A Real Estate Investment Trust (REIT) is a pass-through company that owns income-producing real estate and must, under U.S. tax law, distribute at least 90% of taxable income as dividends. But "REIT" is not one asset — it is a dozen-plus distinct businesses sharing a tax wrapper. Nareit's FTSE Nareit index family recognizes roughly 14 equity-REIT sectors (data centers, diversified, gaming, health care, industrial, lodging/resorts, office, residential, retail, self-storage, specialty, telecommunications, timberland — plus mortgage REITs separately). The six property types in this node's title are the largest and most representative. Their core tension: each sector has a completely different tenant base, lease structure, demand driver, and capital intensity, so treating REITs as a homogeneous "rates-sensitive yield play" is the most common analytical error. An office REIT and a data center REIT can move in opposite directions in the same quarter.

The sectors

Residential — apartments, single-family rentals, student housing, manufactured housing. Tenants are individuals on short leases (typically 12-month apartment leases, monthly for SFR). Demand driver: household formation, job/wage growth, and the rent-vs-own spread. Short leases mean rents reprice quickly — a tailwind in inflation, a vulnerability in soft markets. Operating-expense ratios are structurally high (tenant turnover and on-site management are costly), in contrast to the low-opex net-lease sectors.

Retail — regional malls, grocery-anchored centers, outlet centers, power centers, and free-standing net-lease (single-tenant). Tenants are merchants. Demand driver: consumer spending, e-commerce penetration, and anchor health. Single-tenant net-lease retail uses triple-net (NNN) leases (see below); malls and centers use gross or modified-gross leases plus percentage rent tied to tenant sales.

Office — CBD towers, suburban parks, and life-science/lab buildings. Tenants are businesses on long leases (commonly 5-10+ years). Demand driver: white-collar employment and, structurally, the share of work done in-office. Operating-expense ratios are structurally high (landlord-borne building services under gross/modified-gross leases).

Industrial — warehouses, distribution/fulfillment centers, logistics, cold storage. Tenants are logistics, e-commerce, and manufacturing firms. Demand driver: goods consumption and e-commerce, which requires roughly 3x the warehouse space of brick-and-mortar retail per dollar of sales (a widely cited logistics rule of thumb). Operating-expense ratios sit at the low end (NNN-style leases push costs to tenants).

Data center — highly specialized facilities housing servers, with redundant power, cooling, and physical security. Tenants are cloud and enterprise IT ("hyperscalers" like the major cloud providers, plus colocation customers). Demand driver: cloud computing and, recently, AI training/inference. Capital-intensive and power-constrained; leases are long with built-in escalators.

Healthcare — senior housing, skilled nursing, medical office buildings (MOBs), hospitals, life-science labs. Demand driver: demographics (population aging). Structure is bifurcated: NNN-leased properties (rent from an operator) versus RIDEA structures, where the REIT shares operating income/risk of senior-housing operations directly — a critical distinction, because RIDEA exposes the REIT to operator margins, occupancy, and labor costs, not just contractual rent.

Lease structure — the single most important differentiator

Lease type determines who bears operating cost and how cash flow behaves:

• Triple-net (NNN): tenant pays rent plus property taxes, insurance, and maintenance. Common in single-tenant retail, much of industrial, data centers, and NNN healthcare. Produces stable, bond-like, long-duration cash flow with low landlord opex and built-in escalators — but limited upside when market rents rise and concentration risk if a single tenant defaults.
• Gross / modified-gross: landlord pays most operating expenses; rent is higher to compensate. Standard in multifamily and traditional office. One academic study of U.S. REITs found gross leases collected an average ~12.9% rent premium over net leases, while NNN portfolios were more operationally efficient — a real trade-off, not a free lunch.

Short-lease sectors (residential) reprice with the market fast; long-NNN sectors (net-lease retail, some healthcare/data center) deliver predictability but lag inflation unless escalators keep pace. Weighted Average Lease Term (WALT) is the key disclosed metric for long-lease REITs.

How it's used in practice

Analysts compare REITs within a sector, not across, because the drivers differ. The universal valuation metric is Funds From Operations (FFO) = net income + real-estate depreciation − gains on property sales (depreciation is a non-cash charge that distorts GAAP earnings for asset-heavy businesses). AFFO/Adjusted FFO further subtracts recurring maintenance capex and straight-lined rent — usually the better dividend-coverage gauge. Valuation runs on P/FFO multiples and implied cap rate (NOI ÷ enterprise value) versus private-market cap rates; a REIT trading at a wide premium/discount to its underlying NAV is the classic public-vs-private arbitrage signal. Sector selection is effectively a thematic macro bet: secular winners (data centers on AI/cloud, industrial on e-commerce) versus structurally challenged ones (office under hybrid work).

Adoption, debate & evidence

Sector divergence is now extreme and well documented. Through Q3 2025, office REITs posted roughly -19.7% YTD returns with negative FFO growth and traded around 8.5x P/FFO, while data centers traded near 28.3x P/FFO on AI-driven demand (paperfree.com / Nareit data). For full-year 2024, Nareit-cited figures put specialty at ~+35.9%, data centers ~+25.2%, and healthcare ~+24.2% total return — but industrial was down ~17.7% that year despite its strong secular story, a reminder that a good thesis and a good year are not the same thing.

The genuinely contested points: (1) Office — bulls argue current discounts overprice permanent impairment; bears argue hybrid work is a structural, not cyclical, demand cut. Unresolved. (2) Data centers — the demand is real, but power availability, hyperscaler capex cyclicality, and obsolescence risk on AI-specific buildouts are debated; high multiples price in sustained growth. (3) Rate sensitivity — REITs broadly show a negative correlation to the 10-year Treasury yield (driving the Q4 2024 selloff), but the strength varies sharply by sector: long-WALT NNN names trade most like bonds, while short-lease residential and operationally-driven data centers are less purely rate-bound.

Strengths & limitations

The sector framework's strength is that it forces the right peer set and reveals that "REITs" is a misleadingly broad label. Limitations: sector labels hide structure — two healthcare REITs with NNN vs RIDEA exposure are nearly different businesses; "diversified" and "specialty" REITs blur the taxonomy entirely. The single most common misuse is treating all REITs as one interest-rate trade and ignoring that the demand driver (consumer spending, employment, e-commerce, AI capex, demographics) often dominates the rate signal sector-by-sector. Capital intensity is another trap: REITs must constantly raise debt/equity to grow, so cost of capital and balance-sheet leverage matter as much as property fundamentals.

System relevance

This node anchors the Real Estate (REITs) branch of the Sector & Industry Playbooks. It defines the taxonomy and structural drivers; sibling nodes covering FFO/AFFO mechanics, cap-rate/NAV valuation, and REIT rate-sensitivity should carry the formula depth — cross-link rather than duplicate. For the Augustus agent, the load-bearing caveat is the within-sector rule: a REIT ticker must be evaluated against its property-type peers and its lease structure (WALT, NNN vs gross vs RIDEA), never against the REIT index as a whole. Hard caveat: sector total-return figures above are point-in-time and reverse across cycles — use the structure as durable knowledge, the numbers as dated examples.

Sources

• Nareit — REIT Sectors (reit.com/what-reit/reit-sectors) and glossary; FTSE Nareit sector definitions and 2024 sector total returns
• S&P Dow Jones Indices — "Understanding REIT Sectors" (education PDF)
• ScienceDirect — "Triple-net leased property portfolios and operational efficiency: Evidence from the U.S. REIT market" (gross-lease ~12.9% premium; NNN efficiency)
• Holland & Knight / Rising Realty — NNN vs gross vs modified-gross lease cost allocation
• paperfree.com — "REIT Sector Performance Q3 2025" (office -19.7% YTD, 8.5x P/FFO; data center 28.3x P/FFO)
• Nareit / Fidelity — 2025 outlook; rate sensitivity and 10-year Treasury correlation
• Investopedia — REIT, FFO/AFFO, RIDEA definitions