In our previous post, we attempted to establish several things. Allow us to refresh your memories: First, housing markets of late have been working in extremes. Four years ago, we were discussing record lows in for-sale markets and these days, we are discussing record highs in certain rental markets. Second, averages (or arithmetic means) can be misleading when used as indicators of pricing levels. This is the result of potential biases that can come from outlier units like super-luxury condos or dirt-cheap rooms for rent in a dataset that can serve to bring averages either up or down. The main takeaway was that medians are perhaps better indicators of price levels, as they represent the true middle number in a set.
In the process of playing with the data to conduct last week’s mean versus median test, another key question was begged: Is this intriguing gap between the mean and median number consistent throughout the entire borough of Manhattan, or will the sub-groups within Manhattan behave differently from each other? As we know, Manhattan is a collection of several neighborhoods. As a matter of fact, given the density in the borough, it could be argued that each neighborhood is almost like its own mini-city (think of the sheer number of people living in a neighborhood). Given this, it makes sense that these mini-cities of sorts may have their own rental market dynamics that are distinct from other mini-cities within the borough…or even the borough as a whole.
Before running the same test on certain neighborhoods as we did for the entire borough, we must first solve the eternal riddle of what a neighborhood actually is. Our tenure in this business has taught us that neighborhood boundaries are nothing more than broker brainchildren perched neatly somewhere between art and science. This is contrary to our seemingly naive beliefs from when we were laymen. Because It is difficult to find the consensus as to where these boundaries are, we have found that ZIP codes are the surest indicator of them. While ZIPs don’t perfectly rhyme with broker sentiment, they come close and that is all that matters.
To conduct our intra-Manhattan neighborhood test, we plucked out four ZIPs that represent popular residential neighborhoods in Manhattan. We chose 10024 (Upper West Side), 10003 (Union Square/East Village), 10036 (Midtown West), and 10021 (Upper East Side). Let’s see how things turned out:
With the averages in “red” and the medians in “blue”, one can quickly note that in each of these neighborhoods, the averages are far north of the medians. If you remember, this is similar to what occurred when we tested Manhattan as a whole, which should come as no surprise. As the saying goes, however, the devil is in the details. With a little bit more careful study, some interesting things can be taken away from these graphs. Allow us to point a couple of things out:
First, in some neighborhoods, the averages and the medians did not move in lock-step. An example of this behavior is evident in the 10024 ZIP (Upper West Side). Here, the average price dropped dramatically from February to March while the median price remained flat. A likely cause of this disparity could have been a drop in the number of luxury units on the market during this time period.
Second, we can glean from these graphs that rents aren’t increasing in all corners of Manhattan. Let’s look at 10038 (Midtown West) where prices have been falling since February. The same phenomenon is occurring in 10021 (Upper East) to a lesser extent except for a small bump around the first of April. These two ZIPs contrast the other two where means and medians follow each other likely as a result of a largely homogenous housing stock.
Stepping back from these past two posts and thinking critically, one can conclude that much of the headline-grabbing commentary about rental markets lately is either misleading, exaggerated, or both. While it is indisputable that rental markets are incredibly heated at the moment, there is some dispute as to how heated they truly are. Further, behavior within certain neighborhoods of Manhattan is divergent from the behavior of Manhattan as a whole. For true residential rental insight, a superficial analysis of these markets will not suffice. Stay tuned for further posts, as we will continue to bring novel and thought-provoking insights on rental markets to you.