What Is The Proof That This Covid-19 Thing Really Is A "Crisis," Or That Economic Suppression Is The Solution?
/We are in the midst of an event that is completely unique in the history of our country, and as far as I can tell, in the history of the world: namely, the intentional suppression by governments (in the U.S., both federal and states) of a very large percentage of economic activity, in an effort to control a dangerous disease. There are estimates that the U.S. economy could decline by as much as 38% from its recent peak as a consequence of this great economic suppression. In a matter of just a few weeks, tens of millions of people, many of low to moderate income, have been suddenly thrown out of work; hundreds of thousands of businesses have closed, of which an unknown number may never be able to reopen; and trillions of dollars of value have been lost in the stock market.
Surely this kind of devastating government response would not be undertaken unless this disease represented a true crisis, and unless there was also solid evidence that the economic suppression would quickly end the crisis. But how do you distinguish what constitutes a “crisis” that warrants such a drastic response, versus something that is part of the normal and ongoing pain of human existence? And even if this is a bona fide crisis, why do we think that suppression of economic activity will improve the situation?
In the category of the ongoing pain of human existence, we have the fact that more than 2.8 million people die each year in the U.S. That’s close to 240,000 per month, or 8,000 per day. Sadly, we are all mortal. Also in the category of the ongoing pain of human existence we have the regular annual flu season. Note that I am not saying that Covid-19 is just the flu, or that it is no worse than the flu; rather, I’m just trying to get a benchmark from something that is a major public health issue, but is treated as a regular matter to be dealt with in the ordinary course, rather than as a crisis requiring drastic economic suppression. According to CDC data for the U.S., an estimated 38,230 people died of the flu in the 2016-17 season, 61,099 in the 2017-18 season, and 34,157 in the 2018-19 season. Such numbers are in the range of 1-2.5% of the total annual deaths in the U.S., which is surely significant, but definitely not justifying turning the economy inside out.
So what is the proof that this Covid-19 thing is so different that it really is a “crisis”? With the economy in a 38% downward tailspin, you would think that that proof must be readily available and completely definitive. Wrong. I’ve been looking hard to try to find the proof, and remarkably, I can’t. Let’s consider various points that have been raised.
Without doubt there was one thing that got this started by scaring the bejeezus out of people sufficiently to motivate the economic suppression, and that was models of disease propagation created by self-proclaimed scientific experts, showing vast numbers of imminent deaths. The now infamous kick-off was the study published by Neil Ferguson and others at Imperial College London on March 16, projecting up to 2.2 million deaths in the U.S. From Common Dreams, March 17: “An alarming scientific report compiled by British researchers and shared with the Trump White House warns that, in the absence of drastic and coordinated government action, the novel coronavirus could kill as many as 2.2 million people in the United States alone.” Alarming indeed! But was there any reality to this projection? As I have pointed out many times in the context of models predicting climate doom 50 or 100 years from now, a model is correctly viewed as a scientific hypothesis, that only becomes usable for policy-making as it is validated through accumulation of data. At the time it was published, the Ferguson/Imperial College model had essentially no validating data to back it up. Nevertheless, it marked the start of the shut-down orders that have devastated the economy.
Almost immediately Ferguson backed off his projection. But the new go-to gurus quickly became the IHME (Institute for Health Metrics and Evaluation) at the University of Washington. In March that group was projecting something in the range of 100,000 to 200,000 deaths in the U.S. — a 90+% reduction from the Imperial College/Ferguson figure, but still fairly scary. Then a quick series of reductions followed: down to 93,531 on April 2, and then 81,766 on April 6. By April 8 the projected number was down to about 60,000. Wait a minute, now we’re talking about fewer deaths than the 2017-18 flu season.
The IHME model also covered projections for needed hospital beds. The following chart (compiled by John Nolte at Breitbart) shows the rapid decline in the projections as the month of April has progressed, plotted against actual hospital bed usage in green. It looks like even the April 9 projection, just a week ago, is almost twice too high.
We are regularly bombarded with statistics on death rate (number of deaths divided by number of people infected) and numbers of deaths — but how real are these numbers? On April 14 the number of reported daily deaths in the U.S., after running around 2000 per day for several days, suddenly shot up to over 6000 (worldometers data). Oh, that is the day that New York City decided to add to its totals some 3778 new deaths where “decedent [...] had no known positive laboratory test for SARS-CoV-2.” There is also the question of so-called “co-morbidities” — heart disease, diabetes, obesity; so how many of the people reported to have died of coronavirus would have died at about the same time even without the virus? The incentives for reporting deaths as “caused by” the virus have become so powerful (e.g., federal money, politicians wanting to justify their extreme measures) that it is now highly likely that the death totals are getting substantially inflated. But by how much? It’s impossible to know. At the same time, the total number tested is only now starting to approach 1% of the population (3.25 million tested as of April 16), meaning that the death rate could easily be a tenth or less of what you get by dividing the number of deaths by the number of “cases.” What percent of the general population has had a mild infection without even being aware of it? At this point it’s impossible to know.
The only way to get a real handle on incremental mortality from the virus is to look at so-called “excess deaths,” that is, for example, total number of deaths in the U.S. in March 2020 versus a benchmark of, say, March 2019; or April 2020 versus April 2019. I can’t find those data available yet. Similarly, to get a handle on death rate among the infected, you need to know not just how many sick people have tested positive, but how many infected people there are among the healthy; and therefore you need to do some substantial random testing of the healthy population. As far as I can find, that hasn’t been done yet.
OK then, how about the proof that enforced lockdowns and economic suppression are the things that have kept the disease from being far worse.
Isn’t it just obvious that the rapid declines in projected hospitalizations and deaths must be the result of all the lockdown orders and social distancing and economic suppression? After all, the authorities started to impose the lockdowns at various dates in March, and by the beginning of April, the projections for hospitalizations and deaths were plumeting. Yes, but post hoc does not prove propter hoc. As readers of this blog know, propter hoc (causation) in the scientific method is not demonstrated by correlation, no matter how close, but rather by exclusion of the null hypothesis.
The null hypothesis here is that a jurisdiction would fare just as well in disease prevalence or deaths with no forced economic suppression. To exclude that null hypothesis, we need to look at examples of jurisdictions that have imposed little to no economic lockdown, and see how those jurisdictions compare to the ones with the extreme lockdowns.
Among all the states, I believe that the one at the farthest extreme in terms of imposing no mandatory economic suppression is South Dakota. A long-time reader and fan of this blog has become Senior Advisor and Policy Director to Governor Kristi Noem of that state, and sends me this link to the governor’s pronouncements. The bottom line is that Governor Noem has strongly encouraged “social distancing” and good hygiene, but has imposed no mandatory restrictions on the operation of businesses, and leaves it up to the people individually to determine what is reasonable conduct in the circumstances. To date, South Dakota has had all of 1311 cases of Covid-19, and 7 deaths. Granted, South Dakota is a small population state; New York’s population is about 22 times larger (20 million versus 900,000). But New York has had 16,251 deaths, which is more than 2000 times as many deaths as South Dakota. Another way of expressing the mortality statistic is that New York reports more than 800 deaths per million population, and South Dakota reports about 9 deaths per million population. Could it be that South Dakota is just at an earlier point in the cycle, and cases and deaths are only now starting to surge? Actually, cases have plateaued and are declining in both states.
For a somewhat broader comparison, John Hinderaker at Powerline yesterday collects data for five states in the Upper Midwest: Minnesota, Wisconsin, Iowa, North Dakota and South Dakota. Of these five states, Minnesota and Wisconsin have imposed strong versions of mandatory business restrictions, whereas Iowa and North and South Dakota are at the other end of the spectrum, with South Dakota the least restrictive of all, as already noted. Hinderaker notes that as of April 11 the people at IHME predicted far higher fatality rates for the three less restrictive states, presumably reflecting a significant component in their modeling based on the degree of restrictions. (Of course you will not be surprised to learn that the IHME model is completely opaque and they won’t tell you what factors go into their supposedly expert projections.). Then on April 13 IHME drastically revised down the projected number of deaths at least in North and South Dakota (369 down to 32 for ND, 356 down to 181 for SD), and revised Minnesota notably upward (442 to 656). Hinderaker’s surmise is that the projections for North and South Dakota had become completely ridiculous, given the tiny numbers of deaths to date (9 and 7 respectively). Even with the downward revisions, the IMHE projections for these states still appear absurdly high. In terms of actual deaths to date, and deaths per million, the figures are: Minnesota 94 deaths (16 per million); Wisconsin 182 deaths (30 per million); Iowa 60 deaths (19 per million); North Dakota 9 deaths (12 per million); and South Dakota 7 deaths (9 per million). From those numbers, it certainly appears that the advantage goes to the states with the less restrictive business suppression policies.
Then there’s Sweden, the country in Europe with the fewest mandatory restrictions. Restaurants are still open! It has 1203 deaths to date, which is about 120 per million population. That compares very favorably to most of the major European countries with severe lockdowns (UK, Italy, Spain, France), let alone to New York at about 800 deaths per million.
If lockdowns and business suppression don’t seem to be showing measurable benefits in the fight against the disease, what is going on here? Yesterday a prominent Israeli scientist went public with a new study (in Hebrew) that claims to show that the virus has a life cycle where it plays itself out in any given location over the course of about 70 days.
A prominent Israeli mathematician, analyst, and former general claims simple statistical analysis demonstrates that the spread of COVID-19 peaks after about 40 days and declines to almost zero after 70 days — no matter where it strikes, and no matter what measures governments impose to try to thwart it.
This study is based not on a theoretical model, but rather on analysis of available data to date. The author is Isaac Ben-Israel, identified as head of the Security Studies program in Tel Aviv University and the chairman of Israel’s National Council for Research and Development. The Homeland Security News Wire website quotes Ben-Israel as follows:
While [Ben-Israel] said he supports social distancing, the widespread shuttering of economies worldwide constitutes a demonstrable error in light of those statistics.
Nothing I can find in data available to date shows any superiority of the business-suppression-will-save-us hypothesis over Ben-Israel’s hypothesis.
In a situation where accurate data are desperately needed, we find our public policy in the grip of doctors, supposedly experts, who think they understand what is going on because they have created what they think are sophisticated models. See also, climate change.
