Real Estate website Trulia released a report yesterday claiming to show how much rising rents are squeezing out the poor and young by showing how many of them are moving out of the 10 major US metros. Anyone who follows this blog (Hi, Mom) will know that I agree with the basic premise that housing is too expensive and it is having a major impact on urban demographics, but I don't think Trulia actually proved how many people are moving or that they are moving because of higher rent. To be fair, I have found previous Trulia reports to be well constructed and useful and don't mean to dump on them too much. They have an enviable amount of data (and 'data scientists') at their disposal and I can appreciate wanting to merchandise that as much as possible, but I think the report actually does a disservice to the greater discussion of affordability.
I have two problems with the report. The first is the methodology used to calculate their thesis (it almost seems like they had a conclusion in search of a thesis) and the second is how, given the title they chose for the report, they didn't come close to addressing the full scope of the problem of affordability in these major cities.
Let's start with the data. The authors of the report claim that the young (millennials 18-34) and the poor (less than $30k annual income) are being priced out of the 10 major metros and they prove this by showing how many of them are moving away. They come up with the term "move-away rates relative to expectation" to show how these two cohorts are leaving these cities at higher rates than their relative population shares would suggest. They even explain their reasoning by saying that if 60% of people in a city have brown eyes, you would expect that 60% of people moving away would have brown eyes.
I'm not even sure this assumption fits the cliche "correlation is not causation" but it certainly doesn't make much sense when you look at it in detail. Why is there any reason to assume that move-away rates should correspond directly with a selected cohorts relative population share of a city? Why would you expect 60% of people moving away to be brown-eyed if they are 60% of the population? Maybe there is statistical evidence that would suggest this, but the report doesn't offer any to compare their assumptions to.
More likely, they drew data from a single time period (the 2014 ACS) that showed significant differences between the young and poor moving away compared to other age/income groups and thought that not only was this significant, but it strongly represented being 'priced-out." The "compared to what" that is so important to data analysis is therefore grossly lacking here.
It's not actually interesting to compare move-away rates between the young and old or the rich and poor in the context of their report - a report claiming to show the young and poor being priced out by rising housing costs. It would be much more insightful to compare their current numbers with historical numbers of the young and poor's moving trends to see if there really is a dramatic difference. If this were true, then we could start to consider why this is the case now. But they don't actually prove that this is the case.
Just to focus on the young in more detail as an example, as this recent US Census report shows, they already move much more often than other age groups as they go to school, find/change jobs, or start a family. There also doesn't appear to be any significant change over the last ten years in these trends, even accounting for the dip during the recession. In fact, going back a few decades, the mobility rates have continued to decrease across age and income. The US Census makes the further distinction between 18-24 year olds, who move more than other age groups and people in their late 20s, who move the most of any, but the Trulia report does not, which makes it harder to determine what they are trying to prove. Basically, Trulia might be correct that the young/poor are moving more than other age/income groups right now, but they don't make the case that they are moving more than their specific demographic data trends would suggest.
Even if they were able to prove that more young/poor are moving today than previous times, they don't show that rent prices are the reason. In these 10 US metros, the 'out-migration' of the young represented in the report (it appears that the report does not address net out-migration, which would presumably complicate their the argument further) could easily be explained by job changes, moves to the suburbs, or any other factor that may or may not be linked to housing costs in these metros. They don't make any attempt to link their data analysis to these explanations or any explanation (nor do they attempt to explain where they are moving to, which would be insightful). Again, this is not to say that housing costs are not causing some young people to leave these cities, but the report don't make this argument despite its title.
The second problem with the report is how it misrepresents the crisis of affordability in these 10 metros and incorrectly lumps the housing challenges facing the young and the poor together. This is partly a problem of their broad definitions of young and poor but more so it is a problem of relying on the idea that each group has a Tiebout 'vote with your feet' type-agency to move when it becomes too expensive.
There is no doubt that some younger people and poorer people are leaving these cities because they can't afford to stay in them. Again, I don't disagree with this premise generally. But the larger truth is that rising housing costs aren't leading most of these people to leave cities, instead the poor are concentrating in poorer neighborhoods and the young and poor are both absorbing the higher rent burdens while forgoing other purchases.
It is therefore much more accurate to stay most of the poor and some of the young are being squeezed-in rather than squeezed-out. The truth is, gentrification does exist, but how much it displaces people (the young, poor, or otherwise) is much harder to determine and there is simply not enough evidence to say that it's causing people to leave a city altogether.
Even in the context of the report, the data shows the huge difference between young people moving and poor people moving. Again, this makes sense. The young move for many reasons but can generally absorb the financial cost of moving (let alone the social cost) more than the poor can. The young also have vastly different employment opportunities than the poor have, which creates more mobility options. They are also less likely to have children or family ties to a specific area. It is simply not helpful to compare the two groups' moving trends to make a claim about prices squeezing people out.
Regardless of their differences, in many cases, accepting a greater rent burden is still the better financial decision than moving for young and poor alike. Therefore determining the percentage of these people moving to other neighborhoods in the same city or accepting a higher rent burden compared to those moving away would much more illustrative to public policy discussions. That data already exists, so incorporating it would have been relatively easy.
Trulia shouldn't be faulted for raising issues of affordability in major cities but, even though their reports fall into a broader marketing and engagement strategy rather than policy prescriptions, I am surprised about how poorly executed this was. This report missed an opportunity to leverage their resources to introduce insightful data to a much needed public discussion.