housing.macroread.com
Methodology

How we compute housing fear & greed.
Public data, two gauges, four countries.

Most sentiment trackers are black boxes. We publish every component, every weight, every public data source, and every honest limitation. The methodology is the product — if it isn't auditable, it isn't useful.

The two-gauge framework

Stock-market sentiment is roughly one-dimensional. Fear and greed track price momentum, breadth, options skew, junk-bond spreads, and the like — all signals that point at the same question: is the market priced for optimism or for pessimism right now?

Housing is different. A city can be wildly expensive and cooling (think US coastal markets in 2023). A different city can be affordable and surging (think Calgary 2024-25). The same market can be both unaffordable and stagnant after a rate-hike cycle. Forcing those into a single number throws away the distinction that actually matters for buyers and observers.

So we publish two scores, each 0-100, for every market:

Affordability (0-100)

How reachable housing is for a typical buyer right now, normalized against the market's own ten-year history. The components, weighted 50/30/20:

  • Price-to-income (50%): Median home price divided by median household income. Compared to the same market's ten-year range. The most-cited affordability metric globally.
  • Mortgage payment burden (30%): Monthly amortized payment on the median home at current rates, as a percentage of median household income. Captures the rate-hike effect that price-to-income misses.
  • Price-to-rent (20%): Median home price divided by annualized rent on a comparable property. Cross-checks via the rental equivalent.

A score of 80 means the market is more affordable than 80% of the last decade for that same market. A score of 20 means the opposite. It is not a cross-market comparison — Toronto and Austin each get scored against their own histories, not against each other. That distinction matters and we'll re-emphasize it below.

Momentum (0-100)

How hot the market is right now. Components weighted 35/25/20/20:

  • Local tightness indicator (35%): Active listings divided by trailing-three-month sales pace. Tight inventory drives momentum.
  • Real (inflation-adjusted) 3-month annualized price change (25%): Official price index, deflated by CPI, annualized from the most recent three months — captures recent acceleration or deceleration.
  • Sale-to-list ratio (20%): Average final sale price as a percentage of original asking price — captures whether buyers are paying over or under asking.
  • Days on market (20%): Median days from listing to under-contract. Lower values indicate hotter momentum, so this component is inverted before being combined.

Same 0-100 scale, normalized against the market's own ten-year history.

How they combine

We don't combine them. The whole point is that they often disagree and that disagreement is the analytical signal.

Classification uses absolute economic thresholds on the raw inputs, not on the percentile-normalized scores. A market is “unaffordable” when the mortgage payment burden exceeds 35% of median household income, and “affordable” below 28%. Momentum is “hot” above +2% real annualized price growth and “cool” below −2%. These cutoffs come from mainstream mortgage industry practice (28%/35% are the classic housing-cost-burden lines) and a conventional definition of “flat” real growth.

Five labels result:

  • Low affordability + High momentum — bubble watch. The market is unaffordable and still running hot.
  • Low affordability + Low momentum — expensive stagnation. Often where high-cost markets sit after a rate-hike cycle.
  • High affordability + High momentum — recovery underway. Cheap and heating up.
  • High affordability + Low momentum — buyer's market. Cheap and cooling.
  • Transitional — at least one dimension is in the middle zone (payment burden between 28% and 35%, or real annualized growth between −2% and +2%). Common for markets near a turning point. We label these honestly rather than forcing them into one of the four corner regimes.

Where each city sits today is on its city page.

Per-country differences

Each country uses its own headline price index, mortgage benchmark, and listings data, because there is no single global housing dataset and the locally-canonical sources are usually the best ones.

United States

  • Price index: Case-Shiller (FRED CSUSHPISA national + metro components NYXRSA, SFXRSA, MIXRSA) and FRED ATNHPIUS series for Austin/Phoenix
  • Median price for affordability: Zillow ZHVI smoothed per metro, FRED MSPUS for national
  • Mortgage rate: Freddie Mac PMMS 30-year fixed (FRED MORTGAGE30US)
  • Inventory and days-on-market: Zillow Days-to-Pending per metro (Redfin Data Center planned, currently unavailable)
  • Rent index for price-to-rent component: Zillow Observed Rent Index (ZORI) per metro
  • Median income: US Census ACS metro-level estimates (2023 release)

Canada

  • Price index: Statistics Canada New Housing Price Index (Table 18-10-0205) for Canada national, Toronto, Vancouver, Calgary, Montréal
  • Mortgage rate: Bank of Canada 5-year posted via the Valet API
  • Inflation reference: StatsCan CPI All-items (Table 18-10-0004)
  • Median income: Statistics Canada 2021 Census metropolitan area median household income
  • Median price for affordability: CREA monthly metropolitan release (refreshed periodically in code)

Australia

  • Price index: Australian Bureau of Statistics Residential Property Price Index (RPPI), quarterly, by Greater Capital City Statistical Area
  • Mortgage rate: Reserve Bank of Australia cash rate (table F1.1) plus a typical variable-rate spread of two percentage points
  • Inflation reference: ABS All-groups CPI (monthly, dataflow CPI_M)
  • Median income: ABS 2021 Census metropolitan median household income
  • National series: population-weighted average of the five capital cities tracked here (ABS publishes capital-city RPPI but not a separate national row)

United Kingdom

  • Price index and average price: HM Land Registry UK HPI (averagePrice and housePriceIndex), monthly, by region (London, Manchester / North West England, Birmingham / West Midlands, Edinburgh / City of Edinburgh, Bristol / City of Bristol)
  • Mortgage rate: Bank of England base rate plus a typical 5-year fixed spread of one and a half percentage points
  • Inflation reference: ONS CPIH (Consumer Price Index including Owner Occupiers' Housing costs)
  • Median income: ONS Annual Survey of Hours and Earnings (ASHE) 2024 regional median, scaled to a household approximation

What we deliberately don't do

  • We don't predict prices. No "buy now" or "sell now" framing anywhere on the site. We publish where the market sits today and let you decide what that means.
  • We don't make cross-city comparisons claim more precision than they have. Each city's gauge is normalized against its own ten-year history. A Toronto affordability of 18 means Toronto is at the harder-than-history 18th-percentile for itself, not that Toronto is harder to buy in than Sydney at 22.
  • We don't include commercial real estate. Different drivers, different audience, different data sources. Out of scope.

Data licensing and reuse

Our scores are published under Creative Commons Attribution 4.0 International (CC BY 4.0). You can embed, quote, screenshot, or republish them in any context (commercial or not) provided you credit the source.

Underlying public data is attributed to its original sources on each market page and consolidated on /data-sources.

Frequently asked
Why two gauges instead of one?
Housing markets are two-dimensional in a way that stock-market sentiment isn't. A city can be both unaffordable and cooling (a bubble that is starting to break). Another can be both affordable and surging (a recovery underway). Forcing those into a single number erases the most useful distinction. Affordability tells you whether prices are reachable. Momentum tells you whether they are moving. Together they give you a regime read.
How often is the data updated?
Three cadences. Daily: listed homebuilder and REIT stock prices, our only truly real-time signal. Weekly: mortgage rate releases (Freddie Mac PMMS, BoE composite, BoC posted, RBA series), MBA application volume, Australian auction clearance rates, Rightmove asking-price index. Monthly: official price indices (Case-Shiller, CREA HPI, CoreLogic HVI, Nationwide HPI), NAR existing home sales, NAHB builder sentiment.
Why not include rental yield, vacancy, or other data?
V1 holds the methodology to four affordability components and four momentum components per market so the calculation stays auditable. Rental yield, vacancy rates, mortgage origination volume, foreign buyer share, and household formation are all real signals we may layer in for later versions. Adding them now would be premature without a v1 baseline to compare against.
Why does each country use a different price index?
Because there isn't one global index. Case-Shiller is the gold standard in the US. Canada uses CREA's national HPI with a clear methodology. Australia's CoreLogic Daily HVI is the closest thing to a real-time index anywhere. The UK has four competing indices (Nationwide, Halifax, ONS, Land Registry) and we lead with Nationwide because it releases on the last working day of the month with the shortest lag and is built on actual mortgage approvals. We cross-check against the others where possible.
What about commercial real estate, REITs, or industrial property?
Out of scope for housing.macroread.com. The data sources, drivers, and audience for commercial real estate are different enough that a separate site would serve them better. Listed REITs appear here only as one momentum component — they're a leading indicator for residential, not the subject.
What are the limitations of this data?
Honest list: monthly indices lag by 2-8 weeks. Different country indices use different methodologies (repeat-sales, hedonic, valuation-based) so cross-country comparison is directional, not exact. Median prices smooth over mix-shift (a market with more luxury sales will look hotter even if individual property values are flat). City-level affordability uses metro-level median income, which understates affordability in the central city itself. We document these on each page where they matter.
Is the data free to use?
Yes, under Creative Commons Attribution 4.0 International (CC BY 4.0). You can embed, quote, screenshot, or republish the gauges in any commercial or non-commercial context provided you credit the source as "housing.macroread.com." The underlying public data is from sources we attribute on each page.