Insights—Price Correlations with Other Factors

Here is a list of all the insights pages on the site.

This exploration will get a little more technical, with a little more statistical jargon. But we'll try to resurface periodically with practical insights.

The goal here is to explore the correlation between price and other factors. Does warmer weather imply higher cost of housing? One would assume so. What about property taxes? They might make investment less attractive and thereby lower the price of housing. Does more walkability increase prices? What about climate risk? Is higher risk reflected in the price, or have markets not yet figured out that these areas are in danger?

First we'll introduce a list of correlations, then we'll delve a bit more into each. As a reminder for those of you who weren't paying attention in high school stats, correlation is a measure of the extent of the relationship between two quantitative variables. The number ranges from -100% to 100%, with -100% and 100% being 100% negative and positive correlations, respectively. For instance, the number series 1, 2, 3, 4, 5... and 2, 4, 6, 8, 10, 12... have an 100% positive correlation. However, this kind of correlation is rarely found in the wild. In practice, correlations over 50% are usually considered strong, while the 20-50% range is moderate. Correlations under 20% are not usually viewed as significant.

Correlation between home prices and other factors in our database:

Some of these are unsurprising. For instance, cities with lower crime have higher property values. But there are some that are a bit more interesting.

The first reasonable expectation shattered is that warmer weather usually means higher prices. Think of the Sun Belt and you think of glittering gated communities, with elegant golf courses and polo shirts worn year round. And yet it is actually moderate weather that costs the most—California's year-round balminess springs to mind. Warmer winter weather indeed commands a premium, but warmer summer weather is correlated with an equally steep discount. To put this in more quantitative terms, cities where the mean July high exceeds 85 degrees Fahrenheit have a median price of $400,000. Cities with cooler summers average $452,000—a huge difference! Meanwhile, winter weather is even more striking: cities where the average January low is below 35 only have a median price of $323,000, while cities above that threshold average $509,000—a nearly 60% jump!

Although climate change correlations with price were muted on the surface, the results under the covers were a bit more interesting. Higher risks of most factors, as you might expect, did correlate with lower prices. However, fires and sea level rise, perhaps associated with the scenic West and beachy Florida respectively, actually were correlated with significantly higher prices, over 35% in each case.

Politics also makes a difference. Probably in keeping with the expense of blue New York and California, cities without a majority of Republican votes in the 2020 election had median home prices of $465,000, while the redder cities only averaged $329,000, a dazzling 30% discount. Excluding New York and California, the Democratic median dropped to $385,000, while the Republican median only fell to $316,000, reducing the "political discount" to around 18%.

Interestingly, state income tax has a positive correlation with home prices, but property taxes have a negative correlation. However, it's hardly worth buying a less expensive house just to be chained to a higher property tax bill, so this insight may not be that useful for most readers.

In terms of walkability and school ratings, they are numerically cleaner, so we can build linear regressions to see how much you pay for each additional point. If you don't remember what linear regressions do, don't worry—we'll simplify it for you.

The walkability findings are shocking: for every 1% increase in walkability, you pay an additional $4,900. The 20 cities with the top walkability scores have four with a median price over $1 million and six more with a median price over $750,000!

Meanwhile, for every one-point increase in our one-to-ten school ranking mechanism, you pay an additional $71,600! Raising kids right really is expensive. To further drive the point home, take a look at home prices by school ranking:

Of course, correlation does not imply causation, and there are other major factors that might be influencing both the price and the school rankings, such as income, crime rate, infrastructure, etc. That being said, this is still a striking finding for those looking to purchase a home.

Hopefully this deep dive helped you figure out where to move. If you don't have kids or you send them to private school or homeschool, you can save money by avoiding the really reputable school districts. If you're willing to tolerate really cold or really hot weather, that could save you $100,000! To get personalized recommendations on where to move based on all these factors and more, take the quiz.