Why It Matters
You're evaluating a market you've never invested in. The broker says rents are strong and demand is rising. Before you trust the pitch, you pull the ACS. What you find: median household income in the target zip code is $38,200, down 4% over three years. Renter-occupied housing is 71%. Vacancy has crept from 6.2% to 8.9% over five years. That's not a strong market — that's a distressed one with eroding fundamentals. The ACS didn't tell you whether to buy or not. But it told you the broker's story needed more scrutiny. That's exactly what this dataset does: it gives you the demographic and economic baseline that CoStar, Realtor.com, and Redfin don't provide. Use it before committing to any new market.
At a Glance
- Conducted by: U.S. Census Bureau — part of the federal statistical system
- Release schedule: 1-year estimates (areas 65,000+ population); 5-year estimates (all geographies, including small tracts and zip codes)
- Key data for investors: Rental vacancy rate, median household income, housing tenure (own vs. rent), gross rent as % of income, population change, household size
- Geographic depth: State → county → city → zip code → census tract (roughly 4,000 people per tract)
- Lag: 1-year estimates released roughly 12 months after the reference year; 5-year estimates released December annually
- Cost: Free via data.census.gov and the Census Bureau API
How It Works
What the ACS actually measures. The Census Bureau surveys approximately 3.5 million households per year — every address in the country cycles through over a 5-year rolling window. Respondents answer questions on housing costs, employment, income, commute, education, family structure, and whether they own or rent. This continuous sampling is what makes the ACS different from the decennial census: it produces annual estimates, not once-per-decade snapshots. For investors, the relevant tables are B25002 (occupancy status), B25003 (tenure), B25070 (gross rent as percentage of household income), B19013 (median household income), and B25004 (vacancy by reason).
The 1-year vs. 5-year tradeoff. The 1-year ACS covers geographies with 65,000 or more people, which means most smaller cities, suburbs, and rural markets won't appear in 1-year estimates at all. For those areas — and for zip code or census tract analysis anywhere — you need the 5-year estimates, which pool five years of survey data to produce statistically reliable figures for small geographies. The tradeoff is currency: 5-year estimates released in December 2024 cover the 2019–2023 window. That's not the same as current. For fast-moving markets, supplement ACS baseline data with more recent indicators from FRED and BLS.
Housing tenure as a demand signal. One of the most investor-useful ACS tables is the owner vs. renter breakdown by geography. A zip code where 65% of households rent tells you that the rental market is deep and established — landlords have a large addressable tenant pool. A zip where 85% of households own tells you renting is culturally non-dominant in that area, which affects both absorption of new units and tenant quality expectations. Neither is automatically good or bad. But you need to know which market you're entering before you underwrite a deal.
Income trends as affordability stress-testing. The ACS median household income figure is your affordability anchor. Take your target monthly rent and divide by 0.3 (the standard "30% of income" threshold). The result is the monthly income a tenant needs to afford your unit without being rent-burdened. If that number exceeds the local median, you'll have trouble attracting stable, qualified tenants. ACS table B25070 makes this explicit — it shows what percentage of renters in a tract are already spending more than 30% of income on housing. High rent-burden percentages signal either a supply-constrained market with pricing power, or a distressed market with weak income foundations. Realtor.com and Redfin data help you distinguish between the two.
Real-World Example
Jessica is underwriting a 24-unit apartment complex in a mid-sized Midwest city. The listing broker says the area is "working-class but stable." Before she runs her proforma, she pulls the 5-year ACS estimates for the target census tract.
What she finds: median household income is $41,700, up just 1.1% annualized over five years (below inflation). Renter-occupied housing is 68%. Gross rent as a percentage of household income shows 38% of renters spending more than 30% of their income on rent — already rent-burdened. Vacancy in the tract is 7.4%, compared to a metro average of 4.9%.
Jessica revises her underwriting. She drops the broker's projected rent growth from 4% annually to 1.5%, brings vacancy assumptions up from 5% to 8%, and cuts her expected NOI by $31,000 per year. At the seller's asking price, the deal no longer pencils. She passes. Six months later, a buyer who didn't run the ACS buys it — and spends the next two years fighting vacancy that the data predicted.
Pros & Cons
- Free, government-sourced data with no subscription required — unlike CoStar, which runs $5,000–$10,000+ per year for comparable depth
- Covers all U.S. geographies including small markets, rural areas, and micro-neighborhoods that private data providers don't track with precision
- Income, tenure, and vacancy data in one source — the three fundamentals you need to stress-test any residential rental market
- 5-year estimates smooth out year-to-year sampling noise, making trend comparisons more reliable than single-year private market reports
- Data lag is significant — 5-year estimates released in late 2024 cover 2019–2023, which misses post-pandemic rent spikes and corrections in fast-moving metros
- Survey-based, not transaction-based — vacancy and rent figures reflect self-reported household data, not actual listing or lease transactions the way Redfin or CoStar track them
- The Census Bureau interface (data.census.gov) is notoriously difficult to navigate; extracting specific table-variable combinations requires knowing the exact table IDs (B25002, B25070, etc.)
- 1-year estimates are unavailable for most small markets and suburbs, limiting current-year visibility in precisely the tertiary markets where investors chase yield
Watch Out
Trend direction matters more than snapshots. A 7% vacancy rate is fine in a market trending toward 4%. It's a warning sign in a market trending from 4% toward 9%. Always compare at least two ACS reference points — most data.census.gov queries allow side-by-side year comparisons. A single-year pull misses the most important signal: which direction is the market moving?
Don't conflate metro ACS with neighborhood ACS. A city-level vacancy rate of 5% can mask a census tract with 12% vacancy three miles from the deal you're evaluating. Always drill to the tract or zip code level for residential analysis. Metro-level figures are useful for context; tract-level figures are what you underwrite against.
ACS is a baseline, not a substitute for transactional data. The survey captures demographics and self-reported conditions. It won't tell you what comparable units actually rented for last month, or whether a new employer just announced 2,000 jobs in the submarket. Use ACS alongside FRED for macro indicators and BLS employment data to build a complete market picture before committing capital.
Ask an Investor
The Takeaway
The ACS Survey is the most granular free dataset available for residential real estate market research. It gives you the income, tenure, and vacancy baseline at the zip code and census tract level that makes your underwriting assumptions defensible — or reveals they aren't. Pull it before you enter any new market, supplement it with FRED and BLS for employment and economic context, and triangulate against Redfin or CoStar for transactional rent data. The investors who skip it are underwriting with one eye closed.