Do taxes matter when it comes to employment growth?


In the world of policy, there’s no more hotly debated issue than this.  With state governments (and many adoring media outlets) continually arguing that governments don’t have enough money to do all the wonderful things elected officials and their appointed bureaucracies want them to do, increased state government revenue from increased tax burdens will likely continue to be the marching song from both left and right-wing bureaucracy lovers for, at the very least, the next few years.

But, what do the data show on taxes and employment growth?  Well, here are data, and please take note of this caveat: the plots are the raw results in video form over time.

Now, academics will debate ad nauseum the causes of the observed negative correlation between taxes and employment growth.  Some, likely most, will find various theories to explain why higher tax burdens do not cause lower employment growth.  All of these theories are contrary to the observed correlations.

To outside observers, it comes as no surprise that most of academia would try to dismiss any raw empirical results regarding a negative relationship between taxes and employment growth because, after all, where does academia get its support from?

Additionally, as most reading this article are likely well aware, correlation doesn’t mean causation.  The “correlation doesn’t mean causation” reaction has one big flaw: relatively strong correlation between taxes and employment growth also does not mean there is no causation and relatively strong correlation is a better argument than counter-factual theories.

The bottom line is this: the negative correlation between taxes and employment is there and generally consistent over time.  The question is what elected officials want to do with the results.

With this background and these caveats in mind, here’s what the relationship between state employment growth and income tax burden per household looks like from 1990 to 2012.

As a note, the negative linear line means that, in general, when employment growth is higher (vertical axis), tax burden per household is generally lower (horizontal axis).

Of course, the flip holds as well, meaning that higher tax burden per household (further along on the x-axis) is generally associated with lower employment growth (further down on the y-axis).

Income Taxes per Household from Sheeraz on Vimeo.

Here’s another look at the relationship between taxes and employment growth, with corporate income tax burden replacing income tax burden per household.

State Taxes Burden per Household from Sheeraz on Vimeo.

In addition to the individual income tax and corporate income tax look, here’s what things look like for total state tax burden per household and employment growth.

Corporate Taxes per Household from Sheeraz on Vimeo.

Perhaps unsurprising for people with a real view of the world, but the relationship between taxes and employment growth is consistently negative over time, with only the slope of the relationship shifting.  Additionally, most of the weakening of the relationship is due to the business cycle, rather than some long term mean reversion type of an idea.

Overall, as the videos show, the common sense view that higher taxes are associated with lower long-term employment growth is factually correct.  Policymakers should really think about the raw relationship between taxes and employment growth when they pontificate on all the great things they can do with rich people’s money.