At CTData we examine the strengths and weaknesses of different data sources to help users make data-informed decisions. Most recently, one of our users asked us about the source and collection methods for the Census measure, "Median Home Value by Town." We discovered that for this measure, homeowners are asked to estimate the value of their home if it were up for sale.
You can imagine our surprise at the possibility that the home values for each town may be higher than actual housing values given the known norm that people overvalue their homes. We wanted to compare the Median Home Value to other publicly available data source on home value and believed that Zillow provides just such a resource!
We don't want to keep our findings to ourselves. The data portal below is the product our work to answer a simple user question, and it represents the beginning of a housing data resource for Connecticut. We have aggregated data from both public and private data sources. What you will find:
The average median home price peaked in most counties in 2006 and then steadily declined, reaching a low in 2012. Since the low in 2012/2013, prices have remained relatively flat with only a slight increase.
The graph below aggregates monthly home sales data for all towns and home types by county.
The Zillow Home Value Index (ZHVI) is a smoothed, seasonally adjusted measure of the median estimated home value for towns by type of housing. The ZHVI for single family homes has followed a similar trend in all towns but the slope of the graph for each town has varied.1,2
The graph shows the rise of housing values, peaking in 2006 and 2007 for most towns, and then sharply declining following the 2008 financial crisis. Darien is the only town where the average median 2017 ZHVI surpassed its peak value.
In four towns, the average ZHVI in 2017 in only 70% of the 2006 peak value reached – Ansonia, New London, Waterbury, and Naugatuck. Hartford is excluded from this comparison, as Zillow did not report its home values before 2010. (Click here to see a comparison of 2006 to 2017 ZHVI).
As shown below, towns in Fairfield County (in light blue) have the highest home values, with Litchfield County towns the next highest. In 2017, the average median home values ranged from $101K in Hartford to $1.65M in Greenwich — the range in values doubled when compared to the range in values that existed in 1996.
In the aggregate, rental prices have been on the rise for both single family homes and multifamily homes, with the slight decline in 2017.
However, when looking across the state the price to rent a home can vary dramatically. Zillow calculates a Zillow Rent Index which is a smoothed, seasonally adjusted measure of the median estimated market rate rent. Similar to home values, Fairfield County towns have the highest ZRI values.3
The Rent to Price Index represents the number of years you would have to live in a home where the accumulated net cost of renting exceeds the cost to buy (meaning the number of years you must live in a place such that it makes more sense to buy than pay rent).
The three maps show that over time certain areas of the state have maintained a high break-even point. However, in Eastern Connecticut the ratio has declined such that it takes fewer years to break even.
In the U.S. Census American Community Survey 5-year estimates, data are collected on the value of Owner-Occupied Housing. Value is the respondent’s estimate of how much the property would sell for if it were for sale. These data are then aggregated and a median home value is calculated. Given that people are known to overestimate the value of their homes, we were interested in comparing these data with another source to determine the accuracy of the data.
The scatterplot shows that most Census data estimates are higher than the Zillow Average of the Median Values (dots below the diagonal line). Six towns had Zillow estimated values higher than the census estimates (Darien, Greenwich, Winchester, Washington, Stonington, and Windham).
Two towns were outliers in terms of the percent difference between the two estimates - Hartford and Canaan. In Hartford the median value provided by respondents was 50% higher than the average median values from Zillow.
According to federal guidelines, the household whose rent exceeds 30% of the gross income is considered rent-burdened. In Connecticut, there are 77 towns where average rent-to-income ratio exceeds 30%.
To try and ascertain affordability, we looked at the most recent American Community Survey 5-year estimates, 2012-2016. In particular, we mapped the dataset, Median gross rent as a percent of income. Since these data are survey data, they are not perfect and for some towns, the margins of error are large. Further, the median may not be truly representative of income estimates for households that are most likely to rent. However, it is the best data we currently have to identify the cost of housing. 77 towns have a price to rent ratio that is greater than 30%.
In Woodstock, Somers, Haddam, Middlefield and Mansfield the rent burden exceeds 45%.
To do so, we used Zillow home value indices (ZHVI) for all homes, which include single family residences, condos, and co-ops. Then we applied the 2015-16 mill rate for each town to 70% of the Zillow estimates as shown in the formula:
((ZHVI * 0.7) / 1000 ) * MILL
9 out of the 10 towns with the highest residential tax burden are in Fairfield County. The one town not in Fairfield County is Woodbridge. In Hartford County, Glastonbury, West Hartford, and Simsbury have the highest tax burdens.
Below is a tool that allows you to access all the data we used in this project by town. Select a town from the dropdown menu.