COVID-19, Red vs. Blue vs. Europe

I’ve observed a lot of angry debate about which country/state/political party has handled the COVID-19 crisis the best. As I read various articles online, I struggled to find anything that presented quantifiable evidence as opposed to partisan rhetoric, let alone anything that clearly and succinctly pulled the data together, so last night I did it myself. This article is for you to see what I found.

I organized data from publicly available sources into one table which I present below. The rest of the article explains how I pulled the data together.

How I determined Red vs. Blue vs. Purple States and European Countries

I went to Wikipedia, “Table of U.S. state party statistics as of November 2019“. I pulled the data into a spreadsheet and designated a state as “blue” if the Republican Party had no control over the state house, state senate, or governorship. I designated as “red” if the Democratic Party has no control of the state house, state senate, or governorship. All others were designated as “purple”. For European countries I picked the five largest Western European countries and added Sweden because of its well known practices related to COVID-19. (I didn’t add other smaller countries such as Denmark because it takes the conversation down a rabbit hole.)

How I determined the percentage of population who died

I went to Worldometers for both country and state. I pulled the number who died and put it in my spreadsheet which already had party affiliation and population. I divided those who died by the population of the political unit to make percentages who died.

How I determined the economics

This was much more challenging than I expected.

You cannot compare unemployment data from the U.S. vs. Europe because European countries count people who are employed but with reduced or no hours as “employed”.

It is difficult to compare U.S. GDP results to those of Europe, because the U.S. displays its data on an annualized basis while other countries show quarterly results.

The OECD computes rates for both Europe and the U.S. on the same basis. (Here is a chart.) Nobody seems to be disputing the OECD’s figures. For example, here is an article by CNN using this sort of metric for comparison.

I could only find U.S. state data on an annualized basis. So, I took the U.S. state data and applied a ratio to it, to make it in the aggregate equal to the OECD total US GDP change of -9.0%. The ratio used was “9.0/31.4”. By multiplying this ratio against the annualized GDP per state, I produced a value that in the aggregate would produce the -9.0% GDP that the OECD says the USA experienced. Next, I used the proportion of GDP of each state in 2020Q1 to weight the changes that happened from Q1 to Q2 to produce a weighted average change for Red vs. Blue. vs. Purple states.

For US states, I obtained the following results:

How can you do this?

I’ve given you all the links. If you would like my spreadsheet, contact me, I’ll gladly share. Also, if you detect something wrong with my data, please let me know! I’ll fix/update this article.

Why do I do this?

When I finish doing my day job, I read the news and think about what is going on. I observe a lot of heated political discussion. I am surprised at how hard it is to answer basic questions, so I search the web. Then I think that I might as well share on social media. I’ve done this with regards to COVID-19, wildfires, Joe Biden’s longevity, and the flu.

What is my day job?

Most of my career I’ve been in management of insurance companies. Currently I am helping some insurance companies with some projects, and I am also building algorithms for FraudSpotters. These algorithms are very successful at detecting fraud rings who are trying to scam insurance companies. I believe we have invented something groundbreaking. If you think our algorithms can help you, please contact me and I’ll gladly run them for free on your data to prove the point.

Mathematics of the Flu vs. COVID-19

No alt text provided for this image

The World Health Organization has advice about how to deal with the flu. They also have some articles that compare and contrast the flu with COVID-19. (For a recent interview, click here.)

Likewise, the CDC have information on the flu. (And here.) They also have articles that compare and contrast the flu with COVID-19.

We are all aware of how deadly Coronavirus can be. I wrote an article here. Three weeks later, the CDC produced their own article here. The table below compares/contrasts the results of the two articles. My article (labelled in the table as BT/NY) relied on New York data. The CDC’s estimate appears to be nationwide. From here on, in this article, when I refer to the survival rate of COVID-19, I am referring to the CDC’s figures.

The CDC provides recommendations on how to deal with the situation, as does the World Health Organization. (The World Health Organization also has a myth buster section.) I am providing no additional guidance on how to deal with the COVID-19; I only reference the CDC’s estimate of the COVID-19 survival rate to provide context.

Having said that, how deadly is the flu?

The CDC provides estimates of how many people will show symptoms of the flu and, of these, how many will die as a result. The table below is an image from their website. (Note that “95%Ul” means the upper 95% percentile and lower 5% percentile of their estimate.)

We can take the ratio of symptomatic illnesses to deaths to estimate the survival rate of those who have symptoms.

The above table shows the survival rate for those having symptoms of the flu.

Please note, that this is for all Americans and 40 to 60% of Americans receive the flu vaccination, as the chart below shows. One can only guess what the flu death rates would be without a vaccine.

You may be wondering about the frequency of asymptomatic carriers of the influenza virus. This is difficult to answer. I believe this article, published by the National Institute of Health, provides the best overview of knowledge on this topic. The article shows that academic studies which attempt to estimate this value have a very wide range. The estimates of asymptomatic fraction range from 0% to 100%. For purposes of this article, let’s assume that the asymptomatic fraction is 50%. This would imply that for every symptomatic case of the flu, there is an additional person who is infected with the influenza virus who could theoretically be passing it on to others.

If we adjust our table by this ratio, our mathematics look like this.

Because of the massive coverage of the ongoing COVID-19 pandemic, most people are aware of the lethality of COVID-19. However, many people may not be aware of corresponding data on the flu. The table below compares/contrasts these statistics.

No alt text provided for this image

You can compare the survival and death rates of various age ranges using the table above.

Post script:

You may be wondering how often individuals contract the influenza virus. The image below, copied from Oxford Academic, provides an estimate of the frequency of flu symptoms of 8.3% for all ages (which takes into account the flu vaccine). If we assume a yearly rate of 8.3% of Americans contracting symptomatic flu and 8.3% contracting asymptomatic flu (consistent with our assumption of 50% asymptomatic above), then we are assuming 16.6% of Americans will contract the influenza virus each year. If we assume this 16.6% is constant across the population from year to year, then using a Kaplan-Meier estimate, the median American (assuming 50% asymptomatic) will contract the influenza virus approximately every 4 years (although symptoms may appear on average only every 8 years). (You see this by multiplying (1-.166) until you get below .5)

Some sources on the flu vs COVID-19 can be seen by John Hopkins here, and the CDC here. Note that the CDC says that both the flu and COVID-19 can spread prior to a person becoming symptomatic.

If you found this information useful, please consider reposting.

How to Calculate the Mortality of Joe Biden

Written by Benjamin Turner

This article was published on 10/9/202 on 7:15 am on LinkedIn. It was removed on 9:15 am of the same day by LinkedIn.

Fraudspotters, LLC, nor its employees express any opinion about this article. We are publishing it out of solidarity with an employee.

As an actuary, I’m asked how long to expect Joe Biden to live.

This is a morbid question but relevant. We just experienced a VP debate which may have actually been our presidential debate, and this is a good opportunity to teach people about actuarial mathematics. And, as an actuary, this type of stuff is literally what I do, so it is not morbid to me; it is common place.

So, without further ado…

To calculate the mortality rate go to a published actuary table. An obvious one is found here, and I will reference this table. You should see a webpage that looks like the image below. This is a table produced by the Social Security Administration to project how long people will live for purposes of understanding how much will be paid from Social Security (at a younger age, I worked as a pension actuarial analyst for Mercer Consulting).

The place that says “select a year for period life table: 2016” is referencing the data / time period used for estimation. We are going to use 2016 as this is the latest year offered at this website; however, if you select slightly different years you won’t see much of a change as the mortality rate is not changing much for Americans.

Since Joe Biden is male and will be 78 in November, we are going to obtain data relevant to that age and gender. Below is the data I will be using. In the first column, the first value, which is .047826, is the chance of dying in that age. I will be focused on this column because if we do 1-X on this column we get the survival rate. So, for example, the chance of surviving from age 78 to age 79 is 1-.047826 = 0.952174 = 95.2174% This is similar to the chance of surviving Covid-19 if you are over 70, which is 94.6%.

If we perform the 1-X calculation on all of the relevant ages we obtain the following table shown below. The survival probability is the chance of living to the next age, given that you survived to the given age. So, for example, if Joe Biden lives to age 86 he has an 89.1% chance of making it to age 87.

If we want to calculate the chance of making it several years than we combine the survival rates by multiplying them together. For example, the chance of making it from age 78 through age 79 to age 80 is 95.2% X 94.7% = 90.2%

If we perform this type of mathematics from age to age, we obtain Joe Biden’s chances of making it to a given age. As the table below shows, if Joe Biden serves two terms, his expected chance of surviving to the end is 49.8%. In other words, over a two-term time period, there is a 50.2% chance that Joe Biden’s VP has to be sworn-in. In addition to mortality, there is the chance of becoming disabled, but that is not the purpose of this post.

If you have read this far, you probably are wondering Donald Trump’s chances of making it to the end of four years. Donald Trump is male and turned 74 back in June, so we’ll use male and age 74. As the table below shows, there is a 85.7% chance of survival. In other words, there is 14.3% chance that over the next term, Donald Trump’s VP would need to be sworn- in. (Please note, the purpose of this article is not to compare/contrast Donald Trump’s chance of survival to that of Joe Biden. There are other factors besides their ages that affect their probability of survival. I am not a privy to those factors. This article is to teach the reader the basic concepts of how to calculate the chances of survival.)

Final thoughts/changing topics.

LinkedIn censored my last article about the mortality rate of infectious diseases on the grounds of “misinformation”. They removed it. When I challenged their removal, they gave a “second look” and affirmed their ban of my article. All data in that article was from the reputable sources such as the CDC or the Social Security Administration; I provided no original research. My only contribution was that I simply explained to the reader how to do the math, like I am doing in this article. That article can be seen here.

To LinkedIn, I’d like to coin a phrase by saying the following, “True science never censors, it only counters with the truth.” If you believe something I say is untrue, please counter with something you believe is true.