Nareit Monthly Index Values & Returns - Complete History XLS (1972-2026)

The month-end heartbeat of the US REIT market, unbroken since December 1971. Nareit - the trade association for the REIT industry - maintains the complete monthly history of the FTSE Nareit U.S. Real Estate Index Series in one Excel workbook, and Datadory delivers those histories as typed, analysis-ready rows.

The 'Index Data' sheet runs 665 rows by 42 columns: one month-end per row, six sector blocks across the columns - All REITs, Composite, Real Estate 50TM, All Equity REITs, Equity REITs and Mortgage REITs - each block carrying Total Return, Price Return, Income Return and Index level sub-columns. A companion 'Var' sheet defines every published variable, and a slimmer second workbook isolates the current year.

The scale reads best in the index column itself: All REITs stand at 100 in December 1971; the composite closes at 11,960.151 on July 31, 2026. Between those two numbers sits every drawdown and recovery the asset class has printed, at monthly resolution, including the leveraged mortgage half of the market that rent-collecting equity REITs obscure.

Three slices of the same sheet, exactly as captured during the August 2026 research pass:

# one row: the base month, everything measured from here
File  : the historical workbook     Sheet: Index Data
Date  : 1971-12-31
        all_reits_index : 100            # rebased at inception

# January 1972: the first measured month, full float precision intact
Date         : 1972-01-31
total_return : 1.2203525 %    price_return  : 0.3268945 %
income_return: 0.893458 %     index         : 101.2203525

# February 1972 compounds straight onto it
Date         : 1972-02-29
total_return : 0.9496795 %    index         : 102.1816214

# the newest row in the window: composite, not the base block
Date             : 2026-07-31
composite_index  : 11960.151       # against the December 1971 base of 100
monthly_total_return : 2.07293 %

Two things worth noticing. First, precision survives: returns arrive at full float depth (1.2203525000000104, not 1.22), so rounding decisions stay yours downstream. Second, the base block and the composite answer different questions - the All REITs columns speak to how REITs as an asset class compounded, while the July 2026 composite row shows where fifty-four years landed - and one Date column keys all forty-two columns together.

Seven field groups cover every observation, splitting into a time key, three measure families and a column-block structure.

The time key is Date - one month-end per row, December 31, 1971 through July 31, 2026.

The measure families repeat inside every block: Total Return with dividends reinvested, Price Return without them, Income Return for the dividend share alone, and Index, the compounded level rebased to 100 at the December 1971 base. Because the trio is mechanical, decomposing any month into income versus price is subtraction, never estimation. In January 1972, for instance, All REITs returned 1.2203525 percent in total, of which 0.893458 points were income - roughly 73 percent of that month's return arriving as dividends.

The column blocks are the entities being measured: six investment-sector aggregates including the dedicated Mortgage REITs block, laid out across the sheet's 42 columns, with the 'Var' sheet documenting what each one means. The total return versus price return distinction is the load-bearing one for anyone modeling REIT income.

The seven-group spine covers every published cell; the extensions below become additional fields on request because they depend on the cut you specify:

1. Rate-conditioning joins. The return panels aligned against Federal Reserve H.15 benchmark series - fed funds against 10-year constant maturity gives the classic steepness factor behind mREIT book-value swings.
2. Collateral context. Quarterly Z.1 mortgage-debt balances joined alongside returns, so the sector's pricing can be read against the stock of mortgages financing it.
3. Constituent resolution. Membership lists reaching back to 1991 from the companion Nareit indexes record, mapping a sector-month cell toward the companies that produced it.
4. Derived views computed on delivery. Growth-of-$100 curves, rolling windows, drawdown series, rolling dividend yields or mortgage-versus-equity spread panels rebuilt to your date range.

• Geography: the United States - specifically the constituents of the FTSE Nareit U.S. Real Estate Index Series, the listed US REIT universe. A national benchmark, not a global composite.
• Temporal: December 1971 through July 2026 - 665 consecutive month-end observations in the historical sheet, with a companion workbook isolating January through July 2026. Over five decades, spanning every REIT-market cycle on record at monthly grain.
• Granularity: one row per month-end, six sector blocks wide, each with its Total/Price/Income/Index quartet riding the same row. No interpolation, no modeled fills - the sparse early years are simply the years before the series existed.

One honest boundary: these 42 columns carry investment-sector blocks only. Property-sector and subsector breakouts - Office, Data Centers, Apartments and peers - live in the annual Nareit returns record from 1994 onward rather than inside this workbook.

API, files, or your warehouse. Daily, weekly, or hourly.

You pick the channel and the cadence; the field semantics above hold across all three. Rows arrive flattened - one observation per month-end with every block's measures normalized into typed numeric columns, the variable definitions already resolved - so no spreadsheet-cell wrangling reaches your pipeline. Most teams take the full 665-month history once; it is small enough to hold in memory whole, and each subsequent pull appends cleanly because the schema never moves. Reshaping, filtering and derived panels are cut at delivery, not left as homework.

• Asset allocators and quant researchers treat the monthly investment-sector panels as the canonical REIT performance record, backtesting equity-versus-mortgage allocations across five decades of months rather than a handful of annual prints.
• Risk managers convert the compounded index columns into drawdown and volatility series per sector, measured on a consistent grid since 1971.
• Fixed-income and rates analysts condition mortgage REIT returns on funding-rate regimes - the spread business mREITs are actually in.
• Journalists, academics and students reach for the series whenever a claim needs the industry's own monthly numbers behind it, methodology attached.

The quant backtesting workflow is where the record earns its keep, and the deeper narrative lives in our guide to historical mortgage REIT total returns.

• Investors & Quant Researchers score highest at relevance 4, the strongest rating this dataset earns across the catalog's persona packs: a fixed-schema monthly panel with dividends separated from price feeds allocation and drawdown models directly.
• Journalists, Academics & Students land at relevance 3 on citation strength - the quotable version of REIT history, one publisher deep.
• Market Researchers & Consultants, Developers & Data-Product Builders and Data Scientists & ML Engineers each carry relevance 2: a citable benchmark, a stable performance spine for products, and a clean monthly panel for regime modeling respectively.

If you are weighing it against the wider shelf, the best mortgage REITs datasets ranking places it third of seven in this slice, and the investors and quants pack shows how it pairs with the Fed releases that flank it.

Three pairings complete the picture. The mREIT Sector Overview supplies the live roster - roughly 30 vehicles with an aggregate yield of 13.14 percent as of month-end July 2026 - so the historical return stream meets the companies currently printing it. The indexes and market-performance landing adds the 36 annual constituent lists back to 1991, which is how survivorship gets caught before it flatters a backtest. And when the packaged-exposure view is useful, the REM ETF record tracks the FTSE Nareit All Mortgage Capped Index as a single instrument.

Two knowable caveats. Property-sector detail lives in the annual record, not these columns. And where this record overlaps the broader Nareit Data & Research Hub, the tradeoff is simple: one deep 665-month table against a doorway holding many artifacts.