More charts about poverty and cities

Here comes more laz-e-boy social science from Alex.  Following up on my response to Steve Berg’s Strib op-ed from a couple weeks ago, I dug up the poverty rate for the 50 metro areas with 1 million population or more.  From that I was able to calculate the spread between central city poverty and metro poverty, which seems like a better measure for concentration of poverty than central city poverty alone.  I then plotted them differentiating by region, to see if any regions bucked national trends.  Actually, I added the regions mostly for a dose of color, as the four regions defined by the census are so broad that they’re almost meaningless.

First up is a chart that also appeared in my last post, but this time it’s spritzed up by some regional color.  It certainly seems to show a trend of central cities with higher poverty rates having lower rates of population growth, although the West may be exempt from this pattern.  Although the trend lines in all of these charts are for all the cities grouped together, it also looks like the trend may be more pronounced in the South.

Central city poverty also appears to correspond with metro area growth, although seemingly more weakly than central city growth, except maybe in the Midwest.  In fact, there are several Western cities on the left end of the chart that had quite low central city poverty rates but also didn’t grow much at all – San Diego, San Francisco, San Jose, and, interestingly for Berg’s thesis, Seattle.  This last town, Berg’s role model for the Twin Cities, had a poverty rate eight points below the national average (which is 21%, Seattle’s rate is 13%) but saw growth exactly at the national average of 13% (beating MSP by only 2.5 points despite our ten point higher central city poverty rate).

If central city poverty correlates with population growth to some degree, is that because of concentration of poverty or just because of poverty in general?  To answer that question I plotted total metro population in poverty against central city population growth.  Lo and behold, the correlation seems to have gotten much weaker.  This is a good time to point out that this is about as lazy a statistical analysis as you can get – technically you’re supposed to do a regression in order to make the variables comparable, but I’m not real clear on how to do that and didn’t want to bother to find out.  Plus I think it’s useful to have the actual numbers on the chart rather than their statistical translations.

The last raspy bark of this fruitless tree of data finds a puffy cloud instead of a relationship between total metro growth and the metro poverty rate.  Or does it?  Does the slight incline of that trendline prove that the capitalist machine thrives on an army of the poor to feed its satanic mills?*

Let’s get back to a more useful line of inquiry.  What does it matter how high the poverty rate is in the central city if it’s also high in the whole metro?  To get at concentration of poverty, we have to look at the difference between central city and metrowide poverty rates.  The above chart seems to show that cities where poverty is concentrated in central cities tend to have a lower growth rate overall, which means that metros should be concerned about their ghettos.  Once again, though, the trend seems to be strongest in the Midwest and South, the latter of which is a bit surprising because so many Southern cities annex their suburbs.  And the West and the Northeast appear to not follow the trend, so I’m not sure I’m convinced.

Comparing concentration of poverty with central city population growth is a bit more convincing.  The trend line is a bit steeper, and the West appears to fall in line, leaving only the stubborn Northeast to buck the trend.  In case you’re wondering, Memphis is where the central city has a lower poverty rate than the metro as a whole – perhaps because Memphis city continues to annex its suburbs and therefore reflects an urbanized population, whereas the metro area includes rural areas as well.  Also, Detroit is the city with a central city poverty rate 20 points higher than the metro rate, which with its -25% population change rate would seem to be Exhibit A for Berg’s thesis.  Hartford may be a prime example for the opposing viewpoint, since it beats Detroit for concentration of poverty, yet managed to grow its central city 3% (amounting to a bit more than 3,000 people).

In Summation

Regardless of whether concentration of poverty has any effect on population growth, I agree with Berg we have a more prerogative to fight poverty, or as he put it, “stabilize poor neighborhoods” for “ethical” reasons.  Unfortunately doing so will require a change in attitudes to the city and citybuilding to basically a polar opposite of what currently exists.  Like Berg I think that expanding transit is a small but important step in that direction – auto-dependence furthers the cycle of poverty because poor people tend to buy less expensive cars which tend to have higher maintenance costs and need to be replaced sooner – and I’d add that it’s more achievable than most of the other options because the only area in which our political system is capable of finding consensus is in the construction of large public works projects.  That’s one of the reasons I started this blog – “working” towards a fairer society by complaining about the auto-orientation of our transportation system – and I was hoping that this series of graphs would have been more help towards this goal, but I don’t seem to have the time or knowledge to really pull it off.  If anyone reading this does, here’s the data I gathered – give it a go.

*No.  No it doesn’t.

About these ads

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s