America’s Urban Distress: How Much of the Problem Can We Blame on Liberal Politics?


Much of the distress in America’s cities is linked to regional developments, as shown by the unemployment data in this post. Population loss is a particular factor for the Northeast and Midwest (discussed here), while public pension shortfalls may be the biggest challenge for cities everywhere (covered here).

But what about politics? Does ideology play a role?

By popular demand (at least based on comments on the earlier articles, especially on Zero Hedge), below are two charts that compare unemployment rates to political bias for 218 cities.

Although the figures for political bias are from a 2005 report, they’re still relevant in my opinion, considering that:

  1. City-wide biases are unlikely to change rapidly.
  2. Political actions have long-term effects on economic performance.

Some readers will argue that correlation isn’t causation, while others are likely to conclude that causation runs from unemployment to politics and not the other way around.

I believe there’s some causation in both directions. But for now, I’ll just push the charts out there and step out of the way while you form your own conclusions…

UpdateAfter the charts, I’ve added the list of 25 most liberal cities from the Bay Area for Voting Research report.

cities and politics 1


If you prefer a (non-Talebite) statistical view:


cites and politics 2

And here are the most liberal cities, circa 2005.  The top-ranked city is a former automobile manufacturing powerhouse located in the upper Midwest.  The photo (above) shows the Gary City Hall – the second most liberal city.  Berkeley, California – the third-ranked city – happens to be the home of one of CYNICONOMICS’ biggest fans.

list of most liberal cities

Bookmark and Share
This entry was posted in Uncategorized and tagged , , , , , , , , , , . Bookmark the permalink.

8 Responses to America’s Urban Distress: How Much of the Problem Can We Blame on Liberal Politics?

  1. steve says:

    With all of your data, why did you suspiciously omit the top 25 cities with the most conservative/republican bias….

    I wonder if the results would go against your seemingly biased agenda.

    • ffwiley says:

      Actually, all the cities in the voting preferences study were included in both charts. I updated the post with the list of liberal cities at the end because I thought they may be of particular interest. Real suspicious that I would follow a post about liberal politics (see title) with a list of liberal cities, huh?

      But hey, I’m more than happy to report the top 25 conservative cities for anyone interested in both extremes:

      1 Provo Utah
      2 Lubbock Texas
      3 Abilene Texas
      4 Hialeah Florida
      5 Plano Texas
      6 Colorado Springs Colorado
      7 Gilbert Arizona
      8 Bakersfield California
      9 Lafayette Louisiana
      10 Orange California
      11 Escondido California
      12 Allentown Pennsylvania
      13 Mesa Arizona
      14 Arlington Texas
      15 Peoria Arizona
      16 Cape Coral Florida
      17 Garden Grove California
      18 Simi Valley California
      19 Corona California
      20 Clearwater Florida
      21 West Valley City Utah
      22 Oklahoma City Oklahoma
      23 Overland Park Kansas
      24 Anchorage Alaska
      25 Huntington Beach California

      I also noticed the link that I included had the most liberal and most conservative cities but not the full study. Here’s another link with many more lists that I “omitted”:

  2. PaperIsPoverty says:

    What’s in this model? Just unemployment and liberal % of vote? If so, I’m afraid this is just not interesting. The whole strength of regression analysis is that it allows you to separate out the contributions of different predictive factors. In sociological models you’ll often see a dozen or more variables because there are so many related factors. If this is a simple two-variable regression I’m afraid the R-square is just meaningless. It is possible to have a simple two-variable correlation disappear or even switch direction once other confounding variables are included. I’m not suggesting that would happen here, I am just pointing out the general importance of controlling for important factors. If you take a sample of elementary school children and you don’t control for their ages, you can find a stunning correlation between reading scores and shoe size. Stick age in the model and it’s gone.

    • ffwiley says:


      I could come up with a long list of additional caveats to “sociological models” of all shapes and sizes, but that won’t stop me from wondering how the most troubled cities line up against measures of political bias and showing the results here.

      That said, I’d be happy to include a multi-variable regression with a different sign on political bias if you can create one that makes any sense. (I’d be checking a few things and especially collinearity since you’ll be looking to flip the sign.)

      Or better yet, instead of using an example that has nothing to do with urban unemployment, maybe you could suggest a variable (or more) that would conceptually cause the relationship between liberal policies and unemployment to reverse. It’s not as easy to do when you’re not starting with a nonsensical pair of variables like reading scores and shoe sizes. In fact, I think you’ll find it difficult. If you’re determined to refute the idea that liberal policies have contributed to unemployment, I would have thought you would have just argued that the causation is in the other direction.

  3. shutupnsing says:

    Great post FW! PisP & Steve would debate the H & O content of water in the Mojave as they were dying there of thirst…The left “owns” nothing that could lead to their own accountibility!

  4. agonzo says:

    This post also ignores how “conservative” areas receive more in federal subsidies via agriculture assistance than liberal areas.

    • russ says:

      I’d say the agricultural subsidies going to the top 25 conservative cities is negligible.

      There’s certainly no good reason Detroit couldn’t start getting in on the agricultural gravy train since they’ve got plenty of open space. Other than the fact that the ag subsidies wouldn’t make a dent in their debt problem even if the subsidies went directly to the town coffers.

Comments are closed.