Property Values in Major US Cities since 1984- Part 2

Following up to my previous post about property values in US cities, I wanted to correlate annual house values with different variables such as population, size of economy, crime, education, and income. I’m very interested to see which of the listed variables home values correlate with the most.

Let’s begin.

The first chart shows total population of each city in 2014 correlated with house values for the same year (Q2).

Figure 1 – Total Population vs. Home Prices, 2014Pop

The largest cities in the US do not generally command higher house prices, as seen in Figure 1. However, there is a stronger correlation for example when population and house price growth is analyzed. The following chart shows house values correlated with the Gross Domestic Product in 2014 of each city.

Figure 2 – Gross Domestic Product vs. Home Prices, 2014GDP

The most interesting thing I notice after looking at both Figure 1 and 2 is that the two charts are almost identical. In fact, when you correlate total population and GDP,  you get an almost perfect correlation. Despite this, the strength of a city’s economy shows a stronger correlation with home prices than population. The next chart shows all crime per 100,000 residents correlated with house prices.

Figure 3 – All Crime per 100,000 residents vs. Home Prices, 2014Crime

Overall crime does not correlate too strongly with house prices. But when I broke up the different categories of crime, burglary showed the strongest correlation at R^=0.27 (not too surprising). Interesting to note the outliers with high house prices and high levels of crime such as San Francisco, Oakland, Wahsington DC and Seattle.

Figure 4 – Education Attainment vs. Home Prices, 2014 

Education attainment clearly shows the strongest correlation with house prices thus far. Interesting outliers include Washington DC, Boston, Denver and San Bernardino. Washington DC and Boston are very well recognized for the nationally significant institutions that are based there (i.e., the federal government in DC, and prominent education institutions like Harvard, MIT, Boston University and Northeastern in Boston), which result in high education levels of their residents. Interestingly enough, the high education levels in both Washington DC and Boston do not translate into the highest house prices. It almost seems like nice weather has an important influence on house prices. San Bernardino in contrast surprised me a little as I did not know too much about the city. The city has a large portions of its residents living below the poverty line, which might influence its low levels of education attainment.

Figure 5 – Annual Median Income vs. Home Prices, 2014

The second strongest correlation with home prices is income levels, as shown above. This shouldn’t come as a surprise to anyone, as wage levels will dictate whether a home is affordable to a prospective buyer or not. As seen in Figure 4 and 5, the same cities generally exhibit high education and income levels – notably San Jose (where Silicon Valley is), San Francisco, Washington DC and Boston. In fact, when you correlate median annual wages with % population over 25 with a bachelor’s degree or higher, the variables show a very strong correlation to one another.

It would also be interesting to correlate the size of the city in area to its house prices, as recent analyses have determined that geographical constraints influence housing supply. As well, I would be curious to correlate the house price data with average temperatures for the cities – as it’s evident that the housing markets in cities with nicer climates have rebounded much faster since the most recent recession than cities with colder climates.

Data Sources:

Population data –

GDP data –

Crime data –

Education attainment data –

Income/Wages data –




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s