#TBT: Cause and Correlation and You’ve Still Got to Do Something People Want
Happy #TBT. To celebrate, we’re sharing an oldie-but-goodie post from Jim: Cause and Correlation and You’ve Still Got to Do Something People Want.
I was very interested to read through a short piece posted by Gary Steuer, who is the Chief Cultural Officer for the City of Philadelphia, about the city’s (modest) rebound in population [circa 2011]. Here’s a key snippet:
“The new Census numbers for Philadelphia are in, and the city managed to actually record a population increase, the first in 50 years. And while the increase was tiny — 8,456 residents, which represents a .6% increase to 1,536,006, the reversal of the decades-long decline is huge.”
Given the uncertainty of the accuracy of statistics on this scale, a .6% increase is probably statistically close to a wash, but the point is that Philadelphia’s population decline has stopped, and that’s a healthy sign.
Steuer seems to have a grasp on the complexity of the question of why, but puts forward a few hard-to-refute ideas and then a couple that are easier to debate. First, it’s clear that in-migration from Hispanics and Asians have at least replaced the out-migration from whites. I’d add that it’s pointless to discuss, statistically speaking, the population growth in Philadelphia without discussing these two factors, because net of the group in these two groups, Philadelphia would have lost a substantial portion of its population over the last decade.
Second, he points out that some of the largest growth has come in the Center City area, where the retail, restaurant and arts assets are most dense. This shouldn’t surprise anyone, as it’s happening elsewhere in the United States. Someone should (and may already have) crosstab the first trend (toward Asian and Hispanic residents) and the second trend (where the neighborhoods of growth are, including Center City) to see if people coming to Philadelphia are moving to these neighborhoods or if people who were already in Philadelphia are moving to those neighborhoods.
Far less clear is the role of arts organizations in this improvement. What we’d all like to believe is that if arts organizations go into an area, it’s a leading indicator of improvement in that area. The problem, of course, is that it can also work in reverse: If an area starts to improve, arts organizations could be drawn to it as an attractive place to find a good audience before rents and costs get too high.
And that’s why the difference between cause and correlation is so important to understand. It may be true that the neighborhoods with the population growth have more arts organizations in them, but that doesn’t mean those arts organizations very presence caused that increase. It’s probably also true that the neighborhoods that saw the most population growth have more hybrid cars in them, but no one would suggest that the hybrid cars caused the population increase. (By the way, I think Steuer understands this, and so isn’t making the mistake of saying that the arts organizations are the cause, though I do feel he’s suggesting that it’s possible. Fair enough.)
So what’s the lesson for arts organizations? I think it’s that nothing about a neighborhood suspends the basic laws of nature. You’ve still got to do good strategic marketing in the sense that you have to deliver something that people want, and there have to be enough of those people show up and pay the bills in one way or another. Being in an undesirable neighborhood can be a big boulder for an organization to overcome, so making some kind of assumption that the presence of an arts organization will improve a neighborhood is risky. You only hear about the happy stories where it kinda worked out. All the ones where the organization quietly vanished tend not to get big coverage.
There’s an investing maxim (I think Keynes said it) that goes like this: The market can stay irrational longer than you can stay solvent. In other words, you may be right about the potential of a neighborhood to improve, but it can stay blighted longer than you can keep the lights on.
I’m not saying that people shouldn’t consider new neighborhoods. I’m assuming cutting-edge artists always will and that there’s something about an extraordinary artist’s sensibility that sniffs out the neighborhoods on the way up before the rest of us do. I’m just saying if you confuse correlation and causality, you could be making a very expensive mistake.