Did the Affordable Care Act Actually Destroy or Redeem Obama's Approval Ratings?

By Sharon Hoeck

The passage of the Patient Protection and Affordable Care Act (also known as the ACA or Obamacare) in 2010 marked the first significant healthcare reform in the United States in decades. The ACA reforms the way health insurance is sold, aiming to “put consumers back in charge of their healthcare” by attempting to make health insurance more affordable for people with pre-existing conditions, to expand insurance coverage to young adults just graduating from college, to guarantee that individuals have the right to appeal insurance companies’ decisions not to pay their claims, and to expand Medicare. Despite its good intensions, the law has been met with opposition from individuals and politicians across the ideological spectrum who contest that the federal government is overreaching its authority and that healthcare reform should be a state-by-state decision.

Obama and his supporters argued that the electoral victory in 2012 showed that the majority of the American people approved of the law and its reforms. Republicans disagreed, ultimately shutting down the government for weeks in the fall of 2013 in an attempt to effectively repeal the ACA by defunding its programs. Reflecting this ideological split for politicians, the media has also taken sides. Depending on your preferred news source, the ACA is either the best thing or the worst thing that has ever happened to healthcare in this country and will either be the highlight or the bane of Obama’s presidential legacy. Politicians have avoided healthcare reform for years, as changes to insurance policies and Medicare are generally viewed as a dangerous issue with the potential to ruin political careers.

But does the ACA really polarize the American as much as some media outlets and politicians would have us believe? In this post, I will address this question by examining the effect of the ACA on Obama’s presidential job approval rating. To do this, I compare two states: one that has instituted a state exchange through the ACA (Minnesota) and one that has not (Wisconsin). My hypothesis is that, if the ACA is as polarizing as some lead us to believe, Minnesota’s presidential job approval ratings must have changed significantly either for the better or the worse after it passed the law implementing a state exchange. 

I chose Minnesota and Wisconsin as my comparison cases because of their similarities apart from their implementation of the ACA’s state exchanges or Medicaid expansion. Both states, for example, had similar percentages of uninsured individuals prior to the ACA’s passage as well as similar median ages and similar percentages of individuals employed by small businesses. Additionally, despite Wisconsin’s more conservative state-level politics, the states have similar presidential voting records over the past twenty-three years. Most important for my purposes, however, is the similarity between the two states in terms of their approval of the job that Obama was doing as president before Minnesota passed its health exchange law in late October of 2011 (Figure 1). Both states show the decrease in approval that naturally follows a president’s election, and both states show ratings that leveled off in 2010 only to rise slightly in 2011. I can therefore assume that if Minnesota had not established a state exchange in October 2011, its presidential job approval ratings would be similar to those of Wisconsin (see Appendix A for a more detailed explanation of my rationale). Because Minnesota’s law was not signed until the end of 2011, the difference in approval rating trends after that year could be the effect of Minnesota’s full compliance with the ACA (see Appendix B for information on the approval ratings used).  

While Figure 1 shows Minnesota’s presidential approval ratings decreasing at a slower rate than those of Wisconsin following Obama’s election to a second term in 2012, Table 1 shows that this difference in trend is not statistically significant and is therefore more likely due to random noise in the model than any systemic difference in opinion caused by Minnesota’s implementation of a state exchange (see Appendix C for a note on my method).

Table 1. Average Effect of State Exchange Legalization

DV: Presidential Approval Ratings	Pre-Exchange	Post-Exchange	Difference
Minnesota	47.95	48***	0.05
Wisconsin	47.6***	46.1	-1.5
Difference	0.35	1.9	1.55

Note: Differences-in-differences estimates. *, **, *** indicate statistical significance at the 10%, 5%, and
1% level respectively.

Despite the statistical insignificance of my results, comparing Wisconsin and Minnesota shows that the passage of the ACA did not result in a systematic change in public approval for the president. This result brings into question the validity of the idea that healthcare reform is a polarizing and potentially politically deadly issue for our representatives to address. The implementation of the many of the ACA’s provisions did not occur until January of 2014 and sufficient data reflecting public opinion after the law was fully enacted is not yet available. However, future data will only reflect the American public’s opinions on the ability of the law to function properly and make a difference in people’s lives. The data that I used shows that the act of drafting and passing comprehensive healthcare reform in and of itself will not destroy or resurrect a politician’s approval ratings. In this instance at least, the American public is showing themselves to be more levelheaded than some politicians and media outlets assumed. The functionality of the ACA and its programs will have much more of an effect on Obama’s legacy and long-term approval ratings than the political firestorm that surrounded the ACA’s passage. Legislators should therefore reconsider avoiding any discussion on issues that they previously believed to be off-limits or a political “third rail.”

            

Appendices

Appendix A: State Similarities

According to the U.S. Census Bureau, the percentage of Minnesotans without health insurance in 2009 measured at 8.8. Wisconsin’s percentage was not significantly higher at 9.5. A similar percentage of the population in both states had the potential to gain access to health insurance through the ACA, and would therefore affect approval ratings similarly.

The median age in Minnesota, at 37.4, is only slightly lower than Wisconsin’s 38.5 (U.S. Census Bureau, 2010). A similar percentage therefore stand to gain coverage through the ACA provision that allows young adults to remain under their parents’ health insurance until their 26^th birthday.

Additionally, in 2011, the Census Bureau marked the percentage of the total employed population working for small businesses at 22.9 in Minnesota and 23.5 in Wisconsin. Because the ACA mandates that businesses employing more than fifty people full-time must share health insurance responsibilities with their employees by 2016, I limited the above percentages to firms that employ between 50 and 499 people. Many have voiced the concern that small businesses, when forced to share health insurance costs with their employees, would be compelled to either fire many of their staff or significantly reduce their hours. Wisconsin and Minnesota have similar levels of small business employment and therefore relatively similar proportions of the population that would see the ACA as a threat.

Electoral results from both states are also similar, with Obama winning both states in 2008 (with 54.06% and 56.22% of the vote in Minnesota and Wisconsin respectively) and in 2012 (with 52.65% and 52.83%). Further, the popular vote (though not the electoral vote) in the two states has gone to the same candidate in every presidential election since 1992. And while Wisconsin currently has a republican governor and Minnesota has a democratic governor, the state house of both states changed party hands in 2010 elections (from Democrat Jim Doyle to Republican Scott Walker in Wisconsin and from Republican Tim Pawlenty to Democrat Mark Dayton in Minnesota). The two states are therefore prone to similar party affiliation and to a similar inherent approval or disapproval of Obama’s actions.

Despite the differences between the two states, therefore, the major systemic factors in both states that could affect their respective presidential approval ratings are similar enough to allow me to assume that if Minnesota had not established a state exchange in October 2011, its presidential job approval ratings would be similar to those seen in Wisconsin’s.

Appendix B: Data

All presidential job approval data comes from Gallup. The following is paraphrased from Gallup’s website concerning their collection method: Gallup tracks the percentage of Americans who approve of the job that Barack Obama is doing as president. Daily results are calculated from telephone interviews with approximately 1,500 individuals and are based on a three-day rolling average. Margin of error is ± 3 percentage points. 

        

         Table 2. Summary Statistics

	Mean	Std. Dev.	Min.	Max.	Description
Minnesota	50.23	5.74	45.4	61.5	Annual Averaged Presidential Job Approval Rating, 2009-2014
Wisconsin	48.53	5.10	42.4	57.7	Annual Averaged Presidential Job Approval Rating, 2009-2014
Observations	6

           Source: Gallup Daily Presidential Job Approval Ratings 2009-2014

Appendix C: Method

When running the Difference in Differences regression that produced the values in Table 1, I used a lead measure of the exchange variable. For example, had my results been significant, I would have been able to use the presence of an exchange in Minnesota in 2012 to predict presidential approval ratings in Minnesota in 2013. In this way, I can justify the assumption that the ACA influences approval ratings, not vice versa.