Skip to main content

How does the Case-Shiller home price index measure housing trends?

The Case-Shiller Home Price Index is the gold standard for tracking home prices over time. Created by economists Karl Case and Robert Shiller (and now maintained by S&P Dow Jones Indices), it measures the repeat-sale prices of single-family homes in the United States. By tracking the same homes over time, the Case-Shiller index avoids the "composition problem" that plagues other home price statistics: if expensive homes sell more often in a given month, the median price can jump without any home actually becoming more expensive.

For policymakers, investors, and homeowners, the Case-Shiller index is essential. The Federal Reserve watches it closely when setting interest rates; investors use it to spot real estate bubbles and busts; homeowners use it to understand whether their home is appreciating or depreciating relative to the broader market. The index comes in three flavors: the National Home Price Index (covers all 50 states), the 20-City Composite Index (major metropolitan areas), and 19 individual metro-area indices (down to the city level).

Quick definition: The Case-Shiller Home Price Index measures the price changes of single-family homes by tracking repeat sales (the same homes sold multiple times), eliminating composition bias and providing a reliable, long-term measure of housing market trends.

Key takeaways​

  • The Case-Shiller index uses "repeat sales" methodology: it tracks homes that sold twice (or more) and calculates price changes for the same property, eliminating composition shifts.
  • The index is rebased to 100 in January 2000, making it easy to calculate total appreciation or depreciation since that date.
  • Year-over-year price changes are more reliable than month-to-month; the index can swing ±2 to ±3 percent month-to-month due to seasonal and data-lags, but year-over-year smooths this noise.
  • The 20-City Composite is the most closely watched version; individual metro indices reveal regional variation (Western and Southern markets are more volatile than Midwest and Northeast).
  • The index lags real-time sales data by two to three months because it only includes homes with recent prior sales; newly-built homes aren't included until they've been resold.

How the repeat-sales methodology works​

Most home price indices use "hedonic" methods: they estimate a home's price based on attributes (square footage, bedrooms, bathrooms, age, location) and track how those attribute-prices change over time. The problem: if bigger homes sell more often in a given period, the average price can jump even if a typical home is no more expensive.

The Case-Shiller index uses a different approach. It tracks pairs of sales for the same home sold in different time periods. If a home sold for $200,000 in January 2010 and again for $280,000 in January 2015, the Case-Shiller methodology captures that 40 percent price change. By multiplying across thousands of repeat-sale pairs, the index measures the pure price appreciation (or depreciation) of the housing stock.

This approach has advantages: it's transparent (you can understand exactly what's being measured), it avoids composition bias, and it's based on actual transactions. But it has a lag: the index can only include homes that have been sold at least twice recently. Newly-built homes that haven't been resold yet aren't included. Homes in very illiquid or slow-turnover markets (some rural areas) are underrepresented.

The index is calculated monthly for metros and quarterly for the national index. Data are released with a two-to-three-month lag (December data are released in late February).

Why the index is rebased and how to interpret it​

The Case-Shiller index is rebased to 100 in January 2000. This makes interpretation intuitive: if the index is 200, homes are worth double what they were in January 2000; if it's 150, homes are worth 50 percent more. The national index stood at 119.7 in January 2000, 226.2 at the peak of the housing bubble in July 2006, fell to 150.4 at the trough in February 2012, and climbed back to 314.8 by March 2022 (the most recent peak).

To calculate total appreciation since January 2000, the formula is: [(Current Index − 100) / 100] × 100 percent. For example, if the index is 310, total appreciation is (210 / 100) × 100 = 210 percent. If someone bought a median-price home in January 2000 for $300,000, the same home would be worth roughly $900,000 in March 2022.

Year-over-year price change is calculated differently: [(Current Index / Index 12 months ago) − 1] × 100 percent. If the March 2022 index is 314.8 and March 2021 was 281.6, then year-over-year growth is (314.8 / 281.6 − 1) × 100 = 11.8 percent. This metric smooths seasonal noise and reveals the underlying trend.

Regional variation and the 20-city index​

The national Case-Shiller index is a population-weighted average of all metro areas. But regional variation is huge. Western metros (San Francisco, Los Angeles, Phoenix, Las Vegas, San Diego) have experienced the most volatility: they surged the most during the 2003–2006 bubble, crashed the hardest in 2008–2012, and rebounded the fastest in 2012–2022. Southern metros (Miami, Tampa, Atlanta, Charlotte) have shown similar but slightly less extreme patterns.

The Midwest (Chicago, Cleveland, Detroit) and Northeast (New York, Boston) have been more stable. They experienced less appreciation during the bubble (never got as overheated) and less depreciation during the crash (people still need housing in cold climates). As of 2022, Western metros had fully recovered and exceeded prior bubble peaks by 20–40 percent, while Midwest metros had recovered but not exceeded bubble peaks by much.

The 20-City Composite Index includes: Atlanta, Boston, Charlotte, Chicago, Cleveland, Dallas, Denver, Detroit, Las Vegas, Los Angeles, Miami, Minneapolis, New York, Phoenix, Portland, San Diego, San Francisco, Seattle, Tampa, and Washington, D.C. This subset captures the most liquid, most-reported-on markets. Individual metro indices are available for all 19 (the 20th is the national index).

For investors and homeowners, understanding regional variation is crucial. "Home prices are up 10 percent" tells you nothing if you live in Miami (where prices might be up 20 percent) versus Cleveland (where they might be up 3 percent). Always compare your metro to the national index.

The 2008 crash and the 2020–2022 boom​

The Case-Shiller index tells the story of two decades of American housing clearly. From January 2000 to July 2006, the national index rose 116 percent—an extraordinary bubble fueled by loose lending, speculation, and the belief that home prices never fall. Western metros saw even faster gains (200+ percent in Las Vegas, Phoenix, and San Diego).

Then came the crash. From July 2006 to February 2012, the national index fell 33 percent—a decline comparable to the Great Depression. Western metros crashed even harder: Las Vegas fell 65 percent, Phoenix fell 50 percent, Los Angeles fell 45 percent. Homeowners who bought in 2006 had lost half their home's value by 2009.

The recovery was slow but steady. From February 2012 to March 2022 (10 years), the national index climbed 109 percent. By mid-2022, homes were worth more than double the January 2000 baseline. The wealth effect was enormous: trillions of dollars of equity was created as home prices recovered, fueling consumer spending and GDP growth.

Then, in 2022–2023, as the Fed raised interest rates from 0 to 4.25 percent, the Case-Shiller index began to decline—falling about 5 percent from peak by end of 2023. This decline was much shallower than the 2008 crisis (33 percent) and was regional: Western metros fell 15–20 percent while Midwest metros fell only 2–5 percent. By early 2024, the decline had stabilized, and regional variation suggested that prices had found new equilibrium at lower but sustainable levels.

Bubble identification: When does price growth become dangerous?​

Economists and investors constantly debate whether housing prices reflect fundamentals (income growth, population, cost of building) or are in a bubble (prices have decoupled from reasonable valuation). The Case-Shiller index is the primary tool for answering this question.

One common approach: compare home prices to household income. In January 2000, the median home price was roughly 2.8 times median household income. By July 2006, this ratio had climbed to 4.5 times income—clearly unsustainable. By February 2012 (post-crash), it had fallen back to 2.9 times income (fair value). By March 2022 (pre-rate-hike), it had climbed back to 4.8 times income—again, stretched.

Another approach: compare price-to-rent ratios. If you can rent a home for $2,000 per month but the purchase price is $400,000, the annual gross rent is $24,000 and the price-to-rent is 16.7. Historically, price-to-rent ratios above 18 suggest prices are overextended; below 12 suggests undervaluation. By July 2006, price-to-rent ratios had climbed to 20+ in many metros—a clear warning. By February 2012, they'd fallen to 10–12 (undervalued). By 2022, they'd climbed back to 16–18 (stretched again).

Using these metrics, the Case-Shiller index revealed two clear bubbles: 2003–2006 and a smaller one 2017–2022. Both were preceded by Fed rate hikes that cooled demand and prices.

The lag in the Case-Shiller data​

A crucial limitation of the Case-Shiller index is its lag. Because the index only tracks repeat sales, it can only include homes that were recently resold. The data are published with a two-to-three-month lag, meaning February data are released in late April. During fast-moving market periods (like 2022, when rates surged), this lag can make the index seem less responsive than it really is.

For example, in 2022, real estate agents and investors could see prices cooling in real time (offers below asking price, homes sitting on market longer). But the Case-Shiller index, lagging by two to three months, showed prices still rising through October 2022. By November and December 2022, the index finally turned negative. This lag meant that many observers thought housing prices were holding up longer than they actually were.

For real-time price signals, traders and investors supplement the Case-Shiller index with faster data: median listing prices from Zillow or Redfin (updated daily or weekly), price-per-square-foot from MLS data (updated monthly), or the Pending Home Sales Index (released monthly, leading the repeat-sales index by one to two months).

What the Case-Shiller index tells the Federal Reserve​

The Fed watches the Case-Shiller index as a wealth barometer and inflation signal. Rising home prices boost household wealth and consumer confidence, spurring spending and inflation. Falling home prices reduce wealth and spending, deflationary effects.

The Fed's reaction function considers the Case-Shiller index in two ways:

As an inflation indicator. Home prices (and rents, which move with home prices) are part of the inflation indices the Fed targets (the PCE and CPI). Strong house-price gains often precede strong rent inflation 6–12 months later. The Fed watches price-to-income and price-to-rent ratios to identify overheating.

As a wealth indicator. The Fed knows that a 10 percent drop in home prices reduces household wealth by roughly $1 trillion (for 125 million homes). This wealth loss feeds through to consumer spending with a lag of 6–12 months. By lowering rates or buying bonds when home prices are falling sharply, the Fed tries to cushion the wealth loss.

In 2022, as the Fed raised rates to fight inflation, it accepted that home prices would fall. The Fed wanted demand to cool, prices to moderate, and new supply to catch up to demand. By early 2023, Case-Shiller data showing modest price declines (5–6 percent from peak) suggested the Fed's policy was working: prices were moderating, supply was normalizing, and the market was rebalancing.

Reading the monthly Case-Shiller report​

S&P Dow Jones Indices releases the Case-Shiller data monthly, typically near the end of the month. The release includes:

  • National index (month-over-month and year-over-year changes). Month-to-month changes of ±0.3 to ±0.5 percent are normal; beyond ±1.0 percent warrant attention.
  • 20-City Composite (same metrics). Often moves differently from the national index if certain metros are particularly strong or weak.
  • Individual metro indices. All 19 metros allow regional analysis.

For interpretation, focus on year-over-year changes (smooths seasonal noise) and compare to income growth (typically 2–3 percent annually). If year-over-year home price growth exceeds 5–7 percent for two consecutive years, prices are moving above fundamentals and bubble risk is rising. If year-over-year growth is negative for two consecutive quarters, a price correction is underway.

Real-world examples of Case-Shiller inflection points​

In 2005–2006, the Case-Shiller index rose 15 percent year-over-year—unsustainable by any fundamental measure. Economists warned of a housing bubble. Policymakers dismissed warnings. By mid-2006, the index growth rate began to slow; by 2007, it turned negative. Most observers didn't realize the crash had begun until 2008–2009 when losses were massive.

Fast-forward to 2020–2021. As the Fed cut rates to zero and Congress passed stimulus, home prices surged. The Case-Shiller index rose 20–25 percent year-over-year, the fastest growth in 30 years. Warnings of a bubble emerged again. But this time, the Fed moved preemptively: it raised rates starting March 2022. The Case-Shiller index peaked in March 2022, began declining in late 2022, and by end of 2023 was down about 6 percent from peak. The decline was shallow compared to 2008 because the Fed acted sooner.

This comparison illustrates how the Case-Shiller index is used: identifying unsustainable price growth (bubble detection) and gauging whether policy responses are working (is the correction orderly or chaotic?).

Common mistakes when reading Case-Shiller data​

Ignoring the two-to-three-month lag. When you read Case-Shiller data on March 31st, you're reading January data. In fast-moving markets, real-time price declines can be much sharper than the Case-Shiller index suggests.

Comparing absolute index levels across time without adjusting for income. The index at 300 (March 2022) doesn't directly compare to 150 (February 2012) without knowing whether incomes also doubled. Always calculate price-to-income ratios to assess affordability.

Assuming national trends apply locally. The national index rising 3 percent year-over-year tells you nothing about your metro. Your market might be up 10 percent or down 5 percent. Always check individual metro indices.

Over-interpreting month-to-month changes. A single month of -1.5 percent is not a crash; it's normal volatility. Look at three-month or six-month trends to identify true turning points.

Forgetting that the index excludes new construction. The Case-Shiller index only tracks repeat sales, so newly-built homes aren't included until they're resold (typically 5–10 years later). This means the index can miss rapid appreciation in new construction or rapid depreciation in older stock.

FAQ​

Why do the Case-Shiller national index and 20-City Composite sometimes move differently?​

The national index is population-weighted, so large metros (New York, Los Angeles, Chicago) dominate. The 20-City Composite includes 20 specific metros and weights them equally. If small markets boom while large markets slump, the two indices move differently. The 20-City Composite often reflects bubble/bust dynamics more dramatically because it includes high-volatility metros like Las Vegas and Phoenix.

What is the "median home price" reported by NAR, and how does it differ from Case-Shiller?​

The NAR (National Association of Realtors) reports median home prices based on actual transactions in a given month. It's faster (released monthly, same month as sales) but susceptible to composition bias. If expensive homes sell more often in a given month, median price jumps. Case-Shiller, by contrast, tracks repeat sales of identical homes, eliminating composition bias, but lags by 2–3 months. Case-Shiller is more reliable for trends; NAR is better for real-time signals.

Can the Case-Shiller index be used to predict future price movements?​

Not really. The Case-Shiller index measures past price changes; it doesn't forecast future ones. However, it does reveal whether prices are stretched (using price-to-income or price-to-rent ratios), which can suggest a correction is coming. But timing corrections is nearly impossible; the only reliable signal is interest rates and lender conditions.

How does the Case-Shiller index account for renovations or home improvements?​

The repeat-sales methodology assumes the home is the same quality each time it sells. If a home was renovated between sales, the price change includes both the market appreciation and the renovation value. This is a limitation; the Case-Shiller index can overstate true market appreciation if renovations are common. Hedonic indices that adjust for quality changes don't have this problem, but they have composition bias instead.

How quickly do the Case-Shiller metro indices recover after recessions?​

The 2008 housing crash took about six years to recover (February 2012 to February 2018 to regain July 2006 levels). Western metros were faster (by 2017); Midwest metros slower (by 2018–2020). This recovery period is important because it shows how slowly housing bubbles deflate. Even with a massive 33 percent crash and years of zero growth, it took nearly a decade to climb back to fair value.

Summary​

The Case-Shiller Home Price Index measures the change in prices of single-family homes by tracking repeat sales of the same property over time, eliminating composition bias that affects other home price metrics. The index is rebased to 100 in January 2000, making it intuitive: a value of 300 means homes are worth three times what they were in 2000. The index lags real-time data by two to three months (published with 2–3 month delay) but provides the most reliable measure of long-term housing trends. Three versions exist: the national index, the 20-City Composite (major metros, equally weighted), and 19 individual metro indices. Regional variation is substantial: Western metros have been most volatile (greatest boom and bust swings), while Midwest metros are more stable. Year-over-year price growth exceeding 5–7 percent for sustained periods signals bubble risk and often precedes crashes. Price-to-income and price-to-rent ratios reveal when prices are overextended relative to fundamentals. The Federal Reserve watches the Case-Shiller index as a wealth indicator and inflation signal; rising prices boost consumer spending; falling prices reduce it.

Next​

→ Consumer confidence index