Management of Companies and Enterprises (NAICS 55111): An Investor's Primer

A Histometrics rollup primer for public-market and private investors. It synthesizes the three child-industry primers (551111, 551112, 551114) and our federal ground-truth statistics for this level. Figures are reported facts with sources; statements about the future are labeled as judgments in the wording.

---

1. Overview

This is the industry of the corporate parent and the corporate head office — the thin ownership-and-management layer that sits on top of operating businesses rather than running a storefront of its own. In the North American Industry Classification System (NAICS — the U.S. government's scheme for sorting businesses into industries), code 55111, "Management of Companies and Enterprises," gathers three national industries: the parents of banks, the parents of everything-else, and the head offices that actively manage a company's own units [2].

Three ideas carry through the whole level, and every child inherits them:

• These establishments have no outside customers. They hold stock, allocate capital, and administer subsidiaries; value flows up to owners as dividends, distributions, and gains — not as sales to the public [2].
• The right yardstick is not payroll — it is the assets controlled. By federal count the whole level pays ~$497 billion of wages a year [1], yet a single company inside it (Berkshire Hathaway) carries a market value near $1 trillion [11], and the banks sitting under just one of the child codes hold roughly $24 trillion in assets [6]. The government statistics measure the overhead; the wealth lives one layer down.
• The product is capital allocation and governance. Across all three children, an investor is really underwriting one question — does this corporate center create more value than it costs?

Two ways to participate, and both run through every child:

• Public route: shares, preferred stock, or bonds of a listed parent — a bank holding company, a diversified conglomerate, or a serial acquirer. There is no pure-play index for the level; exposure is name-by-name plus sector ETFs.
• Private route: the far larger population — privately held bank holding companies, family offices, personal holding companies, and private-equity (PE) platform vehicles. Most entities in this level are private and never appear in the headline stats.

---

2. What's inside — the three child industries and how they differ

The level splits into three national industries that look similar on paper — all are "parent layers" — but differ sharply in size, economics, ownership, regulation, and how you invest. The contrast is the whole point of this primer.

The one-sentence takeaway: 551114 owns almost all the people and payroll; 551111 and 551112 own almost all the passively held assets. The active managing offices employ ~97.5% of the level's workers because holding shells barely employ anyone — yet those near-empty shells (the bank and non-bank parents) sit atop trillions of dollars in banks, insurers, railroads, and portfolios. Employment and economic weight point in nearly opposite directions, and mixing them up is the classic error with this level.

Two dividing lines keep the three straight:

• Bank vs. non-bank separates 551111 from 551112. A parent that controls a bank is 551111 and must register with the Federal Reserve; a parent that holds anything else — an insurer, a utility, a family's real-estate LLC — is 551112 [ch. 551111][ch. 551112].
• Hold vs. manage separates both holding codes from 551114. The moment a parent's head office stops merely holding stock and starts directing and managing its subsidiaries, the Census Bureau reclassifies it into 551114 [ch. 551114]. A real corporate group often has both a holding office (551111/551112) and a managing office (551114).

---

3. How big the level is

Our federal ground-truth file (stats-55111.md) draws on the U.S. Census Bureau's County Business Patterns (CBP), which counts establishments with paid employees, their employment, and their payroll — not consolidated assets, revenue, or investment returns [1]:

These level totals equal the sum of the three children exactly — establishments (643 + 6,582 + 42,331 = 49,556) and employment (7,837 + 89,908 + 3,757,975 = 3,855,720) both reconcile to the ground-truth figures, and payroll ties to within rounding. The averages are calculations from the table, not separately reported federal metrics, and payroll is compensation paid, not revenue. Average pay near $129,000 is roughly double the ~$69,400 national average across all industries in 2023 [ch. 551114] — this is a white-collar concentration of executives and finance, legal, human-resources (HR), tax, risk, and information-technology (IT) staff, not a broad workforce. Our file contains no level-wide figure for the industry's assets, revenue, profit, or capital spending, so we state none.

Read the undercount before quoting any of this — it is severe, and unevenly so across the children.

1. CBP counts only employers. A large share of holding companies (551111 and 551112) have no payroll at all — they are paper parents whose entire job is to hold stock, and whose few corporate staff sit on a subsidiary's books. Those entities never enter the 49,556 count. Census's Nonemployer Statistics program does not fill the gap either: it explicitly excludes all of Sector 55 [4]. Third-party databases suggest on the order of ~18,000 firms in the 551112 code alone [ch. 551112] — a hint at how many employer-less shells sit outside the federal picture — and the Federal Reserve supervised ~3,747 bank holding companies at year-end 2024 [7], versus only 643 employer establishments under 551111. The undercount bites hardest exactly where small, private, and individual ownership dominates.
2. The operating businesses are counted elsewhere. The banks, insurers, railroads, and factories that these parents own are classified under their own industry codes. The people who actually do banking (~2 million in commercial banking) or run the trains are not in this level [ch. 551111][ch. 551112].
3. Payroll is the wrong ruler. The level's economic weight is the assets it controls — think of Berkshire's ~$1 trillion market value against the level's ~$11 billion of holding-company-office payroll in 551112, or the ~$24 trillion of banking assets governed through 551111 [11][6]. Treat the CBP totals as a measure of the head-office overhead layer, not of the wealth beneath it.

---

4. The investable universe — where value concentrates across the children

The paradox of this level: 551114 has essentially all the payroll but no pure-play security, while 551111 and 551112 have almost no payroll but are the things that trade. Public value concentrates in the two holding codes.

551111 — bank holding companies: the largest and deepest investable pool. Every U.S. global systemically important bank (GSIB — a bank regulators judge critical to the world financial system) is a bank holding company: JPMorgan, Bank of America, Citigroup, Wells Fargo, Goldman Sachs, Morgan Stanley, Bank of New York Mellon, and State Street [9]. The listed roster runs from mega-caps (JPMorgan, ticker JPM; Bank of America, BAC) down through super-regionals (PNC, U.S. Bancorp/USB, Truist/TFC) to hundreds of small- and micro-cap community-bank holding companies, plus custody banks and consumer-finance parents. Diversified exposure comes from bank exchange-traded funds (ETFs — baskets you buy like a stock) such as KBWB and KBE (large banks), KRE (regionals), and XLF (broad financials) [ch. 551111]. Beneath the listed names lies a long private tail: thousands of family-, employee-, mutual-, or OTC-owned bank holding companies (Arvest, MidFirst, First National of Nebraska), where federal control rules cap how much of a bank private capital can own without itself becoming Fed-supervised [ch. 551111].

551112 — other holding companies: fewer names, but the marquee capital allocators. Few industries are so dominated at the top by identifiable public names — Berkshire Hathaway (BRK.A/BRK.B), Brookfield (BN), Loews (L), Markel Group (MKL), Seaboard (SEB), Icahn Enterprises (IEP), and Compass Diversified (CODI) [ch. 551112]. Watch the structure: some are ordinary C-corporations, while others are limited partnerships (LPs) that issue a Schedule K-1 tax form instead of a 1099. There is no mainstream pure-play holding-company ETF, so public exposure is stock-by-stock. The far larger population is private: single-family offices (Deloitte counts ~8,030 worldwide and ~3,180 in North America as of 2024 [13]), personal holding companies, and privately owned parents such as Koch, Cargill, and Mars.

551114 — managing offices: no ticker, only proxies. You cannot buy the managing-office industry, because a head office is overhead, not a business that earns outside revenue [ch. 551114]. The closest expressions are companies whose corporate center itself is the thesis — diversified holding companies and serial acquirers where capital allocation is the product (Berkshire, Brookfield, Roper Technologies/ROP, Danaher/DHR, Markel), or the opposite trade: betting a bloated corporate layer gets broken up [ch. 551114]. Notice the overlap — Berkshire, Brookfield, Loews, and Markel appear as proxies for both 551112 and 551114, because a great holding company and a great managing office are often the same firm seen from two angles.

The pattern: liquid public value sits overwhelmingly in the two holding codes (551111 for regulated-bank exposure, 551112 for diversified capital allocators), while 551114 — the employment giant — is investable only indirectly through those same names.

---

5. How the money works

None of these establishments sells a product, so the usual industry metrics (same-store sales, occupancy, load factors) do not apply. Owners make money in three distinct ways, one per child:

• 551111 (banks): the parent earns almost nothing itself — its value is the bank underneath. The engine is net interest margin (NIM), the spread between what the bank earns on loans/securities and pays on deposits; industry NIM was 3.39% in Q4 2025 [ch. 551111]. Cash flows up to the parent as dividends (subject to regulatory limits), and the parent's job is capital return — dividends and buybacks. Watch return on equity, price-to-tangible-book, the common equity tier 1 (CET1) capital ratio, and credit quality.
• 551112 (other holdcos): the product is capital allocation — redeploying cash from mature subsidiaries into higher-return uses. Owners earn through upstream dividends (Loews took ~$934 million from its insurer CNA in 2024 [ch. 551112]), insurance float (premiums held before claims are paid — a genuine, industry-appropriate edge for Berkshire/Loews/Markel), and buying and selling businesses well. The scorecard is net asset value (NAV) growth and the sum-of-the-parts (SOTP) discount or premium.
• 551114 (managing offices): a cost center, paid by its own operating units through cost allocations and intercompany charges [ch. 551114]. The metric is overhead efficiency — corporate selling, general & administrative (SG&A) expense as a share of revenue — set against the value the center adds: pooled treasury, a lower cost of capital, tax efficiency, shared services, and capital allocation.

Two mechanics run across all three:

• The conglomerate discount. Because a diversified parent is hard to see through, it often trades ~10–20% below what its parts would fetch separately [12] — though a well-run one (Berkshire is the archetype) can trade at a premium, proving management quality outweighs the arithmetic. When the discount gets wide, break-up pressure builds.
• Double leverage. A parent can borrow at its own level and inject the proceeds into a subsidiary as equity; its debt is then serviced only by dividends flowing up. If a regulator, a covenant, or a downturn blocks those dividends, the parent gets squeezed — a shared risk for bank holdcos and non-bank holdcos alike [ch. 551111][ch. 551112]. Consolidated financial statements eliminate these intercompany flows, so read parent-only statements alongside the consolidated numbers.

---

6. What drives demand

There is no consumer demand curve here; demand is for ownership and management structures, and the drivers overlap heavily across the children:

• Interest rates and the cost of capital. Cheap debt fuels acquisitions and lifts asset values across all three; high rates raise parent borrowing costs, pressure NAV, and — for banks — reshape the net interest margin [ch. 551111][ch. 551112].
• Mergers and acquisitions (M&A). Deals are the oxygen of holding companies and the main source of new managing-office work (integration, reporting, tax, governance) [ch. 551114].
• Enterprise scale and complexity. More subsidiaries, geographies, and legal entities mean more head-office activity — the direct driver of 551114's headcount [ch. 551114].
• The insurance underwriting cycle, for the many insurance-heavy holdcos (Berkshire, Loews, Markel) — hard markets grow float and profits [ch. 551112].
• Wealth concentration and generational transfer, driving the fastest-growing private slice — family offices and personal holding companies [13].
• Regulation and risk. Disclosure, cybersecurity, tax, and compliance raise the value of centralized oversight — and, for banks, required capital directly sets how much they can lend and return [ch. 551111][ch. 551114].
• Automation and AI — a two-way force on 551114. It can shrink the number of people needed for corporate functions (a headcount headwind) even as it makes central functions more scalable [ch. 551114].

---

7. Regulation

Regulation is where the three children diverge most.

• 551111 is among the most heavily regulated businesses in the economy. The Bank Holding Company Act of 1956 and the Fed's Regulation Y require any company controlling a bank to register, submit to Federal Reserve supervision, and obtain approval to acquire or merge [ch. 551111]. The Fed is the "umbrella" supervisor of the consolidated holding company; a source-of-strength doctrine can require the parent to prop up a failing bank subsidiary; and the largest firms face annual stress tests and GSIB capital surcharges [ch. 551111].
• 551112 is lightly regulated as a parent — its oversight is inherited from its subsidiaries' industries. A pure holding company that merely owns stock faces few rules of its own, but insurance, utility, and financial subsidiaries carry their own regulators, and dividends up to the parent can require approval [ch. 551112]. Structural traps matter: the Investment Company Act of 1940 can pull a securities-heavy parent into a restricted regime if "investment securities" exceed 40% of its unconsolidated assets [14], and the personal holding company (PHC) tax plus the accumulated-earnings tax push cash out to shareholders rather than letting it pile up [15].
• 551114 has no industry-specific regulator at all — it is the place where the company complies with everything else [ch. 551114]. Key items: SEC disclosure and internal-control rules for public parents, cybersecurity-incident disclosure, and — distinctively — transfer pricing, where the IRS can adjust the fees a managing office charges its own subsidiaries under Internal Revenue Code Section 482 [16].

Two threads cut across the level: antitrust review (the FTC and DOJ can review the M&A that builds these groups) and the fact that most entities here are lightly regulated shells whose real regulatory weight is borrowed from the operating businesses below.

---

8. Consolidation

Consolidation runs in both directions, and its character differs by child.

• 551111 — a decades-long contraction. U.S. banks fell from a 1986 peak of ~18,083 to ~4,336 in 2025 (down ~13% just since 2020), with new-bank formation nowhere near replacing merger losses [6]. Deal flow is heavily gated by regulators but accelerating: 2025 delivered the fastest pace of bank-merger approvals since 1990 [ch. 551111], punctuated by Capital One's ~$35.3 billion acquisition of Discover (completed May 2025) and a run of regional mergers into early 2026 [ch. 551111].
• 551112 — holding companies are consolidation vehicles. They grow by buying and compounding with permanent capital; a wave of PE-sponsored "permanent-capital" structures has expanded the population. But when the conglomerate discount widens, the pressure reverses into break-ups and spin-offs — the multi-year GE breakup is the marquee example [ch. 551112].
• 551114 — M&A both creates and destroys managing offices. Each merger spawns new head-office work but, more visibly, eliminates duplicate corporate staff — "synergies" are very often cuts to redundant headquarters. Activist-driven break-ups push the other way [ch. 551114]. A distinctive geographic dynamic sits on top: metros compete for headquarters, and relocations have accelerated toward lower-tax states (one tally counted ~96 HQ-move announcements in 2024 versus ~18 in 2023) [18].

The common engine across all three is the same trade-off — cheaper funding, duplicated-cost reduction, and broader distribution on the upside; regulatory delay, overpayment, and messy integration on the downside.

---

9. Risks

• Look-through, not headline, risk. Because these are parents, you inherit the cyclical, credit, and operating risks of the businesses below — a bank's credit cycle (commercial-real-estate exposure is a live worry [ch. 551111]), an insurer's underwriting, an industrial's demand.
• The discount can persist or widen. There is no guarantee the market ever pays sum-of-the-parts; a well-run parent can compound below NAV for years [12].
• Double leverage and dividend blockage. Parent debt depends on upstream cash that a regulator, covenant, or downturn can cut off — a shared vulnerability of bank and non-bank holdcos [ch. 551111][ch. 551112].
• Key-person and succession risk. Value is often tied to one capital allocator; Berkshire's Buffett-to-Greg-Abel handover at the start of 2026 is the level's most-watched transition [10].
• Opacity and fraud at subsidiaries. Diversification hides problems: Compass Diversified fell ~59% in a day in May 2025 on accounting irregularities at its Lugano unit, later restating results by more than $750 million [ch. 551112].
• The corporate center as first cut. For 551114, overhead is the classic early target in a downturn, and corporate/administrative roles are among the most exposed to automation and AI [ch. 551114].
• Regulatory and capital change. For banks, unresolved capital rules (Basel III "Endgame") and the stress-test overhaul directly move lending capacity and payouts [ch. 551111].
• Data and undercount risk. The federal statistics capture only the employer offices; private, nonemployer, and family-structured entities — much of this level — are not fully measured [4][ch. 551114].

---

10. How to invest, and the outlook

Public routes, by child.

• 551111: common stock of a listed bank holding company (analyze the consolidated parent — price-to-tangible-book, ROE, NIM sensitivity, CET1, credit); preferred stock and bonds issued by the holdco (a mainstay for income investors); or bank ETFs (KBWB/KBE, KRE, XLF) for diversified exposure [ch. 551111].
• 551112: the most liquid single name is Berkshire's B share (BRK.B); build a diversified basket from the listed conglomerates, minding C-corp (1099) versus LP (K-1) structures. No dominant pure-play ETF exists, so exposure is stock-by-stock [ch. 551112].
• 551114: no pure play — own the good corporate center (Berkshire/Brookfield/Roper/Danaher/Markel) or bet against the bad one via break-up and spin-off situations [ch. 551114].

Private routes. Community-bank equity, de novo bank sponsorship, and recapitalizations (551111); family offices, personal holding companies, and PE permanent-capital vehicles (551112 and 551114). Across all three, due diligence should separate parent-only cash and debt from subsidiary debt, map every material subsidiary and liability, examine dividend restrictions and control/related-party terms, and weigh succession.

One caveat that trips people up: depositing money at a bank, or holding a subsidiary's product, is not the same as investing in the parent — and FDIC deposit insurance protects eligible bank deposits, not a holding company's shares, debt, or preferred securities [ch. 551111].

Outlook (forward-looking judgments, not settled facts). The level is structurally durable but will not behave like a high-growth industry, and returns will stay highly dispersed:

• 551111: the path of interest rates sets bank margins, and the 2025–26 merger wave among community and regional banks looks set to continue — though bank-level capital limits still bind payouts [ch. 551111].
• 551112: the M&A and insurance-pricing cycles drive earnings, the Berkshire post-Buffett transition is a bellwether for how markets price key-man succession, and the wealth-transfer wave should keep expanding the private end regardless of public markets [10][13].
• 551114: AI-driven productivity likely compresses head-office headcount even as HQ relocations redraw the geographic map — a necessary, moderately growing layer under real cost pressure [18][ch. 551114].

The enduring signal is the same across the whole level: this is a capital-allocation and governance industry. Its returns track the quality of the people allocating the capital — and the assets sitting beneath the thin office layer — far more than the payroll the federal statistics happen to measure.

---

Sources

Citations marked "[ch. NNNNNN]" point to the corresponding Histometrics child-industry primer (551111, 551112, or 551114), where the underlying primary sources are listed in full.

1. U.S. Census Bureau. County Business Patterns, 2023 — NAICS 55111 and its components 551111 / 551112 / 551114 (establishments, employment, annual and first-quarter payroll). Level totals per Histometrics ground-truth file stats-55111.md. https://www.census.gov/programs-surveys/cbp.html
2. U.S. Census Bureau / Office of Management and Budget. 2022 North American Industry Classification System Manual — Sector 55 and industries 551111, 551112, 551114 (scope, hold-vs-manage line, exclusions). https://www.census.gov/naics/reference_files_tools/2022_NAICS_Manual.pdf
3. U.S. Census Bureau. County Business Patterns — Methodology (employer establishments only; coverage and undercoverage). https://www.census.gov/programs-surveys/cbp/technical-documentation/methodology.html
4. U.S. Census Bureau. Nonemployer Statistics (excludes Sector 55, Management of Companies and Enterprises). https://www.census.gov/econ/overview/mu0500.html
5. U.S. Small Business Administration. Table of Small Business Size Standards (NAICS 551111 = $38.5M; 551112 = $45.5M average annual receipts). https://www.sba.gov/document/support-table-size-standards
6. Federal Reserve Bank of St. Louis. Banking Analytics: Banks Experience Asset Growth amid Ongoing Consolidation (~4,336 banks; ~$24.2T assets; long-run bank count). 2026. https://www.stlouisfed.org/on-the-economy/2026/may/banking-analytics-banks-experience-asset-growth-ongoing-consolidation
7. Congressional Research Service. Bank Holding Companies: Background and Issues for Congress (R48291; 3,747 BHCs at year-end 2024). 2025. https://www.congress.gov/crs-product/R48291
8. Board of Governors of the Federal Reserve System. Regulation Y / Bank Holding Company Act (12 CFR Part 225). https://www.ecfr.gov/current/title-12/chapter-II/subchapter-A/part-225
9. Office of Financial Research. Largest U.S. Bank Holding Companies — G-SIB figures. https://www.financialresearch.gov/gsib-scores-chart/
10. Berkshire Hathaway. 2024 Annual Report (business segments; Greg Abel to succeed Warren Buffett as CEO at the start of 2026). 2025. https://berkshirehathaway.com/2024ar/2024ar.pdf
11. Market-data aggregators (CompaniesMarketCap, MacroTrends, StockAnalysis). Approximate market capitalizations, early 2026 — Berkshire Hathaway ~$1 trillion and peer holding companies. https://companiesmarketcap.com/
12. LegalClarity / McKinsey & Company. The conglomerate discount — definition, calculation, and corporate-center value. https://legalclarity.org/what-is-a-conglomerate-discount-and-how-is-it-calculated/; https://www.mckinsey.com/capabilities/strategy-and-corporate-finance/our-insights/is-your-conglomerate-discount-a-performance-discount-or-a-communication-problem
13. Deloitte Global. The Family Office Insights Series — Global Edition (~8,030 single-family offices worldwide; ~3,180 in North America, 2024). 2024. https://www.deloitte.com/global/en/about/press-room/global-edition-explores-the-rapid-expansion-family-offices-and-offers-vision-of-the-future-landscape.html
14. U.S. Securities and Exchange Commission. Investment Company Registration and Regulation Package (40%-of-unconsolidated-assets test). https://www.sec.gov/about/divisions-offices/division-investment-management/investment-company-registration-regulation-package
15. Legal Information Institute (Cornell Law). 26 U.S. Code § 541 — Imposition of personal holding company tax (20% rate). https://www.law.cornell.edu/uscode/text/26/541
16. Internal Revenue Service. Transfer Pricing (Internal Revenue Code Section 482). https://www.irs.gov/businesses/international-businesses/transfer-pricing
17. U.S. Census Bureau. 2022 Economic Census — Sector 55 (revenue reported largely as intra-company management fees). https://www.census.gov/data/tables/2022/econ/economic-census/naics-sector-55.html
18. CRE Daily. Headquarters Relocations Rise Sharply (~96 HQ-move announcements in 2024 vs. ~18 in 2023; Fortune 500 HQ concentration). 2025. https://www.credaily.com/briefs/headquarters-relocations-rise-sharply/
19. Histometrics child-industry primers: Offices of Bank Holding Companies (551111), Offices of Other Holding Companies (551112), and Corporate, Subsidiary, and Regional Managing Offices (551114) — for the underlying primary sources cited inline as "[ch. NNNNNN]."