Developments," report IN12516, 2025 (15% F&A cap; negotiated rates averaged ~58%; court injunction upheld Jan. 2026). https://www.congress.gov/crs-product/IN12516
22. U.S. Department of Health and Human Services, "Federal Policy for the Protection of Human Subjects (Common Rule)," 45 CFR 46 (IRB review, informed consent). https://www.hhs.gov/ohrp/regulations-and-policy/regulations/common-rule/index.html
23. Office of Management and Budget, "Federal Collection of Information" (Paperwork Reduction Act clearance; CIPSEA confidentiality for statistical data). https://obamawhitehouse.archives.gov/omb/inforeg_infocoll/
24. U.S. Government Accountability Office, "Evidence-Based Policymaking: Agencies Need Additional Guidance to Assess Their Capacity," GAO-24-106982, 2024 (Foundations for Evidence-Based Policymaking Act). https://www.gao.gov/products/gao-24-106982
25. The Spokesman-Review, "At least 1,200 grants canceled as NEH cuts target humanities across U.S.," Apr. 8, 2025. https://www.spokesman.com/stories/2025/apr/08/at-least-1200-grants-canceled-as-neh-cuts-target-h/
26. EdWeek Market Brief, "Sizing Up DOGE's Education Contract Cuts: What's Been Targeted," Mar. 2025 (Mathematica and AIR each 15 IES contracts cancelled). https://marketbrief.edweek.org/regulation-policy/an-in-depth-look-at-doge-cuts-to-federal-education-contracts/2025/03
27. Guidehouse, "Guidehouse Completes Transaction with Bain Capital," 2023 (acquired from Veritas Capital). https://guidehouse.com/news/corporate-news/2023/guidehouse-completes-transaction-with-bain-capital
28. Kantar, Annual Report and Accounts 2025 (research-and-data firm controlled by Bain Capital; consumer/marketing focus is mainly adjacent NAICS 541910). https://www.kantar.com/investor-relations
29. Institute of Education Sciences (IES), "What We Do" (education research, statistics, and program-evaluation funding). https://ies.ed.gov/
30. Ipsos, "International Social Research" (public-affairs, social, and behavioral research; much of the business is adjacent NAICS 541910). https://www.ipsos.com/en-uk/international-social-research
31. YouGov, "Financial Reports" (opinion and marketing data via survey panels; primarily adjacent market research). https://corporate.yougov.com/investors/financial-reports/