New York City is known for having a large stock of rent-controlled and rent-stabilized apartments. However, it turns out there may be a new kind of rent control in the city. According to a price tier analysis of the StreetEasy Rent Indices, the very top of the rental market in Manhattan has seen the slowest rate of rent increases over the past six years at just 18.3 percent. While still high, it is much less than the 27 percent increase seen for the overall rental market in Manhattan. In Brooklyn, however, the median rent in the most expensive group increased by 37.4 percent over the past six years, which is faster than the 26.6 percent increase seen for the overall Brooklyn market. In terms of price, as of January 2016, the most expensive segment of the rental market in Manhattan (the top 20 percent) had a median rent of $6,308 per month, while the same tier in Brooklyn had a median rent of $4,239.

The least expensive rental properties in both boroughs have seen much faster rates of rent increases. Between January 2010 and January 2016, the median rent in the least expensive group of listings increased by 39.3 percent in Manhattan and by 44.4 percent in Brooklyn. In 2010, the median monthly rent for Manhattan was $1,504 and $1,248 for Brooklyn. By January 2016, the median rent for those same properties had increased to $2,095 and $1,802 per month, respectively. The median rent in both boroughs in January 2010 and in January 2016, along with the percent change, can be seen in the tables below:

[tableau server=”public.tableau.com” workbook=”rentIndexQuintiles” view=”quintileWindow” tabs=”no” toolbar=”no” revert=”” refresh=”yes” linktarget=”” width=”600px” height=”1170px”][/tableau]

Why the Mixed Directions?

The differing rates of rent increases are due to many factors, but as we have seen the surplus of luxury sales properties may also be affecting this segment’s rental market counterpart. A property that is languishing on the market at the high end could just stay on the rental market until the right buyer comes along.

Why would anyone even consider spending upwards of $6,308 a month in Manhattan, or $4,239 in Brooklyn? Oftentimes buying a property in certain neighborhoods is so expensive, it makes more financial sense to rent. According to StreetEasy’s Tipping Point analysis, the median asking rent in Soho was $6,500 in 2015 and it would take over 30 years for the costs of owning a home there to exceed the costs of renting a comparable one. Similarly, in Dumbo, the median asking rent was $4,225 in 2015 and it would take a median time of just over 13 years for it to make financial sense to buy there.

How We Did It

In this analysis, the StreetEasy Rent Indices are divided into five quintiles, also referred to as price tiers. The Rent Indices are a repeat rental regression model, which means that any property that has been rented at least twice can be included in the index. By including only valid and verified listings from StreetEasy and employing a repeat rentals approach, the Indices emphasize the changes in rent on individual properties and not between different sets of properties.

Splitting the indices up into quintiles by price required assigning each property rented at least twice to a quintile group. To determine the price cut-offs, we looked at all properties with a rented status using StreetEasy data for each year and then split the transaction distribution into quintiles. If 20 percent of the transactions of a given year were below a particular rental price, that price would be the cut-off for the bottom quintile for that year. If 80 percent of the transactions of that same year were below a particular rental price, that price would be the cut-off for the top quintile. Returning to the set of properties with at least two declarations of being rented, using the date of the first rented date and the price cut-offs that corresponded to the year of that first rental, the property was assigned to a quintile according to which price group it fell into. That property would then be included in the index according to its assigned quintile.

If a property had been declared rented more than twice, we allow subsequent declarations of being rented to be reassigned to other quintiles, if necessary. This ensures that changes in the property and its area are reflected in the appropriate index.