Tax Compared to Unemployment in the USA

Conservatives and such, like to say here in the USA, that high taxes make a higher unemployment rate. I tested this theory by comparing the tax revenue of personal income of the individual states, with the unemployment rate of each individual state, for the year of 2007. My results are on this link.

I also did the Pearson coefficient for it to see how relative it was to being either directly proportional (in which case higher taxes is related to higher unemployment), inversely proportional (in which case higher taxes is related to lower unemployment), or not related at all. When I got the number back the result was 0.0757666384… meaning that there is pretty much no relation between the two, if what conservatives say was right then it would have given at least 0.3-0.5, but that’s not the case. The hypothesis I get out of this number is that, in the USA at least, there is no relation between taxes and unemployment.


3 responses to “Tax Compared to Unemployment in the USA

  1. Actually, a perfect correlation in the case of the Pearson coefficient would show up with a +1.0 (i.e., when an increase in variable X also effects an increase in variable Y) or -1.0 (i.e., when an increase in variable X effects a decrease in variable Y). In this case, if the usual conservative hypothesis (i.e., that an increase in the tax burden leads to a higher unemployment rate) were correct, you should see a Pearson coefficient of -1 if such correlation is perfect or, at the very least, something as close to -1 as possible. Yet. what you see is something that is very close to 0, meaning that there is no correlation whatsoever between the two. Thus, if there is no correlation, there cannot be a cause-effect relationship either.

    You will find more information in the Wikipedia entry for the Pearson correlation coefficient:

    Unfortunately, as it tends to be the case in politics, facts and empirical evidence matter little. Neither politicians nor media care about this sort of analysis. It’s too boring and requires an effort on behalf of the audience (not to talk of the people who put it together).

    Congratulations! Nice work. :)

  2. Yep, I stand corrected. If the hypothesis were correct (and, therefore, there is indeed a direct correlation between tax burden and unemployment in the sense that an increase in one also implies an increase in the other, as conservatives usually state), then the Pearson coefficient in this case should be close to +1. In any case, instead of that, it’s extremely close to 0, therefore indicating that there is no correlation, at least in the data that was studied. In other words, that the usual statement that “higher taxes lead to higher unemployment” cannot be considered correct, judging from the empirical evidence.

    As I explained to you during a conversation on these topics, my guess is that the also typical conservative statement according to which “higher taxes lead to less wealth” is not true either. You could analyze that correlation next. :)

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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s