“Never before in human history have entire healthy populations been put under effective house arrest and told when they can go out; where to; for how long; who and how many people they can meet; which businesses can stay open to sell designated goods and provide listed services. Yet after a year of this extreme experiment, data from around the world show that the spread of the pandemic correlates more with geography, demography and seasonality than lockdown stringency and sequencing.”
The West should envy Japan’s COVID-19 response¹

As someone who served 8 years in the United States Marine Corps, I have a deep appreciation for the structure and order of a military lifestyle. Of all the environments I’ve experienced, nothing has provided more rigor and discipline than that first year of service when I went through USMC Recruit Training, Infantry Training Battalion, and the Basic Reconnaissance Course.

Flash forward to 2020, SARS-CoV-2, and all of the non-pharmaceutical interventions (NPIs) that came with it such as lockdowns, mask wearing, and social distancing. It seems reasonable to assume that a military environment is the best way to evaluate the efficacy of such interventions, right? I mean, if an NPI doesn’t work in a military setting, where incredibly strict adherence to protocols can be achieved, then surely it won’t work anywhere? Let’s investigate.

Navy

Let’s start with the first military outbreak that we all watched unfold in March & April of 2020: the USS Theodore Roosevelt aircraft carrier. This ship was deployed to the South Pacific and, following a docking in Vietnam, experienced an outbreak that eventually infected 24% (1,156/~4800) of the crew before the Navy stopped reporting new cases².

Now, it’s interesting enough that a military setting as strict as a sea-bound aircraft carrier would spiral from 1 index case to 1,156 despite orders to mitigate the spread, but perhaps we can forgive the USS Theodore Roosevelt because, after all, it is a ship with tight quarters and poor air flow.

However, there is one aspect of this that is impossible to forget: the U.S. Navy discovered that quarantine does not work. They noticed that sailors were testing positive after receiving a negative test followed by 14 days of quarantine³. In case you need to read that again, here it is: Sailors were testing negative prior to quarantine, completing 14 days of quarantine, and then testing positive on their out-test.

If that sequence of events does not make you lean back a little and ask yourself, “do I really know how this virus works?”, then you are not being intellectually honest. What’s more disturbing is that the Navy declined to make this information public while the nation was mired in draconian lockdowns. Anchors Aweigh!

Army

Let’s skip ahead a few weeks to May and move from the South Pacific all the way to the Peach State. The U.S. Army conducts basic training at Fort Benning in Georgia. During the spring, a batch of 640 new recruits showed up and they all were tested upon arrival. Four recruits tested positive and were isolated, while the rest began a 14-day quarantine period. At the end of the 14 days, recruit training began with strict mitigation measures in place such as social distancing and mask wearing. Within a few days, one recruit became ill, prompting the command element to re-test all 640 recruits. 142 came back positive⁴.

Immediately, a couple interesting points come to mind:

(1) The 22% infection rate at Fort Benning is eerily similar to the 24% infection rate on the USS Theodore Roosevelt. Given that they are incredibly different environments with different mitigation tactics, we might expect them to fizzle out at different infection rates.

(2) When 142 recruits can be infected and only 1 complains of illness, it seems reasonable to wonder if the virus might not be the scourge that it was made out to be.

Above all else, we once again have to grapple with the fact that the cohort tested negative, endured 14 days of quarantine, and subsequently developed an outbreak that infected over 20%. It may seem reasonable to hand-wave this away when it happens on an aircraft carrier, because of the tight quarters and poor ventilation. But when it happens again, on a recruit training base (literally the most disciplined environment on the planet), it’s time for reflection. If these interventions fail in a military setting, why on Earth would we expect them not to fail in a community setting?

By the way, Fort Benning is not the only Army base to experience this sequence of quarantine failure⁵. The confidential nature of military operations means that we may never know just how common this type of military outbreak is.

Marine Corps

With a motto like Semper Fidelis (“Always Faithful”), surely the USMC will achieve perfect compliance with viral suppression, or at least as close to it as humanly possible.

In a study published by the New England Journal of Medicine, the Marines lived up to their motto⁶. Upon arrival at bootcamp, recruits were tested, and about 1% of them were positive. The other 99% were put into a 14-day quarantine and then tested again. This time around, about 2% (of the recruits who tested negative on Day 1) tested positive following quarantine. This is a much lower infection rate than the USS Theodore Roosevelt and Fort Benning, so the Marines are looking pretty good by comparison. Semper Fi!

The problem is that, once again, this was a recruit training environment with strict quarantine preceding the positive tests. How did 2% of the recruits become infected? What does this mean for public policy in the general population?

16 out of 1,848 recruits tested positive upon arrival at the depot. That seems reasonable, right? I remember living it up before shipping out to bootcamp–of course some of these kids did, too. The fact that only 16 were positive is pretty impressive. But more than twice as many (35) tested positive by the end of the 14-day enforced quarantine! How do we explain that more than twice as many recruits tested positive following strict quarantine than tested positive upon showing up? Did quarantine somehow make it worse?

If an environment run by Marine Corps Drill Sergeants can’t even stop the virus, then maybe we need to rethink the strategy of shutting down businesses, schools, sports, arts, etc.

The best part of the NEJM study is that it says the quiet part out loud:

“Recruits were under the constant supervision of Marine Corps instructors. Other settings in which young adults congregate are unlikely to reflect similar adherence to measures intended to reduce transmission.”

Another interesting fact is that not a single one of the positive PCR tests was a result of symptom screening–every single one was discovered during scheduled testing. So, once again, recruits didn’t even know they were infected! So much for this virus being worthy of cultural seppuku.

Can We Do Any Better?

Okay, so the military environment isn’t exactly….bulletproof. Maybe we can achieve better compliance somewhere else ( :::eye roll::: ). Let’s take a good, hard look at what total isolation on an uninhabited exploration base looks like.

I’d like to take you back in time, to a continent far, far away. Let’s travel to Antarctica in 1973. During that year, an expedition of 12 men wintered at a remote Antarctic base. After 17 weeks of complete isolation, half of them developed a cold⁷.

Stop and think about that for a second. In the most remote place on planet Earth, after 4 months of total isolation, a virus still found its way. What does this say about our understanding of viral spread? While we don’t know exactly which virus caused this outbreak, we do know that the common cold is caused by respiratory viral infection (rhinovirus, adenovirus, coronavirus, etc.).

If viral outbreak can occur after 4 months of complete isolation in Antarctica, then what the hell is “Two Weeks to Flatten the Curve” supposed to accomplish? Is this a joke?

We’ve documented similar experiences with SARS-CoV-2.

In July 2020, it was discovered that 57 out of 61 crew members of an Argentinian fishing vessel became infected after a month at sea⁸. And just like the Navy, Army, and Marine Corps, they all underwent 2 weeks of quarantine and tested negative prior to embarking.

Negative PCR test + 14-day quarantine + 30 days isolated at sea = outbreak??

I don’t know about you, but when I learn about stories like these, I am inclined to wonder what is going through Governor Tom Wolf’s head when he decides that closing restaurants for a couple weeks is a good idea. I’m even more inclined to worry about the mental well-being of any person who would support such closures.

Pre-2020 Attitudes Towards Lockdowns

It’s a well-known secret that the concept of lockdowns was evaluated and discarded long before SARS-CoV-2 came along. You can easily find dozens of papers from 2019 and prior which all, without fail, come to the same conclusions:

“There are no historical observations or scientific studies that support the confinement by quarantine of groups of possibly infected people for extended periods in order to slow the spread of influenza. … The negative consequences of large-scale quarantine are so extreme (…) that this mitigation measure should be eliminated from serious consideration.”
Disease mitigation measures in the control of pandemic influenza (2006)⁹

“For pathogens that inflict greater morbidity at older ages, interventions that reduce but do not eliminate exposure can paradoxically increase the number of cases of severe disease by shifting the burden of infection toward older individuals”
Too Little of a Good Thing A Paradox of Moderate Infection Control (2008)¹⁰

My personal favorite is a 2008 study on the efficacy of social distancing in a military recruit training environment. Guess what? It didn’t work:

“In summary, we found no statistically significant effect of a social distancing procedure in which some units were closed to an influx of potentially infectious individuals.”
Exploration of the Effectiveness of Social Distancing on Respiratory Pathogen Transmission Implicates Environmental Contributions (2008)¹¹

While this study focused on a specific adenovirus, its conclusions have broader ramifications for the way we understand transmission of respiratory viruses. If follow-up research had been conducted during the intervening years, it’s possible that some poor chap on the USS Theodore Roosevelt wouldn’t have gotten chewed out quite as bad when a virus inevitably spread. It wasn’t his fault. I guess hindsight is 2020.

The 2020 Lockdown Experiment

At this point I hope you are pretty convinced that lockdowns are, at best, grasping at straws. While it sounds reasonable to believe that separating people should work to mitigate viral spread, it doesn’t really work out that way in the real world. Maybe the virus incubates longer than we think. Maybe people cheat their quarantine. Maybe viral spread is more like aerial pollen, and hiding from it for two weeks doesn’t remove it from the environment. Maybe, maybe, maybe. Nobody knows what’s really going on here, but one thing we do know is that lockdowns just don’t seem to stop the spread of SARS-CoV-2.

Take, for example, state-to-state comparisons within the USA.

How is it possible that locked-down California did so much worse this winter than wide-open Florida? Why are my chances of infection lower at open Disneyworld than at the gates of closed Disneyland? Do lockdowns do what we think they do?

California (red) vs. Florida (yellow) — Daily Cases per 100,000

Keep in mind that California was the first state in the nation to lockdown in March. They literally have been under thumb-and-key this entire time–almost a full year now–and what do they have to show for it? A lower trajectory in the summer surge? Big whoop. You see how the two curves look very different, but the total area under the curves is about the same? That’s because California actually did do better in the summer–but Mother Nature always keeps the receipts. And when She has a debt that is due, She will collect. You can’t hide forever.

(Note that the similarity in total area under the curve implies an interplay between seasonality and herd immunity that is worth chewing on)

How about North Dakota and South Dakota? These two were like peas and carrots until the going got tough, when ND Governor Burgum infamously decided to mask up and restrict businesses just as ND cases were beginning to decline. Meanwhile, Governor Noem down in SD stuck to her guns and made no such interventions. Looking at their charts below, can you tell what, exactly, the North Dakota restrictions achieved? I can’t.

South Dakota (red) vs. North Dakota (yellow) — Daily Cases per 100,000

Speaking of ND and SD, let’s look at their Midwest neighbors and see if we can discern any impact of restrictions.

Wyoming (red), Iowa (black), Minnesota (purple), Nebraska (blue), Montana (green), Wisconsin (yellow) — Daily Cases per 100,000

These Midwest states all have dramatically different approaches and policies. Yet they follow a stunningly uniform curve. It’s almost as if their disparate restrictions had no discernible impact on the trajectory of SARS-CoV-2. Hmmmmm… :::twirls mustache:::

Once I start showing multi-state charts, I always have to throw in the Northeast. Once again, we see a high degree of uniformity, even to the point of picking up on the “Christmas Surge” or the “New Variant” or whatever the heck it was that caused a brief blip to start the New Year.

Pennsylvania (yellow), Delaware (brown), Massachusetts (pink), New Jersey (blue), New York (green), Rhode Island (red) — Daily Cases per 100,000

What I find interesting about this particular Northeast curve is the fact that Rhode Island is such an outlier. When evaluating the efficacy of lockdowns, the hypothesis is that a country/state/city that implements lockdowns should see cases go down relative to places that do not implement lockdowns. Alternatively, when a place has higher cases, are we to assume that they did something wrong? Is there some mystical restriction that Rhode Island is flouting, but the other Northeast states are observing? Well, I checked, and there really isn’t. If anything, Rhode Island has been restricting harder. Hmmmmm… :::twirls mustache:::

No Free Lunch

If I were to try to write my own section on the excessive and disproportionate costs of lockdowns, I’m not sure where I’d even start. Fortunately, I don’t have to. The folks over at Rational Ground have compiled a pretty succinct list of all the irreparable harms being caused by various forms of lockdowns. I encourage you to check it out here: Lockdowns: Pros and Cons¹².

The main takeaway is that even the best Science^TM that we modest humans can muster is completely incapable of assessing complicated tradeoffs such as the cost/benefit ratio of lockdowns. After all, it’s not a scientific question at all, but rather a philosophical and socio-political one. That being said, it would be nice if The Science^TM could at least demonstrate that lockdown actually, you know,…works. Speaking of which, let’s address the singular instance in the whole world where maybe it did…

Elephant in the Room

No discussion on lockdown would be complete and honest without addressing Australia and New Zealand, who appear to have achieved good results with lockdowns. New Zealand is to Team Apocalypse what Sweden is to Team Reality: a rallying cry that instills certainty where there should be supposition.

By all accounts, Australia and New Zealand have succeeded where the rest of the world has failed miserably. They have kept SARS-CoV-2 at bay using lockdown-based interventions. What I mean by this is that lockdowns are their primary tool. They do not use masks (outside of specific outbreaks or in Victoria). They do not restrict businesses. They do not cut off normal life, as long as SARS-CoV-2 is away.

However, once it shows up, they go into full-blown Threatcon Delta. This happens with some regularity, unfortunately, including just this week¹³. Remember how I said that Mother Nature always keeps the receipts? Recurring lockdown is a price they have to pay for their success (but at least, in their case, it does appear to work).

So, what’s the deal? Do Australia and New Zealand prove that lockdowns work? Or, at least, do they prove that lockdowns can work?

Before trying to answer that, let’s address the simple fact that these are island nations in the South Pacific, which makes border enforcement pretty easy. You can’t get there without a plane or boat. This provides them with a massive advantage in that they can very easily enforce border control, which they do. I mean, hey, they might not be interconnected with the global economy, but who cares? There is a cold virus to fight.

“When all you care about is here, this is a good place to be.”
Emile de Becque – South Pacific (1958)

Aside from the fact that Australia and New Zealand have effectively removed themselves from the global stage, there is also the fact that they didn’t just lockdown; they locked down early, which is actually quite unique. My belief is that the efficacy of any lockdown is negatively correlated with the real case-rate at the time of commencement¹⁴. What I mean by that is that if SARS-CoV-2 is already endemic in a population, then lockdowns appear to achieve absolutely nothing. If the virus is rare, then perhaps they can achieve something.

If my belief is correct, then the implications are worth contemplating…

Scenario 1

In a scenario where SARS-CoV-2 has already entered the game, lockdowns conceivably would have good public support because, well, the virus is prevalent and people don’t really like that.

Scenario 2

In a scenario where SARS-CoV-2 is rare (say 1 case per 100,000 residents), nobody is really thinking about the virus, so public support for lockdown would be harder to come by.

The irony and tragedy is that lockdowns won’t really work when there IS public support for them, because the virus is too prevalent, but they could work when there ISN’T.

Barring any explosion of SARS-CoV-2 during their winter months (June – August), Australia and New Zealand have achieved success with the Zero-COVID strategy. That’s worth celebrating, but only with the recognition that these are isolated islands in the largest ocean on the planet and they implemented their lockdown strategies early and often. If it works for them, and they choose that path, then great! But let’s not delude ourselves into expecting such policy to be effective in the USA, where even the most severely locked down states have been unable to achieve comparable success (including the Pacific island of Hawaii).

The Literature

So with all of that being said, what does The Science^TM say about lockdowns? Is the evidence in support of lockdowns as ubiquitous as the “experts” would have you believe? Turns out, the answer is no. There is a growing chorus of literature that confirms what we we learned in Antarctica, 1973: isolation just doesn’t work as well as we’d like.

“It turns out that a similar pattern – rapid increase in infections that reaches a peak in the sixth week and declines from the eighth week – is common to all countries in which the disease was discovered, regardless of their response policies: some imposed a severe and immediate lockdown that included not only “social distancing” and banning crowding, but also shutout of economy (like Israel); some “ignored” the infection and continued almost a normal life (such as Taiwan, Korea or Sweden), and some initially adopted a lenient policy but soon reversed to a complete lockdown (such as Italy or the State of New York). Nonetheless, the data shows similar time constants amongst all these countries in regard to the initial rapid growth and the decline of the disease.”
The end of exponential growth: The decline in the spread of coronavirus¹⁵

“Government actions such as border closures, full lockdowns, and a high rate of COVID-19 testing were not associated with statistically significant reductions in the number of critical cases or overall mortality”
A country level analysis measuring the impact of government actions, country preparedness and socioeconomic factors on COVID-19 mortality and related health outcomes¹⁶

“While small benefits cannot be excluded, we do not find significant benefits on case growth of more restrictive NPIs. Similar reductions in case growth may be achievable with less‐restrictive interventions. …

There is no evidence that more restrictive nonpharmaceutical interventions (“lockdowns”) contributed substantially to bending the curve of new cases in England, France, Germany, Iran, Italy, the Netherlands, Spain, or the United States in early 2020.”
Assessing mandatory stay- at- home and business closure effects on the spread of COVID- 19¹⁷

“Inferences on effects of NPIs are non-robust and highly sensitive to model specification. Claimed benefits of lockdown appear grossly exaggerated.”
Effects of non-pharmaceutical interventions on COVID-19: A Tale of Three Models¹⁸

“Official data from Germany’s RKI agency suggest strongly that the spread of the coronavirus in Germany receded autonomously, before any interventions become effective”
Was Germany’s Corona Lockdown Necessary?¹⁹

“The currently most reliable data strongly suggest that the decline in infections in England and Wales began before full lockdown, and that community infections, unlike deaths, were probably at a low level well before lockdown was eased. Furthermore, such a scenario would be consistent with the infection profile in Sweden, which began its decline in fatal infections shortly after the UK, but did so on the basis of measures well short of full lockdown.”
Did COVID-19 infections decline before UK lockdown?²⁰

“The UK lockdown was both superfluous (it did not prevent an otherwise explosive behavior of the spread of the coronavirus) and ineffective (it did not slow down the death growth rate visibly).”
Comment on Flaxman et al. (2020, Nature, https-//doi.org/10.1038/s41586-020-2405-7)- The illusory effects of non-pharmaceutical interventions on COVID-19 in Europe²¹

“Comparison of the epidemic’s evolution between the fully locked down countries and neighboring countries applying social distancing measures only, confirms the absence of any effects of home containment. … This work thus suggests that social distancing measures, such as those applied in the Netherlands and Germany, or in Italy, France, Spain, and United Kingdom before the full lockdown strategies, have approximately the same effects as police-enforced home containment policies.”
Full lockdown policies in Western Europe countries have no evident impacts on the COVID-19 epidemic²²

“Strategies that minimise deaths involve the infected fraction primarily being in the low risk younger age groups—for example, focusing stricter social distancing measures on care homes where people are likely to die rather than schools where they are not. … results presented in the report suggested that the addition of interventions restricting younger people might actually increase the total number of deaths from covid-19”
Effect of school closures on mortality from coronavirus disease 2019: old and new predictions²³

“Current policy can be misdirected and can therefore have long and even short-term negative effects on human welfare and thus result in not actually minimizing death rates (incorporating externalities), especially in the long run.”
Smart Thinking, Lockdown and COVID-19: Implications for Public Policy²⁴

“For example, the data…shows a decrease in infection rates after countries eased…lockdowns with >99% statistical significance. Indeed…infection rates have declined after reopening even after allowing for an appropriate measurement lag. This means that the pandemic and COVID-19 likely have its own dynamics unrelated to often inconsistent lockdown measures that were being implemented.”
J.P Morgan – Market and Volatility Commentary – Political risks of pandemic, data favors further reopening²⁵

“Restrictions imposed by the pandemic (eg, stay-at-home orders) could claim lives indirectly through delayed care for acute emergencies, exacerbations of chronic diseases, and psychological distress (eg, drug overdoses). … In 14 states, more than 50% of excess deaths were attributed to underlying causes other than COVID-19; these included California (55% of excess deaths) and Texas (64% of excess deaths)”
Excess Deaths From COVID-19 and Other Causes, March-April 2020²⁶

“Comparing weekly mortality in 24 European countries, the findings in this paper suggest that more severe lockdown policies have not been associated with lower mortality. In other words, the lockdowns have not worked as intended”
Did Lockdown Work? An Economist’s Cross-Country Comparison²⁷

“I present a cost-benefit analysis of the response to COVID-19 that finds lockdowns are far more harmful to public health than COVID-19 can be.”
COVID-19: Rethinking the Lockdown Groupthink²⁸

“Our findings … further raise doubt about the importance in NPI’s (lockdown policies in particular) in accounting for the evolution of COVID-19 transmission rates over time and across locations”
Four Stylized Facts About COVID-19²⁹

“The President … has flatly denied the seriousness of the pandemic, refusing to impose a lockdown, close schools, or cancel mass events … Yet the country’s death rate is among the lowest in Europe-just over 700 in a population of 9.5 million”
Covid-19: How does Belarus have one of the lowest death rates in Europe?³⁰

“Consistent with observations that .. lockdown has not been observed to effect the rate…of the country reproduction rates significantly, our analysis suggests there is no basis for expecting lockdown stringency to be an explanatory variable”
Exploring inter-country coronavirus mortality³¹

“The national criteria most associated with death rate are life expectancy and its slowdown, public health context (metabolic and non-communicable diseases (NCD) burden vs. infectious diseases prevalence), economy (growth national product, financial support), and environment (temperature, ultra-violet index). Stringency of the measures settled to fight pandemia, including lockdown, did not appear to be linked with death rate”
Covid-19 Mortality: A Matter of Vulnerability Among Nations Facing Limited Margins of Adaptation³²

“Whether a county had a lockdown has no effect on Covid-19 deaths; a non-effect that persists over time. Cross-country studies also find lockdowns are superfluous and ineffective (Homberg 2020).”
Government mandated lockdowns do not reduce Covid-19 deaths: implications for evaluating the stringent New Zealand response³³

The Science™ on Lockdowns

Navy

Army