Lower Life Expectancy in a State is Correlated with a Higher Share Voting for Trump

A lower life expectancy in a state is associated with a higher share in the state voting for Trump.  The chart above shows the simple correlation, using state-wide averages, between the life expectancy in a state and Trump’s share of the vote in that state in the 2020 presidential election.  States where life expectancy is relatively low saw, on average, a higher share of their population voting for Trump.  Life expectancy was especially low in a set of mostly Southern states that also had a high share voting for Trump (the bottom right corner of the chart).

The figures on life expectancy come from a recently issued set of estimates produced by the CDC.  The CDC estimates are geographically highly detailed, providing estimates down to the census tract level, but I have only used here the overall state-wide averages.  Due to their fine level of geographic detail, the CDC estimates are averaged over several years (2010 to 2015) to smooth out year-to-year statistical noise.  But life expectancy figures generally change only slowly over time (2020 was an exception, due to Covid-19), so figures for 2010-15 will provide a good estimate of what should be considered normal for life expectancy currently (i.e. with the exception of the Covid-19 impact).  The presidential election results are from Wikipedia, where the Trump share is his share in the overall vote in each state (including third party and other minor candidates).

The correlation is a strong one.  The regression equation (shown in the chart) for the relationship has an R-squared of 0.45.  This means that if one simply knew the life expectancy in a state, one could predict 45% of the variation in the share across the states that would vote for Trump.  This is high for such a simple cross-section relationship.  The negative slope of the equation (-0.11) means that every percentage point increase in the share of the vote for Trump is associated with a 0.11 year lower life expectancy.  Or put another way, a state with a life expectancy that is one year less than in another is associated with an expected 9 percentage point higher share of those voting for Trump (where 9 is roughly equal to 1 / 0.11).

Why this correlation?  Note that it is not saying that a high or low life expectancy in itself would necessarily be driving a tendency to vote for Trump or not.  Rather, a number of factors that enter into the determination of life expectancy are quite possibly also factors in common with the views of Trump supporters.  Life expectancy depends on personal factors and decisions (smoking, diet and exercise, obesity, vaccinations, whether to wear a mask to protect oneself and others to reduce the spread of a deadly disease), as well as on decisions made by state and local governments chosen by that electorate   (such as on access to health care, e.g. whether Medicaid should be available for the poor).  Life expectancy also depends on income levels and for any given average income level on income inequality.

And it will depend on the social norms of the region, such as car driving habits (speeding) and access to guns.  Of the factors reducing life expectancy in the US between 2014 and 2017 (mostly offsetting factors that would have, by themselves, led to a higher life expectancy) unintentional injuries accounted for just over half (50.6%) while suicides and homicides accounted for a further 15% (suicide 7.8% and homicide 7.5%).  That is, these non-medical factors accounted for two-thirds of the factors that had a negative impact on life expectancy in this period.

Few would question that better health is better than poorer health.  The high correlation seen here between life expectancy and the degree of Trump support suggests that there are significant commonalities in the various states between behaviors (both personal and social) that lead to poorer health outcomes and support for Trump.

Death Rates due to Covid-19: An International Comparison

A.  Introduction

In an interview in early August, when over 1,000 Americans were dying each day due to Covid-19, President Trump was asked how he could consider the disease to be then under control.  He responded “They are dying, that’s true”, and then went on to say “it is what it is.  But that doesn’t mean we aren’t doing everything we can.  It’s under control as much as you can control it.”

If it were true that the disease was “under control as much as you can control it”, then deaths in the US would be similar (as a share of population) to what they are in other countries around the world.  It is the same disease everywhere.  And it would especially be true now, more than nine months into this pandemic.  While much was still not known in the early months on how best to bring this terrible disease under control, we now know what has worked in other countries plus we have results from numerous scientific studies.

In particular, it has become clear that a highly effective measure to contain the virus is also the simplest:  Everyone should just wear a mask when out in public.  The experience of East Asian countries, which will be examined below and where mask-wearing was common even before Covid-19, is consistent with this.  There are also now scientific studies backing this up, as discussed in an editorial published on July 14 in JAMA – the Journal of the American Medical Association.  Dr. Robert Redfield, the head of the CDC, was a co-author of that editiorial, and in interviews and press conferences since he has made clear that if everyone simply wore a mask when in public, the disease would be brought under control in as little as four to eight weeks.

Dr. Redfield said the same in testimony to Congress on September 16 (although with a more cautious time scale, allowing between 6 and 12 weeks for the pandemic to be brought under control).  Indeed, Dr. Redfield noted in that testimony that wearing of masks could be more effective than even a vaccine, as any vaccine that is developed will likely have an effectiveness of 70% or less.  A mask, if worn, can do better.

But getting most of the population to wear a mask requires political leadership, and that has been sorely lacking under President Trump.  Indeed, under Trump the wearing of masks has been turned into an issue of political identity, and he has even mocked Joe Biden and Democrats generally for wearing them.  Trump also asserted, on the same day as Dr. Redfield’s congressional testimony, that the doctor was wrong in his medical advice on masks.

The sad result is that death rates from Covid-19 in the US are now not simply higher than in many other countries around the world, but higher by a large multiple.  There is no basis for asserting that this disease is “under control as much as you can control it”.

We will examine here what other countries have been able to achieve in comparison to what the US has, basically through a series of charts.  A word on the data:  The figures were all calculated from the reported deaths by country from Covid-19 downloaded from the site maintained by the Center for Systems Science and Engineering at Johns Hopkins University.  The data were downloaded on the afternoon of September 15, with the country data current through September 14.

B.  US Compared to Canada and Europe

The chart at the top of this post shows the number of deaths from Covid-19 per day per million of population (based on a rolling seven-day average ending on the date shown), from January 29 through to September 14, in the US, Canada, and Western and Eastern Europe (with Eastern Europe covering the Baltics through to Albania).

Starting with the US, deaths rose rapidly in late March and early April, peaked in mid-April, and then fell.  This continued until early July.  But then, as a number of states rushed to re-open their economies in May and especially June (with the strong encouragement of Trump), death rates rose again, doubling from their not-so-low early-July lows.  They then came down modestly in August and the first half of September, but remain far higher than elsewhere.

The profiles in Europe and Canada are different in an important way.  While death rates rose early in Western Europe (and to rates higher than what came later for the US), when much was still not known about the virus and how it was spread, they were then brought down to very low rates – well below those of the US.  And they have remained low (at least so far).  This is in contrast to the US, where death rates rose in July as lessons on how to manage the virus were ignored.

Canada followed a similar profile to that of Western Europe, although with an initial peak that came later (and with a substantially lower peak – only half that of Western Europe), with then a decline to low levels that have remained low.  In Eastern Europe, early rates in the spring never rose that high, but then still came down by June.  Since then they have risen some, but to rates that remain well below those of the US (at less than a third of the US rate, as of mid-September).

Breaking this down for some of the major countries of Western Europe:

Rates peaked early and at high levels in Italy, France, and the UK, but then all came down and remained down.  The peak in Germany came at roughly the same time as that of the US (but at well less than half the US rate), and then came down to an extremely low level.  As of mid-September, the death rate in Germany is only 2% of the US rate.  If it’s “under control as much as you can control it” in the US, as Trump asserted, why is it that the death rate, per million of population, can be 98% less in Germany?

There are two special cases in Western Europe that are worth examining – Spain and Sweden:

Rates rose rapidly and to quite high levels in Spain early in the crisis.  Its hospital system was overwhelmed and many died.  But then Spain brought down the rates to very low levels by June and July.  They have, however, trended up since mid-August, as it appears Spain opened up its seasonal tourism industry too rapidly (tourism as a share of GDP is far higher in Spain than in any other OECD member country).  But even with the recent increase, the number of deaths per million in Spain remains less than half (45%) of what the rate is in the US as of mid-September.

(One might also note the negative numbers recorded for the number of deaths in Spain due to Covid-19 for a period in late May, as well as an odd spike up in late June.  The reason for this is that Spain revised its counts of the number who had died from Covid-19 as they later reviewed what had been submitted during the peak of the crisis.  A focus on the statistics was not the highest priority earlier – saving lives was.  It is of course impossible for there to be a negative number of deaths.  But figures are recorded each day for the cumulative number of deaths due to Covid-19, and when that total was revised down on May 25, the daily change in the total (which is the basis for the daily death count) will be negative (and will be negative for a week, as the numbers are seven-day averages).  And a later upward revision in late June will look like a spike up.)

Sweden is also an interesting case as, early in the crisis, it deliberately decided not to mandate closures of restaurants, offices, and other non-essential work locations, but rather left this to be decided by each entity.  But the policy failed:  Deaths from Covid-19 rose to rates well above US levels (and was especially far above the rates of its Nordic neighbors of Norway, Finland, and Denmark, although below the peak levels seen in Italy, Spain, France, and the UK).  The rates then fell relatively slowly in Sweden.  They eventually moved to policies more in line with the rest of Europe, and eventually saw similarly low rates.

D.  US Compared to East Asia, Australia, and New Zealand

As an earlier post on this blog on the number of Covid-19 cases discussed, the countries of East Asia, as well as Australia and New Zealand, show what is possible if serious measures are taken to control the spread of the virus (and possible in a region with more travel and business exposure to China than any other region).  The measures required are not exotic.  Nor did they require resources that others did not have.  All that was required were the standard public health measures used to control the spread of any infectious disease – extensive testing with follow-up tracing of contacts and quarantining of those exposed, plus the normal and widespread use of simple masks.  With such measures, Taiwan was able, for example, to keep open its schools basically throughout (in February it extended its regular Chinese New Year holiday by an extra two weeks, but has since followed its regular schedule).

The result was few cases of Covid-19, and few deaths:

 

The rates for all the countries listed on the chart were plotted.  But they were all so close to zero that, other than for the few names shown, one could not distinguish one from the other.

There was an increase in the rates since mid-July in Australia, and to a lesser extent in Hong Kong (and a far lesser extent in Japan), as some of the earlier controls were eased.  But these have all now been brought back under control.  And even with these outbreaks, the rates never approached the US rates.

E.  Who are the Comparables for the US?

Who, then, might have a record comparable to that of the US?  Among the larger countries:

Donald Trump can be proud to say that death rates in the US have, since June, been lower than the rates in Mexico and Brazil.  The US has not performed as poorly as they have.  The pattern in South Africa is somewhat odd in that its rates were higher than those of the US between mid-July and mid-August, but are now substantially less.  And Russia as well as India have had lower rates throughout.

All this assumes the tracking statistics on deaths from Covid-19 are accurate, and one might question this for some of these countries.  As was discussed above for the case of Spain, such numbers can be difficult to assemble even with resources that the countries here do not have.  But for the ranges in the numbers seen here, the conclusions would still hold even if the rates were substantially higher.  As of mid-September, the South African rate would have needed to have been twice as high, and the Indian and Russian rates three times as high, to reach the US rate.

Note that I have not included China.  If it were added, it would show extremely low death rates per million throughout, with a peak of just 0.1 in mid-February.  But while the deaths from Covid-19 may well have been low compared to others (particularly when expressed per million, given its population), I am not confident they were in fact that low.  Restrictions on the news media and what they can report do not engender confidence.

But overall, to find countries with records on management of Covid-19 comparable to what they have been in the US, one needs to look at countries with per capita incomes that are far below that of the US.  The US has thought of itself as belonging in the top rank of countries.  But for this, the only countries with comparable death rates from Covid-19 are countries that, before Trump, the US had not normally been grouped with.

F.  What Deaths in the US Would Have Been at the Rates Other Countries Have Been Able to Achieve

As noted at the top of this post, President Trump claimed that the disease is “under control as much as you can control it.”  But as we have seen, it is not.  Other countries, facing the same disease, have been able to manage it with far lower death rates than the US has had.  How much of a difference would this have made?

Little was known about the disease early in the crisis, and one can argue that countries were searching then for what best to do.  And after the high early peaks, the rates did come down in the US as well as in Europe and Canada.  But then the US reversed course while rates continued to fall elsewhere.  It is thus this more recent period that most clearly shows the consequences of the choices the US made compared to others.  For the purposes of this exercise, we will therefore look at the period since August 1.

From August 1 to September 14, a period of 45 days, US deaths totaled 40,459.  This is a bit over a fifth (21%) of the total US deaths as of September 14 of 194,493.  It is still a substantial figure:   The number of US soldiers who died in battle in the Korean War totaled 33,739, and the number who died in the Vietnam War totaled 47,434.  But based on the numbers of deaths per million in other countries and regions, how many would have died for a population equal to that of the US?:

If the US had had the number of deaths per million that Romania had over this same period, then 31,700 would have died, or about three-quarters of the number of Americans who died.  If the US had the rate of Albania, about 20,800 would have died, or about half the number of Americans who died.  One might ask that if “it is what it is”, and that “It’s under control as much as you can control it”, why is it that Romania could control it so that there would only be three-quarters as many deaths, and Albania could control it so that there would only be half as many deaths?  Neither Romania nor Albania has the resources the US has, plus they are small and open.

Other cases are more extreme.  If the US had the rate over this period of the EU as a whole, there would have been 5,465 deaths.  Instead, it was 7.4 times higher.  At the rate of Canada, there would have been 2,184 deaths.  Instead, it was 18.5 times higher.  And Singapore and Taiwan both had zero deaths over this period.  The most recent death (as of this writing) was on July 14 in Singapore and on May 11 in Taiwan.  If the US had their rates, there would have been no deaths.

There is of course a wide range here.  Plus things may change.  Infection rates have been rising in Europe in recent days, and increases in death rates may soon follow.  The US has also today (on September 22, as I write this) passed a significant milestone:  More than 200,000 have now died in the US from this disease.  And there are widespread concerns that rates will increase this fall and winter across the Northern Hemisphere in a “second wave”, as more people remain inside and as they become less vigilant as time goes on. One has seen this with prior infectious diseases, particularly those that spread through the air.  There is also increasing pressure to reopen schools for in-class teaching and to fully reopen businesses.

So there is uncertainty on how this will progress.  But based on what we know for the last month and a half, a question to address is why the Trump administration has not been able to do as good a job of reducing deaths from this virus as have the governments of Romania, Albania, Bulgaria, Russia, Spain, Australia, Croatia, Serbia, Luxembourg, Portugal, Poland, France, Greece, Hong Kong, Italy, Sweden, Czechia, Slovenia, the Netherlands, Belgium, the United Kingdom, Canada, Switzerland, Hungary, Austria, Ireland, Japan, Denmark, Lithuania, Germany, Norway, Slovakia, Latvia, Finland, South Korea, Estonia, New Zealand, Singapore, and Taiwan.

The Spread of Covid-19: Trump States vs. Clinton States – An Update

An earlier post on this blog compared the spread of Covid-19 in the states that Trump had won in 2016 to that in the states won by Clinton, with data through June 24.  This post will update those figures to July 16.  The trends have become even clearer.

As seen in the chart above, new cases in the states won by Trump have continued to shoot upwards, at an alarming pace.  They had reached 22,000 new cases per day as of June 24 (based on a seven-day rolling average ending on that date), but have now (as of July 16, just three weeks later) more than doubled to 48,500.  The decisions to rapidly reopen by the governors of such Trump-won states as Florida, Georgia, Texas, and Arizona, as well as others, have clearly been a disaster.  The virus is now spreading rapidly in those states, and some of these governors are now putting back in place (albeit only partially) the social distancing measures that had earlier worked.

Daily new cases are also now clearly increasing in the states won by Clinton.  This trend was still too recent to be clear in the data through June 24.  But the pace of spread in the Clinton states is far below that of the Trump states, and the number of new daily cases in the Clinton states (16,500 as of July 16) is only one-third the number in the Trump states.

The trends in the figures for the number of deaths from Covid-19 have also now become clear:

In the previous data through June 24, the daily number of deaths (again based on seven-day rolling averages) had come down from their mid-April peaks to a relatively flat level as of mid-June.  This had marked a sharp decline of over 80% in the daily number of deaths in the Clinton states (where peaks early in the crisis in New York had overwhelmed the hospital system, at a time when still little was known on how best to treat the extremely sick), and by a lesser but still significant decline (about 50%) in the Trump states.

Since mid-June, the daily number of deaths in the Clinton states has been relatively flat (hovering between about 200 and 300).  But there has now been a significant increase in deaths in the Trump states, rising from a trough of about 280 per day to now almost 500, an increase of about 75%.  And the path points to a continued rise, as one would expect given the even sharper rise in daily new cases (as there is a lag – deaths occur several weeks after when a case is first confirmed).

These trends should be worrisome in the extreme.  They are not the consequence of increased testing in the US, as Trump has repeatedly asserted.  While testing was slow to start in the US (the administration had bungled the roll-out in February and into much of March), there has not been a significant change in test availability since mid or late April, and certainly since May.  The increases in cases started in June.  More people are now being tested because more people are getting sick, and seek a test as they come down with the symptoms.  And the increase in the number of people dying from the disease is certainly not a consequence of testing, but rather of more people becoming sick.

More could be done, but sadly this presidential administration isn’t.  And it would not be all that difficult.  As I had noted in my June 25 post, a relatively easy measure would be for everyone to wear masks.  Since that post, Robert Redfield (the head of the CDC) noted in an interview on July 14 that “if we could get everyone to wear a mask right now, I really do think that over the next four-six-eight weeks we could bring this epidemic under control” (see this YouTube video of the interview, starting at about the 4-minute mark).  He noted that this is not difficult – the problem is just that not enough people do it.

For many of those refusing to wear a mask – some adamantly so – the issue is seen as political.  The problem started with Trump, where at the April 3 press conference announcing CDC guidelines calling on people to wear face masks, Trump simultaneously emphasized that he would not himself abide by those guidelines.  With any other president, this would be unbelievable.  Since then, supporters of Trump have increasingly seen the issue as one of making a political statement rather than as the public health matter that it is.  A recent academic study found that political partisanship is the most important factor in explaining whether or not people will wear masks and exercise other social distancing recommendations, and that this partisan difference has grown over time.

This has even become violent.  In early May, for example, a security guard at a Famlly Dollar store told a customer she would need to wear a face mask to enter, as per the state orders of the time.  She returned with members of her family about 20 minutes later who shot the guard, who died.  More recently, a 43-year old man entering a convenience store without a mask was asked by another customer to put on a mask.  He responded by stabbing the 77-year old customer.  The man then fled, was later spotted by police, and started to attack the policewoman who then shot him.  He died.  And there have been, sadly, a number of such incidents.

Those refusing to wear face masks when in public insist that such a requirement infringes on their “freedom”.  Thus, as a matter of principle, they refuse to do it.  If it was indeed the case that the only one suffering harm from not wearing a mask was that individual only, I would not be so concerned.  But that is not the case – others exposed may then become infected, and possibly even die.  It is similar to speed limits on highways.  If the only one who might be harmed by speeding is the speeder only, I would not be so concerned.  But speeders may harm, and possibly kill, others as well.  Hence we have speed limits and those limits are enforced.

Refusing to wear a face mask under a belief that it is an infringement on freedom, and responding with threats or even violence when asked to do so, is madness.  With true leadership in Washington we would have a president who would act on this.  Not only would that president model responsible behavior by wearing a face mask himself when in public or when meeting with others, he would also call on all his supporters to do so as well.  They might listen to him.  But his refusal to do so speaks volumes itself.