The relation between death and the pandemic economy is a fraught one that has become hotly debated, although with not much clear empirical evidence. I note that recently over on Econbrowser Menzie Chinn has had a series of posts on this matter in various forms. Obviously a big issue has been the claim by the anti-lockdown crowd that not reopening the economy quickly will lead to an increase in suicides by the increasingly large numbers of unemployed people out there. There certainly have been many studies in the past showing a variety of bad social outcomes from high unemployment, including suicides, domestic abuse, drug abuse, depression, and more. There does seem to be some strong evidence of several of these notably higher domestic abuse and depression.
When it comes to suicide and death more broadly, the empirical picture is very murky. Menzie in one of his recent posts reported on a regression he ran covering monthly data from 1998 to very recently that used dummies for months and then unemployment rates and suicides (in the US) and found the an unexpected “wrong sign” with lower suicides correlated with higher unemployment, although this was not a statistically significant result. He provides no explanation for why this odd result seems to be there, but it does show that this is not a simple matter.
Regarding current data on the main question, so far there does not seem to be any data showing a noticeable rise in suicides in the US since the pandemic, with only reports of some increases among medical personnel, who have suffered from overwork, stress, and even guilt, along with fear. That we might be seeing that out of them is completely understandable.
So why might we not be seeing much increase in suicides so far despite all the things going on such as increased depression as well as unemployment and more that would suggest we might expect to see it? Some have suggested a “wartime” effect: people are suffering, but they know others are as well and so rally around the flag to hang in there. This rally around the flag effect even worked for awhile to boost Trump’s polls for a few weeks in late March and early April until people saw how we was botching things, and now his polls are lower than they were before, even as those of some generally unpopular leaders in other hard hit nations like Italy, France, and Spain have seen their poll numbers continue to be noticeably higher than they were previously.
Another element, suggested to me by my medically connected daughters, not all that different from the above, is that people who are depressed feel “validated” because now others appreciate their condition. This is especially relevant for veterans suffering from PTSD and so on.
Before proceeding further, let me note that there are some factors that may lower the death rate during a recession that can offset to some degree, with the importance of these matters of ongoing debate. Probably the most important is the reduction of pollution, which is estimated to kill 200,000 to 300,000 people per year in the US. Obviously pollution reductions now in the US are not going to offset all of that or even most of that, but presumably it does some. An estimate in more heavily polluted China has claimed that the reduction of pollution there due to the pandemic might have saved up to 50,000 to 100,000 lives. That is clearly a large number, and way exceeds the number of people in China who died of the coronavirus.
I note on this my earlier post that apparently at the global level carbon emissions have been estimated to have declined by 17% to a bottom point near April 7, with them rising gradually since with a likely gradual recovery of the world economy (with the US lagging on both the virus and the economic recovery). Probably carbon emissions are highly correlated with other forms of pollution, so this looks like a pretty good indicator of timing on that one, at least globally.
Two other items that have been brought up as reducing deaths during a lockdown/recession are auto deaths and work-related deaths. It is probably true that we have seen some reductions of those in the US in the last few months, but I have yet to see any sort of reliable data on them. Both of those have been trending downwards in recent years gradually in the US, with auto deaths in 2019 a bit over 36,000 with work-related deaths at about 5,100 (construction is the largest with over 1000). Suicides last year were at about 48,000 in the US, so larger than those two put together, but as of now we do not know how any of these have changed recently.
I would add that on this matter of auto versus suicide deaths, the US picture is quite different from the global picture, although there is reason to be skeptical about some of the international data on suicides, which get covered up in many nations (but also in the US to some extent for insurance reasons). Anyway, as of last year global auto deaths were estimated to be about 1.35 million while suicides were at a much lower 800,000. Almost certainly why we have more suicides than auto deaths in the US unlike the rest of the world is that we have way more guns per capita than any other nation, and it is well established in the US that across states there is a strong correlation between guns per capita and suicide rates.
On a speculative note, given that pretty much the entire US is now engaging in varying degrees of reopening despite very few states having actually met the CDC guidelines for doing so, that we may see a perverse effect of an increase in suicides in the near future, even as the economy expands, which it certainly has been now for some time, maybe even as long as a month. For one thing, unemployment continues to rise, despite the economy growing (and the stock market rising especially strongly). But another effect may be the ending of that “wartime” effect. People who are depressed and unemployed may lose that feeling of solidarity and as they fail to get jobs and may become homeless due to failures to pay rent, and so on, they may become more seriously depressed and feeling isolated. I can well imagine such an effect leading to more suicides, although I do not know.
I will make a completely anecdotal observation that somehow this week I have observed several people both that I am in close contact with as well as some more distant, who seem to have sort of been freaking out this week. This is probably just an odd bad luck of the draw, but I am a bit worried that it may reflect my speculation in the previous paragraph, and that whether or not we see more suicides, we may see an increase in the near term of people getting less happy and more depressed. if that is happening, let us hope that it ends soon. This pandemic has been bad enough.