RV Parks and Campgrounds: Finding Seasonality in Public Data

Search for RV park demand data and the results are outlooks: occupancy trends, revenue projections, cap-rate commentary, industry statistics pages quoting one another. This article does something different with the same asset class. It shows how the demand behind RV parks and campgrounds is measured, from federal series that publish on a schedule, because this is the property type whose defining risk, the season, is also its most measurable feature. The Quarterly Census of Employment and Wages counts campground payrolls month by month; the American Community Survey maps the seasonal-home geography the parks share; the National Park Service publishes visitation to the destinations that anchor whole corridors of demand; the Bureau of Economic Analysis values RVing as an activity; and the Federal Highway Administration measures the summer swell of the road itself. None of these series quotes an occupancy rate, and this article does not either. What they establish is stronger: where the season is, how deep it runs, and how to read both for a specific property from sources a credit analyst can open without a subscription.

Definitions before measurement. NAICS 721211, RV parks and campgrounds, covers establishments operating sites for recreational vehicles, tents and cabins; it does not cover the manufactured-housing communities that rent pads as primary residences, a different asset class with a different demand base covered elsewhere in this library. The overlap matters at the margins, because long-stay RV residents blur the line, and the honest inventory decides park by park which side of it each property sits on.

The asset, counted: a slow-growing supply of very small operators

County Business Patterns counted 5,018 RV park and campground establishments with paid employees in 2023, up from 4,587 in 2019 and 4,483 in 2017, growth of 9.4% over four years and 11.9% over six (U.S. Census Bureau, County Business Patterns 2017, 2019 and 2023, 2019 to 2025). Employment rose faster, from 20,902 in 2017 to 26,333 in 2023, and annual payroll rose 58.8%, from $641 million to $1.02 billion. The size structure is the familiar signature of a mom-and-pop asset class: 69.2% of establishments have fewer than five employees, 18.4% have five to nine, and only 3.4% have twenty or more. The Quarterly Census of Employment and Wages, on its different reporting-unit basis, shows the same slow climb: 4,738 private reporting units on annual average in 2019, 5,108 in 2021, 5,529 in 2023 and 5,629 in 2024, with annual average employment of 37,624 and average annual pay of $31,066 in 2024 (BLS, QCEW annual averages, NAICS 721211, 2020 to 2025).

Two features of the count deserve a lender's attention before any demand question. First, supply grows slowly. An asset class that added 9% to its employer establishment count across four years that included the largest surge of outdoor recreation demand in the modern record is an asset class where entitlement, land and seasonality constrain the supply response; the demand shocks of 2020 and 2021 were absorbed mostly by existing sites. Second, the employer count is a floor. The QCEW annual employment figure of 37,624 is an average of a series that swings violently within each year, and the smallest campgrounds, family-run and unstaffed in winter, drop below the payroll threshold entirely; the Nonemployer Statistics, released 15 May 2025 for 2023, carry the no-payroll operators the employer series cannot see (U.S. Census Bureau, Nonemployer Statistics 2023, 2025). Even the pay figure carries the season in it: an average annual pay of $31,066 is not a statement about low wages so much as about short years, the arithmetic of a workforce most of which is employed for five months and paid for five.

The season, measured: payroll doubles every summer

The reason averages mislead in this asset class is visible in one row of public data. Campground employment in 2024, month by month, from the Quarterly Census of Employment and Wages: 25,757 in January; 26,025 in February; 28,112 in March; 33,698 in April; 43,598 in May; 50,763 in June; 53,061 in July; 50,945 in August; 44,046 in September; 38,198 in October; 29,705 in November; 27,576 in December (BLS, QCEW quarterly files, 2024, NAICS 721211, private, 2024 to 2025). The July peak is 2.06 times the January trough. No other property type in this series shows a payroll season that steep in the federal record; hotels, the nearest comparator, flex a fraction as much, because a hotel heats its rooms in January and a campground closes its loops. The curve is not an estimate, a survey or an operator anecdote. It is the census of unemployment insurance records, published quarterly, five months in arrears, for the nation, every state and most counties.

For demand analysis the curve does three jobs. It converts the word seasonality into a number that can be compared across geographies: the same July-to-January ratio computed from a state's or county's QCEW rows tells the analyst whether the local season is the national 2-to-1, a northern 4-to-1 or a Sun Belt inversion where winter is the peak, and the county rows exist wherever disclosure thresholds allow. It dates the shoulder: the climb from 33,698 in April to 50,763 in June and the fall from 44,046 in September to 29,705 in November mark the eight to ten weeks on each side of summer where marginal demand decides a park's year. And it disciplines the pro forma: a twelve-month revenue projection for a property whose industry runs half its January payroll is a projection about roughly 100 operating days, and the debt service those days must carry is the underwriting question, not the annual average that hides it.

Where the season lives: a geography inverted

Every other asset class in this series concentrates where people are. Campgrounds concentrate where they are not, and the state figures put a factor on the inversion. Dividing the 2023 establishment count by the July 2023 population gives 1.48 RV park and campground establishments per 100,000 residents nationally, and a spread wider than any property type this series has measured: Maine at 10.06 per 100,000, South Dakota at 9.26, Wyoming at 8.21, Montana at 7.43, Vermont at 6.18 and New Hampshire at 5.49 at the top; Maryland at 0.45, Illinois at 0.56, Georgia at 0.63, Massachusetts and Kansas at 0.68 at the bottom (U.S. Census Bureau, County Business Patterns 2023 state file, 2025; Vintage 2025 Population Estimates, 2026; MMCG computation). Maine's density is 22 times Maryland's. For car washes the same computation spans a factor of three; for self-storage, about two. The spread is the demand model in miniature: a campground's customers are not its neighbors, so per-capita supply says almost nothing about local saturation and almost everything about which states are destinations. Among the ten most populous states the figures stay low and tight, from Michigan's 2.20 and Pennsylvania's 1.61 down to Illinois's 0.56 and Georgia's 0.63, with California at 0.91 despite holding 357 establishments, the second-largest count after Texas's 484.

The second map confirms the first with a different instrument. The American Community Survey counts housing units held for seasonal, recreational or occasional use: 4,342,127 of them in 2024, 3.0% of the nation's 146.7 million housing units (U.S. Census Bureau, ACS 2024 1-year estimates, tables B25004 and B25001, 2025). By state the share runs from Maine's 14.4% and Vermont's 13.2% through New Hampshire's 8.9%, Alaska's 8.2%, Delaware's 7.6% and Florida's 7.0%, which at 741,429 units is the largest absolute stock of seasonal homes in the country, then Montana at 6.9%, Wisconsin at 5.8% and Michigan at 5.0% with 234,649 units. Set the two rankings side by side and they are nearly the same list. The campground map is the second-home map, not the population map, because both record the same underlying fact: where Americans go when they travel for leisure, and where the land, water and season reward a property type that sells outdoor nights. For a specific site, the county-level version of B25004 in the five-year file is the fastest public test of whether the location is a destination at all.

Read together, the two maps also answer the siting question the per-capita figure cannot. A high-density, high-seasonal-share county is a destination market where competition is the constraint and the season is the risk; a low-density county adjacent to a high-share county is the gap the corridor logic looks for, close enough to the destination to draw its overflow and far enough to own its frontage. The pairing is computable for every county in the country from two public tables and one division, which makes it the rare siting screen that costs an afternoon rather than a report.

The demand base that travels

The activity the parks monetise is valued in the national accounts. The Bureau of Economic Analysis' outdoor recreation satellite account for 2024, released on 5 March 2026, puts outdoor recreation's value added at $696.7 billion, 2.4% of GDP, up from $639.5 billion and 2.3% in 2023; real outdoor recreation GDP grew 2.7% in 2024, compensation grew 5.2% and employment 1.1% (BEA, Outdoor Recreation Satellite Account, 2026). Within the conventional activities, RVing is the second largest at $27.5 billion of value added, behind boating and fishing at $38.4 billion and ahead of hunting, shooting and trapping at $16.5 billion and snow activities at $7.6 billion. These are production-side measures, not park revenues, and the honest use is as a demand-base trend: the activity that fills RV pads has grown through every year of the account's recent record, in current and real terms, and its geography can be read in the same release, which publishes value added and employment for every state.

The destination anchors are counted too. The National Park Service recorded 323,014,305 recreation visits across 406 reporting parks in 2025, with 13,016,577 overnight stays and 26 parks setting individual visitation records, a total 2.7% below 2024's record of 331.9 million visits in a year that included a 43-day federal funding lapse. The 2024 record, 2% above 2023 and just past the prior mark of 330,971,689 set in 2016, went out through the agency's visitor use statistics system without a press release (National Park Service, 2025 visitation news release of 13 March 2026, and the NPS Visitor Use Statistics system, which carries the 2024 record and the park-level monthly detail). For a campground near a gateway, the park-level monthly series in the NPS visitor use statistics system is the single most relevant public demand series in existence: it is the destination's own turnstile, published monthly, park by park, back for decades, and its seasonal curve is the demand curve the property lives on. A memo for a park 20 minutes from a national park entrance that does not chart the entrance's own monthly visits has ignored the best data in the file.

The road season: the same curve on the highway

The season the payroll data measure at the campground gate, the traffic data measure on the road to it. Monthly travel on all roads and streets in 2025 ran from 238.7 billion vehicle miles in February to 297.1 billion in July, a summer peak 24.5% above the winter trough, with June through August each above 286 billion and November and December near 265 billion (FHWA, Traffic Volume Trends, December 2025, Table 2, 2026). The national curve is the shallow version; the rural-recreational corridor version is steeper, and the state traffic monitoring programs that feed the federal series publish the station-level seasonal factors that quantify it. The state DOT traffic counts article explains the machinery: continuous count stations record every month, and the published AADT is the annual average of a series whose July on a lake-country two-lane can run at multiples of its January. A campground's frontage AADT, adjusted with the state's seasonal factors for that functional class and region, is the honest traffic input; the unadjusted annual figure is the same number wearing a disguise.

One refinement matters in the South. The national curve peaks in July because the national inventory is northern, but the Sun Belt runs the season in reverse: the snowbird markets of Florida, Arizona and south Texas fill in January and slacken in August. The test is the same instrument pointed at a different row: the July-to-January ratio computed from a Sun Belt state's own QCEW rows, and from the county rows where disclosure allows, shows whether and how far the local season inverts, and a memo for a winter-peak park that quotes the national summer curve has measured the right thing in the wrong place.

Reading a catchment for a destination asset

The drive-time polygon that serves every metropolitan asset class in this library fails here, and the failure is instructive. A campground's demand does not live within fifteen minutes of the gate; it lives along the routes to the destination and in the metropolitan areas a tank of fuel away. The catchment for a destination asset is a corridor and an origin field: the corridor read from the route network and its seasonal traffic, the origin field read from the metropolitan populations within typical travel distance and their vehicle ownership. The ACS supplies the origin side, household counts and vehicles available by metro from the same tables the trade-area demographics article works with; the travel-center article works the corridor side with the same AADT discipline this article applies. What replaces the ring is a stated pair: the destination's own demand series (a park's visitation, a lake's access counts, a corridor's seasonal traffic) and the origin populations that feed it. Both halves are public, and both halves are checkable, which is more than the borrowed occupancy figure in an outlook report can say.

The local competitive set follows the same two-sided logic. Within the corridor, the inventory is built the way every inventory in this series is built, from the assessor roll and the permit record, with the federal establishment counts as the frame; the public-lands side of supply, the national forest, state park and Corps of Engineers campgrounds that compete with private sites at lower price points, is itself published in the agencies' recreation site inventories. A private park's demand statement that omits the public campground a mile up the shoreline has missed the largest competitor in its market, and the omission is unnecessary, because the public inventory is the easiest half to count.

The long-stay boundary: when a campground stops being one

The hardest classification question in this asset class is not architectural but temporal. A park that shifts sites from nightly and weekly stays to month-to-month and annual occupancy migrates, pad by pad, from hospitality toward housing, and the two destinations carry different demand bases, different regulation and different credit. The public record marks the boundary in three places. The industry definition itself: NAICS 721211 covers recreational stays, while residential pad rental belongs to the manufactured-housing community class, whose demand rests on the housing series and is treated in the manufactured housing article alongside the Census Bureau's Manufactured Housing Survey of monthly shipments and prices. The payroll curve: a true recreational park shows the 2-to-1 seasonal swing measured above, while a park full of annual residents shows a flat one, so the county QCEW curve set against the property's own staffing calendar is a fast consistency test. And the housing tables: where a park's occupants are year-round residents, they surface in the ACS as households, not in the seasonal-vacancy stock, and the county's balance between the two is readable directly from tables B25002 and B25004. A lender does not need to police the boundary for its own sake; it needs to underwrite the asset the property actually is, and the three public markers say which one that is more reliably than the borrower's brochure does.

The pipeline: permits, expansions and conversions

Campground supply changes in three ways, and the county record captures each. New parks arrive through entitlement and site development permits, and because a campground is land-intensive and infrastructure-light, the interval between permit and opening is short, often a single construction season, which makes the permit portal the only meaningful forward look the asset class has. Existing parks expand in loops, and the same portals date the utility and grading permits that precede new sites. And conversions run in both directions: seasonal parks add annual pads in weak markets, and well-located parks reclaim annual pads for nightly rates in strong ones, movements that leave traces in electrical and septic permits even when no new ground is broken. The discipline of counting is the familiar one from the rest of this series, an address-level inventory reconciled between the assessor roll, the permit record and the employer counts, with one addition specific to destination assets: the public agencies' own development plans. A state park adding 200 sites or a national forest closing a campground for hazard-tree removal moves a private corridor's competitive balance as much as any private project, and both are announced in public documents long before they happen. The corridor inventory that carries private and public sites, each with its season length and its dated pipeline, is the supply half of the memo; the measured season and the origin field are the demand half; and the two halves reconcile on the same calendar, which is the property type's entire economics in one sentence.

What the lender reads

A campground loan is hospitality credit with agriculture's calendar, and the lender's file should read like it. The nearest underwriting comparator is the hotel, whose public demand evidence the hotel demand article assembles, but the comparison sharpens at exactly the point the payroll curve measured: a hotel's low season still produces revenue, while a northern campground's produces close to none, so annual debt service is carried by a season whose length the QCEW curve and the destination's visitation series date to the week. The consequences are mechanical. Reserves and payment structures should follow the cash calendar rather than fight it; a stress test that moves the season's start two weeks later and its end two weeks earlier is worth more than one that shaves five points off a notional occupancy; and the expansion case that converts seasonal sites to annual leases changes the asset's class as much as its cash flow, moving it toward the manufactured-housing economics that carry their own demand analysis in this library. On classification, the special purpose examples reported from SBA guidance on SOP 50 10 8 name hotels, motels and marinas among properties requiring the higher 504 contribution; the reported summaries this series relies on do not settle the campground's own treatment, and the SOP itself governs (504 Capital Corporation, 2025; NAGGL, 2025). What the public record contributes to that conversation is the seasonality evidence itself: a payroll curve, a visitation curve and a traffic curve, each sourced and dated, that let a credit committee see the season it is being asked to finance instead of adjectives about it.

The public SBA datasets close the loop on performance. The 7(a) and 504 loan registers, which MMCG Analytics' SBA layer is built on, carry two decades of lending to NAICS 721211 operators, origination by origination, subject to the standing rule that no performance rate is shown for any cohort of fewer than ten loans. MMCG Analytics supplies the data and the analysis; the credit decision rests with the lender.

Method: the five numbers a campground memo should carry

First, the season's depth: the July-to-January employment ratio for the property's state or county from the QCEW monthly rows, set beside the national 2.06, with the operating calendar the borrower actually runs. Second, the destination series: the anchor's own published demand, national or state park monthly visitation for a gateway property, the corridor's seasonal traffic from the state DOT for a route property, stated for the latest full year and the five-year trend. Third, the origin field: households and vehicles available in the metropolitan areas within reasonable travel distance, from the ACS, with the drive time to each stated; a park three hours from four million households is a different asset from one three hours from four hundred thousand. Fourth, the competitive inventory: private parks from the assessor roll and permit record within the corridor, public campgrounds from the agency inventories, each with site counts and season lengths. Fifth, the demand-base trend: the BEA outdoor recreation account's RVing value added and the state's outdoor recreation figures, with the regional income data for the origin metros, because leisure spending follows income and the origin field's income trend is the campground's revenue trend one lag behind.

Each is a measurement with a source and a date. What the five deliberately exclude is any borrowed occupancy or rate figure, because none exists in the public record for this asset class, and the provenance standard that governs this library treats an unsourceable number as an omission, not an input. The gap is real and it is stated; the seasonality, the destination demand, the origin field, the inventory and the spending base are all better evidenced than the missing statistic, and together they answer the question the occupancy figure pretends to.

A worked sequence for one park

Run for a 120-site park twenty minutes from a national park entrance, the sequence goes as follows. The analyst pulls the park unit's monthly recreation visits for the last five years from the NPS visitor use statistics system and charts the season: say a June-to-August peak carrying 55% of annual visits, a shoulder adding 30%, and a November-to-March floor near zero. The state's QCEW rows for 721211 confirm the same shape in payroll. The state DOT's continuous count station on the access highway gives the seasonal traffic factors, and the frontage AADT adjusts accordingly. The origin field is read from the ACS: the three metros within four hours hold 2.1 million households at 1.9 vehicles per household, and the BEA state accounts show outdoor recreation value added in the origin states growing in real terms. The inventory: eleven private campgrounds and four public ones inside the corridor, 1,840 sites in total, 310 of them added since 2019 per the permit record. The memo's demand statement then writes itself in measured lines: a destination drawing roughly three million visits on a dated monthly curve, an origin field of stated size and trend, a competitive inventory of stated depth and season, a payroll season of stated ratio, and a pro forma whose revenue calendar can be laid directly over the visitation curve it depends on. Whether the coverage works at the stated season length is then arithmetic, and every input of the arithmetic is public.

The same sequence exposes the weak deal as quickly. A park on a corridor whose destination series has flattened, whose origin metros are losing households, and whose permit record shows 20% of the corridor's sites added in five years is a market where the season will be fought over rather than shared, and the analyst can show it before closing rather than discover it at the first November payment.

The 2026 read, and the cadence

Read in mid-2026, the public series describe a demand base that is large, grown and normalising rather than surging: outdoor recreation at 2.4% of GDP and rising in real terms, RVing's $27.5 billion of value added second among conventional activities, national park visitation within 3% of its record despite a funding lapse, and a supply side that added establishments at barely 2% a year through the strongest demand period the asset class has recorded. Nothing in those series prices a park. What they set is the posture: demand evidence for this class is seasonal, destination-anchored and origin-fed, and all three legs publish on a knowable calendar, the QCEW quarterly, the NPS monthly, the traffic counts continuously, the ACS and BEA annually. The marina article applies the same destination logic to the water, where a state registration file replaces the visitation series; the asset-class pillar holds the whole family to the same rule. For the analyst, the practical summary is four bookmarks and one discipline: chart the season from the payroll and visitation series, size the origin field from the ACS, count the corridor from the permit record and the public inventories, and let the annual average stay out of the memo.

Frequently asked questions

What public data measures RV park and campground demand?

Four families of series: the QCEW's monthly campground payrolls (NAICS 721211), which measure the season directly; destination series such as National Park Service monthly visitation; origin-field demographics from the ACS, including vehicles available and seasonal-home counts; and the BEA's outdoor recreation satellite account, which values RVing yearly at national and state level. None publishes an occupancy rate.

How seasonal is the campground business?

Campground employment in 2024 ran from 25,757 in January to 53,061 in July, a 2.06-to-1 swing, the steepest payroll season of any property type in this series. The ratio can be recomputed from QCEW rows for most states and many counties, and northern markets run steeper than the national curve.

How many RV parks and campgrounds are there in the United States?

County Business Patterns counted 5,018 establishments with paid employees in 2023, up 9.4% from 2019; 69.2% have fewer than five employees. Campgrounds without payroll appear in the Nonemployer Statistics instead, and public campgrounds operated by federal and state agencies sit outside both counts.

Which states have the most campgrounds per capita?

On 2023 counts against July 2023 population: Maine (10.06 per 100,000 residents), South Dakota (9.26), Wyoming (8.21), Montana (7.43) and Vermont (6.18), against a national 1.48 and Maryland's 0.45. The 22-fold spread reflects a destination asset: supply concentrates where visitors go, not where residents live.

What do seasonal-home counts add to a campground analysis?

The ACS counts 4,342,127 housing units held for seasonal or recreational use, 3.0% of the national stock, with state shares from Maine's 14.4% down. The seasonal-home map and the campground map are nearly the same list, so the county-level seasonal share is a fast public test of whether a location is a destination.

Is there a public occupancy rate for RV parks?

No federal series measures campground occupancy or site rates. A defensible memo builds the demand case from the measured season (payroll and visitation curves), the origin field, and the corridor inventory, and treats occupancy as the pro forma's stated assumption, stress-tested on season length rather than presented as a market fact.

Sources

1. U.S. Census Bureau, County Business Patterns 2023, national and state files, NAICS 721211, released 26 June 2025; County Business Patterns 2017 and 2019 national files, released 2019 and 2021. https://www2.census.gov/programs-surveys/cbp/datasets/
2. U.S. Bureau of Labor Statistics, Quarterly Census of Employment and Wages, quarterly files 2024 (monthly employment) and annual averages 2019 to 2024, NAICS 721211, private, United States, 2020 to 2025. https://data.bls.gov/cew/data/api/2024/1/industry/721211.csv
3. U.S. Census Bureau, American Community Survey 2024 1-year estimates, tables B25004 Vacancy Status and B25001 Housing Units, released 11 September 2025 (table-based summary files). https://www2.census.gov/programs-surveys/acs/summary_file/2024/table-based-SF/data/1YRData/acsdt1y2024-b25004.dat
4. Bureau of Economic Analysis, Outdoor Recreation Satellite Account, U.S. and States, 2024, released 5 March 2026. https://www.bea.gov/data/special-topics/outdoor-recreation
5. National Park Service, 2025 visitation news release (323,014,305 recreation visits, 13,016,577 overnight stays, 26 park records), 13 March 2026; NPS Visitor Use Statistics system for park-level monthly series and the 2024 total. https://www.nps.gov/orgs/1207/03-13-26-2025-visitation-statsitics.htm
6. Federal Highway Administration, Traffic Volume Trends, December 2025, Table 2 (monthly vehicle miles 2025), 2026. https://www.fhwa.dot.gov/policyinformation/travel_monitoring/25dectvt/25dectvt.pdf
7. U.S. Census Bureau, Vintage 2025 Population Estimates, state totals (NST-EST2025-ALLDATA), July 2023 populations, 2026. https://www2.census.gov/programs-surveys/popest/datasets/2020-2025/state/totals/NST-EST2025-ALLDATA.csv
8. U.S. Census Bureau, Nonemployer Statistics 2023, released 15 May 2025. https://www.census.gov/newsroom/press-releases/2025/2023-nonemployer-statistics.html
9. 504 Capital Corporation, SBA 504 Loans for Special Purpose Properties and Real Estate (guidance on SOP 50 10 8 classifications), 2025. https://504capital.com/blog/financing-special-purpose-properties-sba-504-loans/
10. National Association of Government Guaranteed Lenders, SBA Notice Revising SOP 50 10 8 (Procedural Notice 5000-872764), 2025. https://www.naggl.org/sba-notice-revising-sop-50-10-8/
11. MMCG Research, SBA 7(a) Performance Series: MMCG analysis of the public SBA 7(a) loan register, 2026. https://mmcganalytics.com/sba-default-rates/
12. Bureau of Economic Analysis, Outdoor Recreation Satellite Account 2023 release (value added $639.5 billion, 2.3% of GDP), 20 November 2024. https://www.bea.gov/news/2024/outdoor-recreation-satellite-account-us-and-states-2023