Taxi and Ridesharing Services (U.S.) — Industry Primer

NAICS 2022 code 485310. NAICS is the North American Industry Classification System, the federal standard for grouping businesses by activity.

1. Overview

This is the business of moving people on demand by car or van — no fixed route, no fixed schedule. It spans two worlds that federal statistics lump together: the old world of metered street-hail and dispatch taxis, and the newer world of app-based ridesharing (Uber, Lyft) that now dominates it. In 2010 this was a fragmented, locally licensed, largely cash business; today a single app-based platform can coordinate tens of millions of trips a day worldwide [1].

The industry has one useful dividing line running through it: the marketplace versus the vehicle operator. Platforms generate demand, set or influence pricing, and dispatch rides; drivers, fleet owners, and operators supply the vehicles, labor, insurance, and local permits. Understanding almost everything else about the sector — the economics, the labor politics, the regulation — starts from that split.

There are public and private ways to participate. The public market offers two direct plays — Uber (a diversified global platform) and Lyft (a near-pure U.S./Canada rideshare bet) — plus indirect exposure to autonomous vehicles (AVs — self-driving cars) through Alphabet (Waymo), Tesla, and Amazon (Zoox). The private side is the traditional taxi world — medallion owners, city fleets, regional dispatch companies, fleet roll-ups — plus venture-funded robotaxi startups. Most of that private side is illiquid, locally licensed, and in the case of legacy taxi assets, distressed. Details in sections 4 and 10.

2. What it is and how it's structured

Scope. NAICS 485310 covers establishments that provide passenger transportation by automobile or van, not on regular routes or schedules. It explicitly includes taxicab owner-operators, taxi fleet operators, taxi dispatch/organization services, and ridesharing / ride-hailing services — including the app-matching (arrangement) services themselves [2].

What it excludes (adjacent NAICS codes) [2]:

• Limousine service — 485320. Chauffeured luxury car service is a separate industry.
• Charter bus — 485510; special-needs (disabled/elderly) transport — 485991; scheduled airport/hotel/destination shuttles — 485999.
• Car-pool and van-pool arrangement — 488999.
• Passenger-car rental without a driver — 532111.
• Intercity/rural bus and urban transit systems sit elsewhere in subsector 485.

So a "ride" in this code is the point-to-point, on-demand car trip — whether you flagged it on a corner or summoned it on a phone.

Ownership mix. The structure is barbell-shaped:

• At one end, a handful of enormous, venture-born technology platforms that own almost no cars and employ almost no drivers — they run the matching software, set pricing, and take a cut.
• At the other end, thousands of tiny operators: single-car owner-drivers, small taxi fleets, and medallion holders (a medallion is a transferable city license to operate a taxi).
• In between, dispatch/software companies that connect riders with licensed operators, and private-equity-backed operators that consolidate local fleets while preserving local licenses and contracts.

The federal concentration data confirm the top-heaviness: the four largest firms collect about 85% of industry receipts, and the top 20 collect 88.9% [3] — a duopoly with a long tail.

Crucially, the workforce is overwhelmingly independent contractors, not employees. App drivers are classified as self-employed 1099 workers (named for the U.S. tax form issued to contractors, versus the W-2 form for employees), which is central to how the economics — and the legal risk — work (sections 5, 7, 9). Federal statistics do not publish a clean public-versus-private ownership split, and they undercount the smallest operators (section 3).

3. How big it is

Federal business statistics for this code, from our ground-truth figures:

(CR4/CR8/CR20/CR50 are concentration ratios — the share of receipts held by the largest 4, 8, 20, and 50 firms. SBA is the U.S. Small Business Administration; its $19 million figure is a federal contracting threshold, not a market-size estimate. The Herfindahl-Hirschman Index (HHI), a finer concentration measure, is suppressed for this code, so we do not report it.)

The undercount — the single most important caveat for this industry. These federal figures dramatically understate the real economic footprint, for three structural reasons:

1. The labor is contractor labor. County Business Patterns counts paid employees on a W-2 payroll. But the millions of people who drive for Uber and Lyft are independent contractors, so they do not appear in the 22,790 "employees" or the $1.45 billion "payroll." Broad estimates put platform-based gig drivers in the low single-digit percent of U.S. adults — i.e., several million people [6]; Massachusetts alone counted roughly 70,000 ride-hail drivers [7]. The 22,790 employee figure mostly captures dispatch-office staff and the platforms' own payrolled corporate/engineering workers, not the driving workforce.

2. Sole-proprietor drivers are excluded. Owner-drivers with no employees are "nonemployer" businesses and fall outside the establishment counts above entirely; the Census measures them in a separate program [8].

3. Receipts ≠ consumer spending. The $14.37 billion receipts figure largely reflects the platforms' net take (commissions/service fees) plus taxable fleet revenue — not the gross fares consumers pay. Gross consumer spend on U.S. ride-hailing runs in the tens of billions of dollars a year; third-party market estimates for U.S. ridesharing range from roughly $28 billion (2024) to $58 billion (2025) depending on whether they count gross bookings or net revenue [9][10]. For scale, Uber's global gross bookings across all its businesses reached $193 billion in 2025 [1].

Bottom line: treat the federal receipts and payroll here as a floor. The genuine scale of activity — measured by trips, gross fares, and people driving — is many times larger, and the standard business census cannot see most of it because the industry runs on contract drivers and a few giant platforms.

4. The investable universe

Only a few names give direct public exposure; most of the traditional industry is private and fragmented.

Public — direct:

Public — indirect (autonomous-vehicle optionality, not primarily rideshare companies):

There is no large, pure-play U.S. public taxi-fleet company and no meaningful "taxi" ETF (exchange-traded fund); indirect exposure otherwise runs through AV developers, vehicle suppliers, insurers, payment processors, and car-rental firms, or through broad technology / consumer-discretionary and thematic autonomous-mobility funds.

Major private-market owners and platforms:

• WHC Worldwide / zTrip — a privately held ground-transportation holding company, formed specifically to consolidate regulated passenger transport; it acquired zTrip in 2019 and says it operates more than 2,800 vehicles with more than 3,400 drivers across 38 cities in 20 states [16].
• Curb Mobility — a private taxi-technology platform (booking, payment, dispatch, fleet management); it reports more than 100,000 drivers on its network and more than 10 million monthly trips, though it is chiefly a technology/network provider rather than the owner of most vehicles [17].
• Local medallion and fleet owners — the largest private universe: city-based operators, franchisees, and owner-operators. No authoritative federal ranking of these private owners is published. Premium/niche players (e.g., Alto, Revel) round it out. If you want the classic taxi business itself, it is a private, distressed, illiquid, locally licensed asset class — not a stock.

5. How the money works

Two very different profit engines sit inside this one code.

The platform (marketplace) model — Uber, Lyft. These are asset-light software marketplaces. The core metrics:

• Gross bookings — the total value of fares riders pay (the top line of activity).
• Take rate — the percentage the platform keeps as revenue after paying the driver. Because the platform often acts as the driver's agent, reported revenue is far below gross bookings. Independent analysis pegs the effective U.S. take at roughly 40%+ on average once fees are included (Uber's has risen toward ~42% since it moved to "upfront pricing"), though it varies widely trip to trip [18].
• Net revenue ≈ gross bookings × take rate. Uber's global revenue was ~$52 billion on ~$193 billion of bookings in 2025 [1]; Lyft's was ~$6.3 billion [13].
• Active riders and trips. Uber reported 202 million monthly active platform consumers (MAPCs) in late 2025, up 18% year over year [1]. More riders and drivers in the same city create a liquidity flywheel: denser supply means shorter waits and higher car utilization, which lifts margins.
• Contribution margin and adjusted EBITDA (earnings before interest, taxes, depreciation and amortization) — the profit left after the direct costs of a trip, chiefly insurance (a very large line item), payment processing, and rider/driver incentives. After a decade of losses, both platforms now generate positive operating profit; Uber posted ~$5.6 billion of operating income in 2025 [1]. The biggest costs are incentives (used to seed liquidity) and insurance/claims; the biggest lever is scale — once a city is dense, each additional trip is highly profitable.

The fleet / traditional-taxi model. A fleet operator or medallion holder earns metered fares or contracted transportation revenue and pays for vehicles, drivers, fuel or electricity, maintenance, insurance, permits, dispatch, and financing. The right metric here is not revenue per ride but revenue per vehicle-hour or vehicle-day after empty miles. Useful operating measures include vehicle utilization and empty-mile percentage, trips per active vehicle-hour, driver retention and earnings, insurance cost and claims frequency per mile, maintenance cost per mile, contract renewal rates and customer concentration, and cash conversion (a fleet with high reported revenue but weak cash conversion is not attractive). A medallion was historically the core asset — a scarce, appreciating license — but that scarcity collapsed when app-based supply flooded the streets (section 8).

The driver's economics (the microbusiness). A driver's revenue is fares + tips + platform incentives; costs are fuel, vehicle depreciation, maintenance, and insurance; net earnings are what's left after the platform's take. Because drivers are contractors, the platform carries almost none of these costs — which is exactly why the model scales and why driver-pay politics are so charged.

6. What drives demand

Ride demand is largely discretionary and tied to activity and cost trade-offs:

• Urban density and limited/expensive parking — trips concentrate in cities.
• Travel and events — airports, hotels, tourism, business and convention travel, nightlife, and weather.
• Households without cars and gaps or disruptions in public transit.
• Cost of the alternatives — car-ownership costs, parking, gas prices, and transit quality push riders toward or away from hailing.
• Smartphone penetration and habit — the app model expanded the total market well beyond old taxi demand [9].
• Safety — avoiding drunk driving is a durable demand source.
• Institutional demand — corporate accounts, hospitals, hotels, and government voucher programs, which are more defensive than leisure trips.
• Driver supply — supply is demand's mirror: gig-labor availability rises when the broader job market loosens and falls when wages elsewhere rise, directly affecting wait times and prices.

The business is cyclical through travel, employment, and discretionary spending, but local supply-demand balance matters more than national GDP: a platform can grow rides while margins deteriorate if it must subsidize drivers or cut prices.

7. Regulation

Regulation is primarily local and state-based, layered, and central to the investment case. Cities and counties commonly control taxi licenses, medallions, metered fares, vehicle standards, inspections, driver credentials, airport access, and curb usage. States commonly regulate the app platforms, insurance, background checks, worker rules, and reporting.

• Transportation Network Company (TNC) laws. App-based rideshare is regulated under "TNC" statutes (a TNC is a company that uses an online app to connect passengers with drivers using their personal vehicles). These typically require operating authority, driver background checks, accessibility plans, operating reports, and specific insurance minimums. Examples: New York requires $1.25 million of liability coverage per occurrence while a passenger is in the car [19]; California requires at least $1 million of primary commercial coverage during the pickup and passenger periods [20].

• Driver classification — the single biggest legal variable. The whole platform model depends on drivers being independent contractors, not employees. California's AB5 (2019) threatened that by applying a strict "ABC test." The industry responded with Proposition 22 (2020), a ballot measure keeping app drivers as contractors while adding a pay floor and limited benefits; the California Supreme Court upheld Prop 22 in 2024 (Castellanos v. State of California) [21][22]. Uber and Lyft both still disclose continuing litigation and regulatory uncertainty over employee status, wages, benefits, and collective bargaining [1][13] — the fight is not over.

• Pay and take-rate rules. Cities and states increasingly set minimum driver earnings (New York City's Taxi and Limousine Commission pioneered this) and are adding benefits and protections [7][23]. In Congress, a proposed Empowering App-Based Workers Act would require per-trip take-rate disclosure and cap the platforms' take at 25% — a forward-looking risk, not current law [18].

• Accessibility. The Americans with Disabilities Act (ADA) imposes obligations on public entities and on private entities providing specified public transportation, which can affect vehicle mix, dispatch, app design, driver training, and contract eligibility [24].

• Traditional taxi licensing. Cities still cap and license taxis (medallions/permits), set metered rates, and mandate wheelchair-accessible service — rules that early ridesharing largely sidestepped, which was the source of the original competitive shock.

• Electric and autonomous vehicles. Robotaxis face a separate, evolving state-and-federal permitting regime for driverless operation — the key regulatory frontier for the industry's next phase. In 2025 the National Highway Traffic Safety Administration (NHTSA) expanded its automated-vehicle exemption program and granted a demonstration exemption to Zoox [25].

8. Competitive dynamics and consolidation

A national duopoly, atop a fragmented base. In the U.S., Uber holds roughly three-quarters of rideshare and Lyft the rest [12] — broadly consistent with the federal CR4 of 85% [3]. One caveat: the CR4 measures national receipts concentration; it shows a long tail of small local establishments alongside a few platform-scale firms, but it does not prove that four brands dominate every individual city. The barriers now protecting the incumbents are the ones they built: rider and driver liquidity in each city, brand trust, and scale economics on insurance and technology. A new entrant must subsidize both sides of the market in every city at once — very expensive.

The traditional taxi's decline. The clearest casualty is the medallion. New York City medallions averaged about $1.32 million in 2014, then fell roughly 90% as unlicensed rideshare supply flooded in — private sales ran $90,000–$200,000 by early 2025, leaving many owner-drivers in bankruptcy; a city relief program had written down over $472 million of medallion debt by early 2025 [26]. Traditional taxis survive in defensible niches — airport queues, street hails, cash riders, corporate and government contracts, and accessibility mandates — but no longer set the market.

Consolidation. Two forces run at once. On the platform side, the majors are rolling up adjacent mobility: in 2025 Lyft bought Europe's FreeNow taxi app (about $200 million) — its first move outside North America — plus a premium chauffeur business [13][27], while Uber continues to acquire and partner globally. On the private fleet side, holding companies such as WHC Worldwide consolidate local operators to centralize dispatch, procurement, compliance, insurance, and technology [16] — though local regulation, driver turnover, insurance claims, and market-specific practices limit the savings from a national roll-up.

The autonomous frontier. Robotaxis are both the biggest competitive threat and the biggest opportunity. Waymo (Alphabet) is far ahead — around 500,000 paid rides a week across roughly 11 U.S. cities, and it both runs its own app and partners with Uber in markets such as Austin and Atlanta [14]. Tesla's Robotaxi and Amazon's Zoox are early and much smaller [14][15]. The strategic question for Uber and Lyft is whether driverless fleets flow through their apps (extending the marketplace) or around them (disintermediation).

9. Risks

• Driver reclassification. A shift to employee status in major states would raise costs sharply (wages, payroll taxes, benefits, insurance, compliance) and is the largest structural risk to the platform model. Prop 22's survival helps, but the issue is live nationwide [21][22].
• Regulatory squeeze. Take-rate caps, mandated minimum pay, benefits, fare caps, permit expansion, airport rules, accessibility mandates, and congestion charges all compress margins [18][23].
• Insurance and liability. Claims and liability are among the largest cost lines; rising accident/litigation costs and inadequate reserves can overwhelm thin margins [19][20].
• Autonomous disruption — two-sided. AVs could remove the driver cost (upside) or route trips around the platforms (downside); timing, capital needs, liability models, and winners are all uncertain [14].
• Price competition. Riders and drivers can use multiple apps at once ("multi-homing"), weakening pricing power.
• Supply shortages. Insufficient driver or vehicle availability raises wait times and forces up incentives.
• Fleet capital intensity. For vehicle operators, cars depreciate, financing costs fluctuate, and utilization can fall during demand shocks.
• Cyclicality. Ride demand is discretionary and softens in downturns, like other travel and leisure spending.
• Safety, cybersecurity, and privacy. A serious passenger-safety incident or data breach can create legal, regulatory, and reputational costs.
• Concentration risk for investors. Lyft is a near-pure U.S. rideshare bet, so it carries full exposure to all of the above; Uber diversifies with Delivery and Freight but is therefore not a clean way to own only rideshare [1][13].
• Data-quality risk. Employer-only federal statistics understate owner-operator activity, while public-company financials include large businesses outside NAICS 485310 — so neither source, alone, sizes the industry cleanly.

10. How to invest and the outlook

Public routes.

• Direct: Uber (UBER) for the diversified global platform; Lyft (LYFT) for a concentrated U.S. rideshare bet. Analyze the relevant Mobility/Rides segments rather than consolidated results, and judge them on gross-bookings growth, take rate, insurance-reserve development, driver supply, and the path from bookings to adjusted EBITDA and free cash flow. Useful valuation tools include enterprise-value-to-EBITDA (EV/EBITDA), FCF yield, and revenue/gross-bookings growth [1][13].
• Indirect / autonomy: Alphabet (Waymo), Tesla (Robotaxi), and Amazon (Zoox) offer AV upside, but ridesharing is a small slice of each — you are buying a much larger company [14][15].
• Funds: no pure taxi ETF; exposure comes via broad tech / consumer-discretionary or thematic autonomous-mobility funds.

Private routes.

• Fleet roll-ups, medallion acquisitions, dispatch software, vehicle financing, insurance capacity, and contracted transportation. Underwrite normalized revenue per vehicle, utilization, driver retention, insurance claims, permit status, contract concentration, maintenance capital spending, debt service, and owner labor — a fleet with high reported revenue but low cash conversion is not attractive.
• Traditional taxi assets — medallions, fleets, dispatch companies — are a distressed, illiquid, locally licensed asset class, not a growth vehicle; enter only with a specific operating or turnaround thesis.
• Robotaxi and mobility startups are venture-stage and largely inaccessible to public investors (the leading ones already sit inside Alphabet and Amazon).

Near-term drivers to watch (forward-looking). (1) The pace and unit economics of robotaxi commercialization — whether driverless fleets prove cheaper than paid drivers, and whether they run through Uber/Lyft or compete with them. (2) Regulatory outcomes on take-rate caps, minimum pay, and — most of all — driver classification. (3) Insurance costs. (4) Consumer discretionary spending and local driver supply.

Outlook — editorial judgment. Demand for on-demand ground transportation should remain durable, but returns will depend on unit economics, not ride volume alone. Public platforms have the stronger network scale and technology advantages, now paired with real profitability; private fleets can offer better control over vehicles, contracts, and service quality. The reasonable base case is continued high-single-to-low-double-digit gross-bookings growth for the incumbents against a slow but potentially transformative rollout of autonomy. The near-term winners are most likely operators — platform or fleet — that combine dense local supply, disciplined insurance management, reliable service, and positive cash flow. Both the biggest upside and the biggest disruption in this industry point at the same thing: the self-driving car.

---

Sources

1. Uber Technologies, "Uber Announces Results for Fourth Quarter and Full Year 2025," 2026. https://investor.uber.com/news-events/news/press-release-details/2026/Uber-Announces-Results-for-Fourth-Quarter-and-Full-Year-2025/default.aspx
2. U.S. Census Bureau, "2022 NAICS Definition — 485310 Taxi and Ridesharing Services," 2022. https://www.census.gov/naics/?input=485310&year=2022
3. U.S. Census Bureau, "2022 Economic Census — Concentration of Largest Firms and receipts, NAICS 485310," 2025. https://data.census.gov/table/ECNSIZE2022.EC2200SIZECONCEN
4. U.S. Census Bureau, "County Business Patterns, NAICS 485310," 2023. https://www.census.gov/programs-surveys/cbp.html
5. U.S. Small Business Administration, "Table of Small Business Size Standards," 2023. https://www.sba.gov/document/support-table-size-standards
6. Goldman Sachs Global Investment Research / GigEconomyData.org, "How many gig workers are there?," 2025. https://gigeconomydata.org/basics/how-many-gig-workers-are-there.html
7. Stateline, "Rideshare drivers win more benefits, protections despite industry pushback," 2025. https://stateline.org/2025/05/22/rideshare-drivers-win-more-benefits-protections-despite-industry-pushback/
8. U.S. Census Bureau, "Nonemployer Statistics," 2023. https://www.census.gov/econ/overview/mu0500.html
9. GMInsights, "US Ride Sharing Market Size & Share, 2025–2034," 2025. https://www.gminsights.com/industry-analysis/us-ride-sharing-market
10. MarketDataForecast, "U.S. Ride Sharing Market Size, Share, Growth & Trends," 2025. https://www.marketdataforecast.com/market-reports/united-states-ride-sharing-market
11. CompaniesMarketCap, "Uber (UBER) and Lyft (LYFT) market capitalization," 2026. https://companiesmarketcap.com/uber/marketcap/
12. Tickeron, "Uber vs. Lyft: U.S. rideshare market share," 2026. https://tickeron.com/blogs/uber-vs-lyft-earnings-preview-as-q4-results-put-ride-demand-and-margins-in-focus-11692/
13. Lyft, Inc., "Form 10-K, FY2025," 2026. https://www.sec.gov/Archives/edgar/data/1759509/000162828026006960/lyft-20251231.htm
14. CNBC, "Robotaxis in 2025: Waymo plots global expansion as Zoox, Tesla roll to the starting line," 2025. https://www.cnbc.com/2025/12/16/waymo-amazon-zoox-tesla-robotaxi-expansion.html
15. InsideEVs, "Waymo's Expansion Is Leaving Tesla In The Dust," 2026. https://insideevs.com/news/788284/waymo-expansion-tesla-cities-2026/
16. WHC Worldwide, "About," 2026. https://whcworldwide.com/about
17. Curb Mobility, "About Us," 2026. https://www.gocurb.com/about
18. National Employment Law Project, "Unpacking Uber & Lyft's Predatory 'Take Rates'," July 2025. https://www.nelp.org/insights-research/unpacking-uber-and-lyfts-predatory-take-rates/
19. New York State DMV, "Information for Transportation Network Company (TNC) Passengers," 2025. https://dmv.ny.gov/business/information-for-tnc-passengers
20. California Public Utilities Commission, "Insurance Requirements for Transportation Network Companies," 2026. https://www.cpuc.ca.gov/regulatory-services/licensing/transportation-licensing-and-analysis-branch/transportation-network-companies/tnc-insurance-requirements
21. Ogletree Deakins, "Supreme Court of California Upholds Law Classifying App-Based Drivers as Independent Contractors," 2024. https://ogletree.com/insights-resources/blog-posts/supreme-court-of-california-upholds-law-classifying-app-based-drivers-as-independent-contractors/
22. American Bar Association, "App Drivers Are Not Employees," 2025. https://www.americanbar.org/groups/litigation/resources/litigation-news/2025/app-drivers-are-not-employees/
23. Human Rights Watch, "The Gig Trap: Algorithmic, Wage and Labor Exploitation in Platform Work in the US," 2025. https://www.hrw.org/report/2025/05/12/the-gig-trap/algorithmic-wage-and-labor-exploitation-in-platform-work-in-the-us
24. U.S. Department of Transportation, "The Americans with Disabilities Act of 1990," 1990. https://www.transportation.gov/civil-rights/civil-rights-library/americans-disabilities-act-1990
25. National Highway Traffic Safety Administration, "NHTSA Issues First-Ever Demonstration Exemption to American-Built Automated Vehicles," 2025. https://www.transportation.gov/briefing-room/nhtsa-issues-first-ever-demonstration-exemption-american-built-automated-vehicles
26. Wikipedia, "Taxi medallion," 2025 (citing NYC Taxi and Limousine Commission data). https://en.wikipedia.org/wiki/Taxi_medallion
27. CNBC, "Lyft to buy taxi app FreeNow for $200 million to expand into Europe," 2025. https://www.cnbc.com/2025/04/16/lyft-to-buy-taxi-app-free-now-for-200-million-to-expand-into-europe.html