About a year ago, I explored the notion that Republican-led states (known for being more relaxed in the area of Covid restrictions) were experiencing a higher rate of Covid-19 deaths than Democrat-led states (known for being more stringent in the area of Covid restrictions). Specifically, I wanted to see if there was a correlation between the severity of a state’s Covid restrictions and their death rate. I found that there was not. Some of the strictest states had higher death rates, and some of the freest states had lower death rates.
A lot has changed since last July, and now that the pandemic is essentially over, I wanted to re-run the data and see how the states fared at the end of the day. This time, I added an additional data set to the mix. We know that the majority of Covid deaths occurred in elderly populations, so states with a higher percentage of elderly people would be expected to have a higher death rate than states with lower percentages of elderly populations. Is that true to experience?
I also wanted to determine if there was a correlation between lockdown severity and lower death rates, and if there was a correlation between a state’s political party and their death rate. Here is what I found:
Democratic vs. Republican states
Democratic states were definitely more restrictive in their lockdown policies than Republican states. Twenty-one of the 25 most restrictive states are run by Democrats, whereas 23 of the 25 freest states are run by Republicans. If more restrictive lockdowns result in a lower death rate, we would expect Republican-led states to experience a higher death rate compared to Democrat-led states. Actually, the opposite is true. Democrat-led states saw a death rate of 174 per 100,000 people (including New York City) vs. a death rate of 168 in Republican states. Contrary to popular belief, then, the death rate was higher in states with greater restrictions. Lockdown severity does not necessarily lead to lower death rates.
Freest vs. most restrictive states
The 25 most restrictive states had an average death rate of 164 vs. 169 for the 25 freest states. That’s a difference of only 5 deaths per 100,000 people, or 3%. However, these figures do not take New York City into account because the site I used to assess lockdown severity did not rank New York City separately (and the death count numbers for New York state does not include the death count for New York City). Given my knowledge of New York City’s lockdown policies, my guess is that New York City would rank very high on the restrictive list. If we account for the death rate in New York City, then, the most restrictive states experienced a death rate of 173, which is 4 deaths higher than the freer states.
No matter how you slice it, the difference in death rate is statistically insignificant. Again, this shows there is no correlation between the severity of a state’s lockdowns and their death rate. This is illustrated very well by a graph from WalletHub. They plotted states according to four groupings:
- Few restrictions and high Covid deaths (red)
- Few restrictions and low Covid deaths (green)
- Many restrictions and low Covid deaths (blue)
- Many restrictions and high Covid deaths (gray)
As you can see below, states are fairly evenly distributed among these four groupings, illustrating the fact that there is no correlation between lockdown severity and death rates:
Top 10 states
The top 10 states with the lowest death rate are split evenly among Democrats and Republicans. The same is true of the top 10 states with the highest death rate. The political party governing a state, then, is not a good predictor of the death rate.
How restrictive a state’s lockdown policies were did not correlate to a lower death rate either. Of the top 10 most restrictive states, 3 had some of the lowest death rates and 3 had some of the highest death rates. Of the top 10 freest states, 1 had some of the lowest death rates and 2 had some of the highest death rates. This is counterintuitive if lockdowns are correlated with lower deaths.
If you average the death rate for the 10 most restrictive states and compare it to the average death rate of the 10 freest states, you’ll find it’s only a difference of 12 deaths per 100,000 people, or a 7% difference (restrictive = 161, free = 173). This does not take New York City into account because the site I used to assess lockdown severity did not rank New York City separately. My guess, however, is that New York City would rank very high on the restrictive list, which would bring down the gap between Republican and Democrat states to virtually zero.
I found no clear correlation between higher elderly populations and higher death rates. Four out of 10 states with the lowest death rates also have some of highest percentages of elderly populations. Ironically, only one of the 10 states with the highest death rate also has one of the highest percentages of elderly citizens.
Freest state vs. most restrictive state
Vermont was the strictest state, and they had the 2nd lowest death rate despite having the 4th oldest population. While this would seem to support the thesis that stricter lockdowns result in lower death rates, Vermont is actually an anomaly.
Iowa was the freest state, but they ranked #33 for death rates despite having a fairly young population (ranked #17 for oldest population).
Oldest vs. youngest states
The 10 states with the oldest populations saw an average death rate of 131 vs. 144 for the ten states with the youngest populations. That’s a difference of 13 deaths per 100,000, or 9%. While this is statistically significant, it is clearly not a main predictor of Covid deaths in a state seeing that 4 of the 10 states with the lowest death rate also have the highest percentage of elderly people and 1 of the 10 states with highest death rate also has one of the lowest percentage of elderly people (8 out of the 10 states with the highest death rate have an elderly population that does not rank in the top 10 oldest or youngest population).
California vs. Florida
Seeing that the media loved to praise California for its handling of the pandemic, while criticizing Florida and Texas’ approach, I want to compare these states against each other to see who had the best approach.
Florida was the 2nd freest state in terms of lockdowns, while California was the 6th strictest state. The difference in the death rate was minimal: 175 for Florida vs. 159 for California. Florida ranked #25 in death rate among all states vs. #19 for California. When you factor in the percentage of the elderly population, however, Florida clearly did better. Florida has the 2nd highest percentage of elderly population in the nation, whereas California has the 6th youngest population. Only 14.3% of California’s population is age 65+, whereas 20.5% of Florida’s population is age 65+.
California vs. Texas
Texas was the 5th freest state in terms of lockdowns, while California was the 6th strictest state. The difference in the death rate was minimal: 176 for Texas vs. 159 for California. Texas ranked #26 for death rate vs. #19 for California. Both states have a similar percentage of elderly population (12.6% for Texas and 14.3% for California), but California has a slightly higher percentage, which means lockdowns may be partially responsible for the slightly lower death rate in California.
Conclusion
Based on this data, I find no clear correlation between the severity of a state’s lockdown or the size of their elderly population and their death rate. At the end of the day, there was little difference in the death rate between Republican and Democrat states, or between restrictive and free states generally. The only clear correlation is between the severity of a state’s lockdown and the political party leading the state. It seems to me that the lockdowns were largely ineffective at preventing Covid deaths.
Here is my data:
| State | Lockdown Severity (1=most severe) | Deaths per 100K | Death rank (1=best) | % of elderly | Elderly % rank (1=highest %) | Governor’s political party |
| Vermont | 1 | 41 | 2 | 19.4 | 4 | Republican |
| Delaware | 2 | 173 | 24 | 18.7 | 5 | Democrat |
| Virginia | 3 | 133 | 14 | 15.4 | 40 | Democrat |
| New York* | 4 | 180 | 28 | 16.4 | 26 | Democrat |
| Washington | 5 | 77 | 7 | 15.4 | 40 | Democrat |
| California | 6 | 159 | 19 | 14.3 | 45 | Democrat |
| Connecticut | 7 | 232 | 45 | 17.2 | 14 | Democrat |
| Maine | 8 | 63 | 4 | 20.6 | 1 | Democrat |
| Rhode Island | 9 | 257 | 48 | 17.2 | 14 | Democrat |
| New Jersey | 10 | 297 | 50 | 16.1 | 30 | Democrat |
| Minnesota | 11 | 136 | 15 | 15.9 | 31 | Democrat |
| Michigan | 12 | 209 | 40 | 17.2 | 14 | Democrat |
| Massachusetts | 13 | 261 | 49 | 16.5 | 24 | Republican |
| Hawaii | 14 | 36 | 1 | 18.4 | 7 | Democrat |
| Pennsylvania | 15 | 216 | 41 | 18.2 | 8 | Democrat |
| Oregon | 16 | 65 | 5 | 17.6 | 11 | Democrat |
| Illinois | 17 | 202 | 37 | 15.6 | 39 | Democrat |
| New Mexico | 18 | 206 | 39 | 17.5 | 12 | Democrat |
| Colorado | 19 | 117 | 10 | 14.2 | 46 | Democrat |
| Kentucky | 20 | 161 | 22 | 16.4 | 26 | Democrat |
| Nevada | 21 | 183 | 30 | 15.7 | 36 | Democrat |
| Ohio | 22 | 173 | 23 | 17.1 | 17 | Republican |
| North Carolina | 23 | 127 | 13 | 16.3 | 29 | Democrat |
| Louisiana | 24 | 230 | 43 | 15.4 | 40 | Democrat |
| Maryland | 25 | 161 | 21 | 15.4 | 40 | Republican |
| Kansas | 26 | 176 | 27 | 15.9 | 31 | Democrat |
| Georgia | 27 | 201 | 36 | 13.9 | 47 | Republican |
| Indiana | 28 | 205 | 38 | 15.8 | 35 | Republican |
| New Hampshire | 29 | 100 | 8 | 18.1 | 9 | Republican |
| Alabama | 30 | 231 | 44 | 16.9 | 21 | Republican |
| West Virginia | 31 | 160 | 20 | 19.9 | 3 | Republican |
| Wisconsin | 32 | 139 | 16 | 17 | 19 | Democrat |
| Utah | 33 | 73 | 6 | 11.1 | 50 | Republican |
| North Dakota | 34 | 200 | 35 | 15.3 | 44 | Republican |
| Tennessee | 35 | 183 | 29 | 16.4 | 26 | Republican |
| Arizona | 36 | 246 | 46 | 17.5 | 12 | Republican |
| Nebraska | 37 | 116 | 9 | 15.7 | 36 | Republican |
| Arkansas | 38 | 195 | 34 | 17 | 19 | Republican |
| Missouri | 39 | 150 | 17 | 16.9 | 21 | Republican |
| Idaho | 40 | 120 | 11 | 15.9 | 31 | Republican |
| Montana | 41 | 155 | 18 | 18.7 | 6 | Republican |
| Mississippi | 42 | 248 | 47 | 15.9 | 31 | Republican |
| Oklahoma | 43 | 186 | 31 | 15.7 | 36 | Republican |
| South Carolina | 44 | 190 | 32 | 17.7 | 10 | Republican |
| Alaska | 45 | 50 | 3 | 11.8 | 49 | Republican |
| Texas | 46 | 176 | 26 | 12.6 | 48 | Republican |
| South Dakota | 47 | 229 | 42 | 16.6 | 23 | Republican |
| Wyoming | 48 | 127 | 12 | 16.5 | 24 | Republican |
| Florida | 49 | 175 | 25 | 20.5 | 2 | Republican |
| Iowa | 50 | 194 | 33 | 17.1 | 17 | Republican |
| NYC | 397 | Democrat | ||||
| *Excludes New York City |
| · Lockdown severity ratings from https://wallethub.com/edu/states-coronavirus-restrictions/73818, as of April 6, 2021. |
| · Percentage of seniors from https://www.prb.org/resources/which-us-states-are-the-oldest, as of 2018. |
| · Covid deaths per state from https://covid.cdc.gov/covid-data-tracker/#cases_deathsper100k, as of June 28, 2021. |
Thinking to Believe 
July 1, 2021 at 2:12 pm
You might wish to fact-checked your results with medical experts and statisticians to see if you missed a few things, because they came up with the opposite conclusion:
https://www.medicalnewstoday.com/articles/covid-19-has-hit-people-in-republican-led-states-hardest-study-finds
https://www.politifact.com/factchecks/2020/oct/01/joe-biden/yes-coronavirus-deaths-red-states-add-second-highe/
https://www.usatoday.com/story/news/factcheck/2020/06/05/fact-check-comparison-covid-19-data-state-politics-false/5252816002/
It’s true the blue states (mostly with Democratic governors) were hit hardest by COVID-19 initially, since the biggest cities tend to be blue and they are the hubs for international travel, making them the most likely locations for initial outbreaks. However, once those states implemented safety procedures and got the situation under control (around July), death rates became highest in red states (mostly with Republican governors).
The sad thing is that if everyone had cooperated and maintained a strict universal lockdown for just 2-3 weeks, the disease would have been wiped out right then, saving trillions of dollars, millions of jobs, and preventing untold misery. But politicizing mask wearing, social distancing, etc. and denying there was even a problem caused almost half the country to defy rational safety procedures. Thus we are where we are now.
LikeLike
July 1, 2021 at 10:16 pm
My death data is up-to-date and coming straight from the CDC. The lockdown severity data is coming from Wallethub, and that was updated as of April. The senior population data is standard data. Political leadership of each state is common knowledge. I don’t think you’ll find any problems with the data itself. The question is my evaluation. So go ahead and copy and paste the data into an Excel spreadsheet and sort it yourself. Tell me if you see any errors in my calculations.
LikeLike
July 2, 2021 at 4:24 pm
Yes, it’s the evaluation I’m questioning, not the data itself. The thing is, the raw data lacks context. As I noted above, the blue states were hit hardest by COVID-19 initially, since they tend to have the most densely populated cities that are hubs for international travel, which made them the first targets for the initial outbreaks and they got hit hardest. The red states were largely spared initially because they are mostly more sparsely populated and less urban. Once the blue states got a handle on the virus, their death rates mostly plummeted while death rates rose in red states. Since your numbers don’t account for population density or any other factors (see the links I provided for more factors), they give the false impression that Republican governors handled the pandemic better. The reality is a far more mixed bag (https://www.cnn.com/2021/03/10/politics/covid-cases-deaths-red-blue-states-late-2020/index.html and https://www.cnn.com/interactive/2020/health/coronavirus-us-maps-and-cases/)
It’s also worth noting that over 99% of all current deaths due to COVID are occurring only among those who have not been vaccinated (https://apnews.com/article/coronavirus-pandemic-health-941fcf43d9731c76c16e7354f5d5e187). And the states with the worst vaccination rates are largely red states, with Mississippi, Louisiana and Alabama at the bottom (https://www.mayoclinic.org/coronavirus-covid-19/vaccine-tracker). Republican governors are the least likely to encourage vaccination, and they’re also the ones most likely to downplay the risks of COVID and ending mask mandates (https://www.usatoday.com/story/news/factcheck/2021/03/15/fact-check-republican-governors-have-mixed-record-mask-mandates/6946193002/). So it should come as no surprise that Republicans are six times more likely than Democrats to refuse the vaccine (https://thehill.com/policy/healthcare/558225-nearly-30-percent-of-republicans-say-they-wont-get-vaccinated-poll), and thus they are the segment of the population most likely to die from COVID. So, it’s fair to conclude that generally Republican governors are NOT handling the pandemic as well Democratic governors.
LikeLike