Case-Shiller home price index explained

What Case-Shiller measures

The index was developed by economists Karl Case and Robert Shiller in the 1980s and is now maintained by S&P Dow Jones Indices using transaction data licensed from CoreLogic. Unlike a snapshot median — which asks “what did the typical home sell for this month?” — Case-Shiller asks: for homes that sold before and sold again, how much did the price change? That repeat-sales design controls for quality differences between a cheap condo sold in January and a luxury single-family home sold in June, giving a cleaner read on appreciation than NAR’s median existing-home price.

Only arms-length resales on single-family homes are included. New construction, foreclosures sold at distressed discounts (in some series configurations), condos (in the headline national index), and commercial property are excluded or reported in separate sub-indexes. Each qualifying pair of transactions enters a weighted regression (the S&P CoreLogic repeat-sales model) that produces month-over-month and year-over-year percent changes for geographic aggregates.

Why repeat sales matter for investors

• Apples-to-apples appreciation — measures price change on identical structures, not a changing sales mix.
• Long history — national data back to 1987; 20-city composite back to 2000; metro series vary by market depth.
• Policy relevance — the Bureau of Labor Statistics uses owners’ equivalent rent and rent of primary residence in CPI; Case-Shiller is the market benchmark economists cite when debating shelter inflation lags.
• Cycle context — peak-to-trough declines during 2006–2012 and the 2020–2022 surge are documented in Case-Shiller, not just anecdotal MLS medians.

Index family: national, 20-city, and metro tiers

S&P publishes several related series each month (typically the last Tuesday, 9:00 a.m. ET, on the economic calendar):

S&P CoreLogic Case-Shiller U.S. National Home Price Index

• Covers the entire United States; the broadest headline for macro traders.
• Published as NSA (not seasonally adjusted) and SA (seasonally adjusted) indexes.
• Base period: January 2000 = 100 for the national NSA series (check release notes — metro bases differ).

S&P CoreLogic Case-Shiller 20-City Composite

• Weighted average of 20 major metros (Boston, Chicago, Denver, Las Vegas, Los Angeles, Miami, New York, San Diego, San Francisco, Washington D.C., and others).
• Often moves more than the national index because coastal metros carry heavy weights and higher volatility.
• Includes a 10-City Composite for a shorter-history, large-metro read.

Individual metro indexes

Each of the 20 cities gets its own index. Sun Belt metros (Phoenix, Tampa, Miami) showed extreme year-over-year gains in 2021–2022; San Francisco and Seattle cooled first in 2023. Metro dispersion is a feature, not noise — national averages hide local affordability crises and builder opportunity.

Tiered indexes (Tier 1 and Tier 2)

S&P also publishes low-, middle-, and high-price tier indexes for select metros, splitting repeat sales by initial purchase price quintile. When low-tier prices outperform high-tier, it often signals investor activity or first-time buyer pressure in cheaper neighborhoods; when high-tier leads, luxury and equity-rich buyers are driving the market.

Seasonally adjusted vs not seasonally adjusted

Housing has strong seasonality: more closings in spring and summer, fewer in winter. The NSA index shows raw month-over-month moves — a January dip may be weather and holidays, not a crash. The SA index removes recurring seasonal patterns so December and June are comparable.

Media headlines usually cite year-over-year NSA percent change because it is intuitive (“prices up 4.7% from a year ago”). Macro researchers modeling month-over-month momentum often prefer SA. When trading the release, know which series the consensus forecast references — Bloomberg typically surveys YoY NSA for the 20-city composite.

Revisions

Case-Shiller revises prior months as CoreLogic feeds additional deed records. Revisions are usually small but can matter at turning points. Unlike GDP, there is no annual benchmark revision of comparable magnitude — still, compare the revision footnote in the press release before reacting to a one-tenth surprise.

Case-Shiller vs NAR median vs FHFA HPI

Three house price indicators dominate U.S. discourse; each answers a different question:

NAR median can rise when cheap homes stop trading (mix shift) even if every individual home’s price is flat — Case-Shiller corrects for that. FHFA covers more zip codes but misses cash buyers and expensive coastal jumbo loans, so it sometimes understates booms in high-cost markets. Use NAR for volume and affordability now; Case-Shiller for appreciation trend with a lag; FHFA for regulatory and GSE exposure.

Lag, release timing, and the shelter CPI link

Case-Shiller is published for transactions that closed roughly two months earlier — the April release reflects deals closing in February and March. That lag frustrates traders who want a live pulse but makes the index reliable: deed data is complete, not estimated from pending contracts.

Shelter is the largest CPI category (~35% of the basket). BLS measures shelter primarily through rent and owners’ equivalent rent (OER) surveys, not Case-Shiller directly. Economists still pair them because OER tends to follow market rents and house prices with a 12–18 month lag. When Case-Shiller YoY decelerates sharply for several months, markets often anticipate future shelter CPI cooling — a key input for Fed policy and TIPS breakeven trades, even though the transmission is slow and imperfect.

Relationship to housing activity indicators

• Housing starts lead Case-Shiller — builders respond to prices with a 6–18 month construction lag.
• Existing home sales volume can fall while Case-Shiller still rises (lock-in, low inventory).
• Mortgage rates affect affordability immediately; Case-Shiller confirms whether sellers actually cut prices or just stop listing.
• New home sales median prices reflect builder incentives and mix; Case-Shiller ignores new construction in the national index.

How investors use Case-Shiller release days

The monthly S&P release includes national and 20-city NSA and SA indexes, plus metro detail. Markets trade the surprise versus economist consensus and the breadth of metro gains.

Typical asset reactions

• Homebuilder equities (XHB, ITB) — accelerating YoY gains support builder margins and land bids; deceleration warns of order cancellations ahead.
• REITs — residential REITs benefit from rent growth tied to house prices; deceleration can pressure apartment REIT narratives if rent growth follows.
• Treasuries — hotter-than-expected Case-Shiller can sell off duration on shelter inflation persistence fears; cooler prints support rallies if growth scare dominates.
• Regional banks — mark-to-market on mortgage servicing and construction loan books; metro detail flags geographic credit stress.
• Consumer discretionary — wealth effect from rising home equity supports spending; falling prices hit home-improvement retailers in affected metros.

Long-horizon investors track six-month annualized rates and metro breadth (how many of the 20 cities show accelerating vs decelerating YoY) rather than trading every 0.1 percentage point miss.

Worked example: Harbor Realty monthly Case-Shiller read

Harbor Realty operates 40 brokerage offices across six Sun Belt and Midwest metros. Each Case-Shiller release morning, the research desk runs a twenty-minute checklist before updating buyer guidance and listing-price recommendations:

1. Read national and 20-city YoY NSA — if both decelerate two consecutive months, flag a pricing-power shift even if local MLS medians still show month-over-month gains.
2. Pull Harbor’s three covered metros (Tampa, Phoenix, Chicago) from the metro table — compare Case-Shiller YoY to internal closed-sale appreciation; divergence wider than 2 points triggers a data-quality review of Harbor’s sample.
3. Check tier splits where available — if low-tier Tampa accelerates while high-tier flatlines, adjust marketing toward entry-level inventory; investors may be crowding the bottom of the market.
4. Cross-check NAR existing-home sales from the prior week — falling volume plus still-rising Case-Shiller YoY confirms lock-in / low-supply regime; falling volume plus falling Case-Shiller signals demand destruction.
5. Note shelter CPI context — if Case-Shiller YoY is 6% but shelter CPI is still accelerating, Harbor warns sellers that Fed policy may stay tight longer; buyers get pre-approval stress tests at rates 50 bps above current quotes.
6. Write one paragraph for the broker newsletter — e.g. “20-city YoY 4.2% (down from 4.8%). Tampa 5.1%, Chicago 2.3%. National deceleration third month; recommend list-price reductions of 2–3% on homes above $600k sitting 45+ days.”

Harbor does not trade the release — they use Case-Shiller to time pricing strategy and buyer education before local MLS medians confirm the turn.

Indicator decision table

Common pitfalls

• Confusing index level with dollars — Case-Shiller is an index (2000 = 100), not median home price in dollars; use YoY or index rebasing for comparisons.
• Ignoring lag — the release describes closings two months ago; pair with mortgage rate moves since then for forward view.
• Using national index for local decisions — national +4% can hide San Francisco -2% and Miami +8%; always check metro tables.
• Expecting instant CPI impact — shelter CPI moves slowly; one hot Case-Shiller print does not mean next month’s CPI surprises to the upside.
• Comparing to new-home median — Case-Shiller national index excludes new construction; use Census new-home data separately.
• Overweighting month-over-month NSA in winter — January and February NSA dips are often seasonal; prefer YoY or SA for trend.
• Assuming price up = volume up — 2023–2025 showed prices resilient while existing-home sales hit cycle lows.

Investor checklist

• Track 20-city and national YoY NSA; note three-month deceleration or acceleration trend.
• Download metro table; count how many cities accelerated vs prior month.
• Compare Case-Shiller YoY to NAR median YoY for same period — large gap implies mix shift in NAR.
• Plot against 30-year mortgage rate changes with 6-month offset for affordability lead.
• Read FHFA HPI for metros outside the Case-Shiller 20 when allocating regional exposure.
• Monitor shelter CPI and OER monthly; note lag vs Case-Shiller peak.
• Cross-check housing starts and building permits for supply response to price momentum.
• Bookmark S&P press release PDF; verify revision footnotes before sizing trades.

Key takeaways

• Repeat sales make Case-Shiller the cleanest measure of U.S. house price appreciation, not a snapshot median.
• National, 20-city, and metro indexes reveal dispersion coastal traders and regional banks need.
• Lag is structural — use Case-Shiller for trend confirmation, NAR and rates for coincident reads.
• Shelter CPI follows with long delay; decelerating Case-Shiller is necessary but not sufficient for near-term CPI relief.
• Pair price indexes with volume (existing-home sales) and supply (starts, months’ supply) for a complete housing cycle picture.

Related reading

• Existing home sales explained — NAR closings, median price, and months’ supply
• Housing starts explained — permits, single-family vs multifamily, and GDP link
• Mortgages explained — rates, payments, and the affordability math buyers feel first
• Consumer Price Index (CPI) explained — shelter weight, OER, and inflation measurement