Making Neighborhoods: New York Region

In 2014, CHPC released the first Making Neighborhoods report and interactive map studying demographic change in New York City in the decade spanning 2000 to 2010. Since then, we have expanded our work to include the metropolitan area surrounding the city.

At the heart of Making Neighborhoods is a statistical method known as cluster analysis. Our cluster analysis method finds commonalities between census tracts across race, income, age, educational attainment, foreign birth, household and family type, and presence of public housing. Our model identified 16 population clusters of census tracts that differed from each other for the year 2000.

If there was one overarching takeaway from our original 2014 Making Neighborhoods study of the five boroughs, it was that the formal, government-drawn boundaries are inadequate for measuring changes in the housing marketplace. The analysis of New York Citys five boroughs showed us many instances where significant demographic groups straddled boundary lines, effectively
splitting them and neutralizing their importance to their official neighborhoods.

There were also interesting changes happening along the Queens-Nassau County border and along the Bronx-Westchester County border. Those transitions compelled us to answer the question of what was happening just beyond city limits.

Using the Making Neighborhoods methodology, we are able to see where clusters of populations shrink or grow. From the thousands of small changes we observed, several distinct trends emerged. By applying the Making Neighborhoods method to the entire regional housing market, five trends mirrored what we saw when we focused only on the city:

• The region lacks a consolidated middle-class Hispanic population cluster. Overall, the regions Hispanic population was the second-fastest-growing, at 19 percentfaster than either of our two population clusters with a Hispanic majority. The clusters that emerged with a Hispanic majority were at low and low-middle income levels, whereas the regions middle-class Hispanic households are found in clusters where the majority race might be black or white.
• In some parts of the region, tracts that in 2000 were home to the upper-middle-income majority black population cluster gave way to a majority black cluster at a lower income level, such as in southeast Queens, the northern Bronx, Hempstead, and Newark.
• Neighborhoods that had a white majority in 2000 saw a further consolidation of white households by 2010, such as Dyker Heights, Maspeth, Jericho, and Norwalk.
• Some neighborhoods with majority white population clusters in 2000 transitioned to an Asian majority by 2010, such as in northern Queens, southern Brooklyn, Plainsboro, and Hicksville.
• Population clusters with no race majority were largely dispersed, giving way to clusters with a clear majorityeither Hispanic, Asian, or whitein 2010, such as in Kensington and Bay Ridge, Brooklyn, and Astoria and Woodhaven, Queens.

There were also new trends that emerged unique to this wider regional study:

• A low-income, majority white population cluster, which did not emerge from our study of the five boroughs, was the fastest growing cluster between 2000 and 2010, found in Brooklyn and pockets of the suburbs in Middlesex County, NJ, and Rockland County, NY.
• Some tracts that were home to a lower-income black population cluster transitioned to a higher-income black cluster, contradicting another trend revealed earlier, but mostly occurring in the same or adjacent areas.
• Some areas in urban neighborhoods or straddling urban and suburban areas transitioned from a white majority to a Hispanic majority. This happened in denser suburbs like Clifton, NJ, Bridgeport, CT, and in Ridgewood, Queens.

Our report details the methods and results of this innovative approach to studying demographic change. The Making Neighborhoods study measures change from 2000 to 2010, as well as presents our thoughts on the implications of the results for the regional housing market looking forward.