It’s no secret that Democratic and Republican politicians often pursue vastly different policies. While it’s common for academics and researchers to study the effects of any given policy, less work has been done to understand the cumulative effects of all of the policy choices made by either party over a given stretch of time. One approach to tackling this question is to analyze the relationship between political party control of state government and a variety of well-being metrics at the state level.
A Tool for Understanding the Relationship
I introduce a dashboard that begins to explore this relationship. Specifically, this dashboard examines how political party control of all three primary levers of a state’s government – the governorship, the state house, and the state senate – is associated with the following metrics pertaining to the health, wealth, and education of the state’s citizens:
- Median Household Income
- Poverty Rate
- Uninsured Rate
- No High School Diploma Rate
- Post-Secondary Education Rate (Bachelor’s Degree +)
- Gross Domestic Product (“GDP”) Per Capita
- Unemployment Rate
The dashboard examines these relationships at the state level over the last 20 years. Each state’s position on the x-axis represents how many years the state government was under complete control – meaning all three primary levers of government under one party’s control, also known as a “trifecta” – by either Democrats or Republicans. For each year a state had a Democratic trifecta, it moves one point to the left. For each year a state had a Republican trifecta, it moves one point to the right. States that fall toward the middle either had a number of years under both Democratic and Republican trifectas over the course of the 20 years, or they had many years under split government control.
Each state’s position on the y-axis represents their average performance on a given metric, with the dropdown menu on the right toggling between metrics. The trend lines for each graph demonstrate how having more years under Democratic or Republican control is associated with better or worse outcomes on that metric. In other words, the trend line indicates the average change in each well-being metric as party control varies. Note that for some of the metrics, lower numbers are better (e.g. Poverty Rate), while for other metrics, higher numbers are better (e.g. Median Household Income). As a result, the direction of the trend line cannot be interpreted the same way for each graph.
To be sure, this dashboard is not intended to imply a causal effect; correlation is not causation. The fact that states with more years of Democratic control have higher median household incomes could be due to policies enacted by Democratic governments, but it could also just mean that wealthier states just choose to elect more Democrats. In other words, there is no way of determining which way the causal arrow goes, or even if there is a causal arrow at all, based on what we see here. Rather, this dashboard is meant to serve as a starting point for investigating the relationship between party control of state government and key well-being metrics.
What the Dashboard Can Tell us
With that being said, the dashboard helps visualize some interesting associations. At first glance, states with more years of Democratic control appear to be associated with better outcomes on five of the seven metrics (Median Household Income, Poverty Rate, Uninsured Rate, Bachelor’s Degree + Rate, and GDP Per Capita), while states with more years of Republican control appear to be associated with better outcomes on the other two metrics (No High School Diploma Rate and Unemployment Rate). However, not all of these relationships are statistically significant at conventional significance levels.
When accounting for statistical significance, states with more years of Democratic control are associated with better outcomes on four of the seven metrics (Median Household Income*, Uninsured Rate†, Bachelor’s Degree + Rate**, and GDP Per Capita†), while no statistically significant relationship exists for the other three metrics.
In a future post, I will seek to explore this relationship further by investigating how each of these metrics evolves over the course of Democratic trifectas vs. Republican trifectas.
† indicates significance at p < 0.1
* indicates significance at p < 0.5
** indicates significance at p < 0.01
Note: Edited on June 10, 2020 to provide more detail on statistical significance.
Find the data and code used in this analysis here.
Photo by Al Case on Flickr
