The strange saga of Massachusetts, Boston and the United States Census Bureau continues
The Boston Globe’s Stephanie Ebbert has an interesting if incomplete article in today’s Boston Globe about the United States Census Bureau’s estimates for the populations of Massachusetts and Boston.
The population estimates aren’t critically important since it is the decennial population counts (the census) that determines the distribution of congressional delegates, distribution of federal funds and the outcomes of other population based decisions. Still, one should be able to expect population estimates to be at least somewhat accurate, because if they aren’t, why bother with them?
The controversy in Massachusetts has primarily centered on the estimates for Boston, which showed a steep decline in the middle of this decade. The city’s mayor Tom Menino challenged the estimates and the Census Bureau revised its estimate in late 2006, without explaining particularly well why it did so, or how it could have been so wrong.
Estimating the city’s population year to year isn’t an easy task. Boston has a steady influx of immigrants, illegal aliens, foreign students, domestic students, and young professional coupled with a steady stream of recent graduates and couples ready to start a family. On the other hand, it’s probably no cake walk to keep up with the population influx in states like California, Arizona and Nevada, either, so the difficulty that the Bureau’s Boston office seems to have in coming up with reliable and defensible estimates strikes me as extra ordinary, or at least noteworthy.
The latest estimate is that Massachusetts’s population grew by 2.1% from 2000 to 2007, compared to 7.2% for the United States.
The gist of Ebbert’s article is that the state of Massachusetts and the City of Boston are spending considerable amounts of money on finding people of may have been overlooked by the Census Bureau’s population estimates. For example, researchers at the University of Massachusetts Donahue Institute believe the state has a group-quarter population of 234,556, “8 percent more than the census is expected to estimate for 2007,” according to Ebbert. Two-thirds of those allegedly unestimated souls live in Boston, presumably as college students. The group quarter population was 221,216 in 2000, according to the population count that year.
Five things stand out about Ebbert’s article:
1) Nobody from the Census Bureau is quoted in it, which is very strange. I don’t know if Ebbert simply didn’t bother to ask anybody at the Bureau or if the Bureau declined to comment.
2) The article pretty clearly violates the embargo that the Cenus places on its pre-release data. I’m guessing that technically Ebbert didn’t do it, instead she was fed the data by either Alvaro Lima, research director at Boston Redevelopment Authority, or by Holly St. Clair, “manager of the metro data center for the Metropolitan Area Planning Council.”
3) The article makes no mention of anybody anywhere demanding some accountability for the shortcomings of the Census Bureau.
4) The BRA seems to be directly contradicting previous claims by the city, by Lima specifically, regarding the inclusion of college dorms built in the city since 2000.
5) The BRA and Donahue Institute cleverly omits a fact that could wipe out a good bit of their argument that the state’s and the city’s populations exceed the Census Bureau’s estimates.
The first three points are what they are, but let’s look a little bit closer at the remaining two.
According to the article, BRA believes the city’s population was at least 616,535 last year, substantially more than the 595,698 the census estimated for 2006 and quite a bit more than what the Census estimate is likely to be (back in March this year I predicted the 2007 estimate will top 600,000). But here’s the interesting thing about the the BRA’s estimate: It doesn’t jibe very well with what the city claimed in 2006, after the first and biggest revision of the population estimate for Boston.
Writes Ebbert:
However, the revised estimate still didn’t take into account the new housing - including 8,400 dormitory rooms - that has been built in the city since the 2000 Census.
Here’s what the Globe reporters Michael Levenson and Yvonne Abraham wrote after the 2006 revision:
BRA officials found that the Census Bureau had underestimated new housing production by more than 3,000 units, and overestimated demolitions by 1,800, said Alvaro Lima, the agency’s director of research.
More by Levenson and Abraham:
Gregory Harper, a Census Bureau demographer, said the agency reviewed the city’s data and agreed that more new housing units such as dormitories should have been factored into the estimate.
A straight-forward question: Did the 2006 revision of the 2005 population take new construction of dormitories into account or not? Did it or did not?
It’s also worth noting that the revision in 2006 was based on an increase of approximately 4,900 units. Those units translated into an upwards revision of 37,000 people compared to the previous estimate. For some reason, the BRA is arguing that the 8,400 college dorm units should count for an upwards revision of less than 20,000, perhaps for as relatively little as 12,000 - 13,000 the underestimate of Boston’s group quarter population that the University of Massachusetts Donahue Institute argues the Census Bureau missed in Boston in 2007, according to Ebbert’s article. The Donahue Institute estimate includes all group quarter populations, not just college students.
Then there’s the matter of the date for the population estimates. The population estimates are specifically for July 1st, unlike the decennial population count which is for April 1. So, it seems reasonable to assume that many of the college students that the BRA and Donahue Institute are waving as proofs of the city’s growing population don’t reside in Boston on July 1. And I don’t mean that as in they’re gone for the summer, but rather that many are gone for good and most of the rest aren’t residing in Boston because they’ve been tossed out of their campus dorms or their apartment leases have expired and they’ve moved on. In fact, it seems to me that one could argue that the more students are housed in dorms, the fewer are going to stay in Boston for the summer instead of going back to their parents on Long Island or wherever they’re from (and where they no doubt are claimed as dependents). So, taking all those thousands of new college dorms into consideration into account, maybe Boston’s July 1st population is smaller now than previously estimated.
The St. Clair mentioned above wrote in an op-ed in the Globe in 2006 that
Last year, the Legislature appropriated Secretary of State William Galvin $100,000 for a population projection program at the University of Massachusetts Donahue Institute. That was a step in the right direction, but a tiny one. Actually, $800,000 is needed — a modest sum compared with all the federal funding we could lose if we are undercounted, not to mention the value of a lost member of Congress
She got her wish. Writes Ebbert:
After spending $700,000 over the past two years, the Legislature is expected to devote $800,000 in the upcoming budget to the census effort.
A 700% budget increase to find a 6% increase in the 3.5% of the population who reside in group quarters - an increase that the 2010 Census may well have found anyway, if there is one to be found. One thing is for sure in Massachusetts: The hack population will never decline.
Declining, however, is enrollment Boston public schools, which is surprising since the Mayor boasted after the revision in 2006 that people are moving to Boston because “they want to live here and raise their families here.” Would you care to revise that statement, Mayor?
