Do Unions Kill Prosperity? Not So Much!

With a correlation of .56, it appears the higher the union membership in a state, the higher that state’s GDP.  Imagine that…

Unions gdp by state

Contrary to the wingnut talking point that “unions destroy prosperity,” the exact opposite seems to hold.

Data is such a bitch…

Some Fun Observations:

  • Lower-Left = Red States
  • Upper-Right = Blue States
  • Sarah Palin’s Alaska is 2nd only to New York in percent of employees who are unionized

Data:

Union density: http://www.bls.gov/news.release/union2.t05.htm

State GDP/capita: http://www.bea.gov/regional/gsp/

Share:

Related Articles

6 thoughts on “Do Unions Kill Prosperity? Not So Much!

  1. Phil, while I’m sympathetic to your point, I think you’re comparing very slight apples and oranges. Using the same data sources as you, but substituting GDP per Employed Worker for your GDP per Capita, I found a much smaller correlation. I think this is an important change, since union representation percentile is a percent of employed workers and not general per capita. While the shape of states on the resulting graph is similar to the one shown above, the r-squared value is only 0.14848. The top GDP/worker state (Delaware) has only the 24th highest union representation/worker. The third highest GDP/worker location (Wyoming) has the 35th highest union representation/worker. But in the end, neither your graph nor mine shows anything about causation versus correlation. That’s still an uphill argument.

    1. Thanks for taking the time to pull the data. Please feel free to send me a copy of your analysis and I’ll consider a post based on it (if the numbers work out). However, I do not necessarily agree that this is apples / oranges. GDP/Capita is a standard measure of productivity. But it’s an interesting point you make.

  2. So Dan, can I conclude that you agree with Phi, that there is no evidence that unions kill prosperity?

  3. Keith, I will agree that using only the data points of gross domestic product per state per worker and percent of union representation per state per worker, there is inconclusive and slight indication that perhaps union representation and GDP rise and fall together. At the very least, GDP and union representation aren’t inversely correlated in these two datasets.

    But in order for us to approach anything resembling evidence, which is to say proof of causation, we’d need annual regressions since unionization began, and not just for the US but elsewhere as well. Furthermore, we’d need to adjust for any number of variables that affect GDP. That’s a tough job.

Comments are closed.