COVID-19 Policy Arguments

This applies only to N95 respirators with valves, which are designed for construction/factory workers or miners. However, N95 masks without valves exist and are used by hospitals. Some municipalities have even banned masks with valves for this reason. See link.

Mechanistic studies find that surgical masks could prevent transmission by infected persons. See links (1, 2)

Cloth masks can act as a physical barrier to prevent large droplets from passing through. See link.

A study finds that homemade cloth masks, although less effective than surgical masks, still significantly reduce the number of microorganisms expelled by wearers. See link.

Even if cloth masks could potentially have no benefit, as long as they are not an active harm it would be better to wear cloth masks than not because there is little cost to doing so.

Although masks can have an effect, it is not sufficient to endanger one’s health; it will be relevant only when going through strenuous physical activity, such as exercising. See link.

Testing alone cannot successfully end lockdowns, because missing even a small number of people with mass testing will allow the pandemic to start again once the lockdowns are lifted. Furthermore, any appreciable false negative or false positive rate will make this problem even worse.

The goal of the CDC and many such similar institutions is not to give the most practical advice, but rather the most health conscious. For example, the CDC recommends or has recommended that steaks always be cooked to at least medium, that women who are not on the pill should never consume alcohol because they may be pregnant, and so on. As such, it is completely appropriate for the CDC to indicate that the safest course of action is to continue with e.g. social distancing even when vaccinated, even if practically speaking it would be generally appropriate for some given individual to do otherwise.

It is impractical to do this because people who have been vaccinated cannot be easily distinguished from people who have not been.

This could be due to issues with fairness. For example, if greenhouses are not considered essential and thus closed, but big grocery stores are allowed to sell the same products, then it would prevent there from being any pent-up demand for those closed greenhouses (thus placing them in even more dire financial straights) and would unduly favor the grocery stores.

If this was really the concern, mandatory temporary furlow and basic income for those who must stop working due to health would be more appropriate than lockdown. If there are people who would be worse off without a job than taking the risk of the pandemic, then removing the option of continuing with their job through lockdown does not improve their situation.

Some abnormalities, such as extremely high demand for funeral urns, showed that the pandemic was not under control at the time it was claimed to be.

In contrast to the situation where everyone must accept lockdown restrictions, it seems like purely a positive good to allow those with no ability to spread or be infected by the virus to live a regular life.

In general, if there are individuals who stand to gain from some intervention, then those individuals should have to pay for that intervention themselves, as is the case with consumer goods.

The methodology of determining the value of a statistical life is not appropriate here. Here is the methodology: ‘Suppose each person in a sample of 100,000 people were asked how much he or she would be willing to pay for a reduction in their individual risk of dying of 1 in 100,000, or 0.001%, over the next year. Since this reduction in risk would mean that we would expect one fewer death among the sample of 100,000 people over the next year on average, this is sometimes described as "one statistical life saved.” Now suppose that the average response to this hypothetical question was $100. Then the total dollar amount that the group would be willing to pay to save one statistical life in a year would be $100 per person × 100,000 people, or $10 million. This is what is meant by the "value of a statistical life.”’ This has issues because people are firstly unable to comprehend and meaningfully reason about very small probabilities, similarly to very large numbers. Furthermore, this approach does not work because people are naturally less willing to pay for extending other lives rather than their own. Although it is possible to make a moral argument about the equality of people’s lives and other factors, this does not show that an unrestricted pandemic would be more economically costly than with lockdown measures.

The methodology of determining the value of a statistical life is not appropriate here. Here is the methodology: ‘Suppose each person in a sample of 100,000 people were asked how much he or she would be willing to pay for a reduction in their individual risk of dying of 1 in 100,000, or 0.001%, over the next year. Since this reduction in risk would mean that we would expect one fewer death among the sample of 100,000 people over the next year on average, this is sometimes described as "one statistical life saved.” Now suppose that the average response to this hypothetical question was $100. Then the total dollar amount that the group would be willing to pay to save one statistical life in a year would be $100 per person × 100,000 people, or $10 million. This is what is meant by the "value of a statistical life.”’ This has issues because people are firstly unable to comprehend and meaningfully reason about very small probabilities, similarly to very large numbers. Furthermore, this approach does not work because people are naturally less willing to pay for extending other lives rather than their own. Although it is possible to make a moral argument about the equality of people’s lives and other factors, this does not show that an unrestricted pandemic would be more economically costly than with lockdown measures.

The methodology of determining the value of a statistical life is not appropriate here. Here is the methodology: ‘Suppose each person in a sample of 100,000 people were asked how much he or she would be willing to pay for a reduction in their individual risk of dying of 1 in 100,000, or 0.001%, over the next year. Since this reduction in risk would mean that we would expect one fewer death among the sample of 100,000 people over the next year on average, this is sometimes described as "one statistical life saved.” Now suppose that the average response to this hypothetical question was $100. Then the total dollar amount that the group would be willing to pay to save one statistical life in a year would be $100 per person × 100,000 people, or $10 million. This is what is meant by the "value of a statistical life.”’ This has issues because people are firstly unable to comprehend and meaningfully reason about very small probabilities, similarly to very large numbers. Furthermore, this approach does not work because people are naturally less willing to pay for extending other lives rather than their own. Although it is possible to make a moral argument about the equality of people’s lives and other factors, this does not show that an unrestricted pandemic would be more economically costly than with lockdown measures.

It was not necessarily known early on in the pandemic that outdoor events would not result in spikes in cases of COVID-19. Given the uncertainty around the pandemic, it may have been better to be safe than sorry.

This is no longer relevant due to the successes of vaccine development.

Although mask-wearing might be appropriate, criminalizing going outside without a mask is not appropriate either due to concerns about civil liberties.

Sex is (generally) consensual and does not occur in public. This makes STDs very different from something like COVID-19, with which people can infect one another by merely standing in each other’s vicinity. One cannot be blamed for being infected with COVID-19 if someone at the checkout line of a store came to close, but the same cannot be said of STDs.

Voting is a right, while eating is a restaurant is not a right.

The number of COVID-19 related deaths clearly demonstrates that COVID-19 is much worse than the flu, although not necessarily orders of magnitude worse. See link.

About one in one thousand children exposed to COVID-19 are affected by a Kawasaki-like disease. Furthermore, that COVID-19 can cause this is evidence that it could potentially do other forms of serious long-term harm, as “bad things” in terms of poor health are correlated since things that harm you one way are likely to harm you in others.

Although hospital beds were available, there was a shortage of PPE, kidney dialysis machines, etc. Therefore, it is not necessarily accurate to say that there was no danger of hospital systems being overwhelmed.

IFRs in the most overwhelmed areas were significantly higher than in areas which were hit less hard. This supports the idea that overwhelmed hospital systems can lead to significantly higher death rates, even if hospital systems were not literally overflowing.

Lockdowns can also save lives that would have otherwise been lost for non-COVID-19 related reasons, such as reduction in violent crimes, some types of driving accidents, other infectious diseases, and so on.

Many of the problems attributed to lockdowns, such as economic contractions, could easily be caused by people choosing on their own to e.g. socially distance and refrain from going out. If most people do not participate in a reopening, then similar problems would occur in any case.

In Italy, a nationwide lockdown was implemented on March 9th. This graph shows the relationship between reported cases per 100,000 people and excess deaths per 100,000 people within each Italian province. The data is from April 15th, and excess deaths are measured relative to 2015-2019 averages. The 41 least infected provinces (as measured by cases per 100,000 people) combine for zero excess deaths in 2020. These provinces account for 20.9 million of Italy's 60.4 million population. This offers suggestive evidence that lockdowns have very little effect on non-COVID mortality in the short-run.

The measures were taken to prevent an even worse situation. Therefore, even if it might appear to be an overreaction, as we cannot observe the counterfactual this is not a good method to evaluate lockdown policies. If no measures were applied, then it is possible than many more people would have died.

States that saw strong, continuous lockdowns such as New York generally had early large outbreaks, which would itself prevent the possibility of future large outbreaks. Therefore, the relationship cannot be identified as causal.

Data on suicide rates shows that they have actually been lower than usual, or at least not higher. See link.

This does not account for exponential growth. It could be possible that the relaxation of restrictions leads to additional cases, but this will not be an obvious spike until long in the future.

It is easily possible that causality runs the other way: because these states had disproportionate deaths, they applied stronger lockdown policies. Furthermore, these states have greater density and are more urban, which could increase susceptibility to COVID-19.

Elected officials have not benefited from the crisis, as it has distracted them from other goals that they might prefer to prioritize. Furthermore, lockdown policies have alienated parts of the electorate, which is undesirable for elected officials.

State governments that have implemented strong lockdowns have received enormous budgetary losses, so there is little financial incentive for it.

Many of those studies have been underpowered or administered hydroxychloroquine too belatedly to be of use. For example, this study shows that hydroxychloroquine reduces risk by about 18% and reduces the risk even more strongly for those who took it earlier. Although it does not reach the 95% significance threshold, a preponderance of non-decisive evidence can still inform individual decisions or even policy.

The WHO claims hydroxychloroquine does not increase risk of death for COVID-19 patients, so there is little cost to using it and a potentially large benefit. See link.

Although naturally causality cannot be clearly determined, countries that have been using hydroxychloroquine, such as Honduras, saw a significant drop in mortality after implementing the treatment.

The CDC only aggregates data provided to them by the states into national statistics and released their interim case definition by early April of 2020. Insofar as states have conflated statistics, they have been violating the CDC guidelines.

While in the U.S. there is insufficient public health infrastructure, in smaller countries with a dedicated apparatus it may be achievable.

Unregulated tests could easily create either hysteria or false confidence if they are not properly functional, administered by untrained users, or even scams.

Distrust of medical institutions and procedures is a major problem, and removing the regulatory authority of the FDA would exacerbate this issue.

If one considers only personal interest, this calculus could easily be changed through policy measures incentivizing vaccination. For example, proposals to condition a "return to normalcy" on evidence of being vaccinated would strongly shift this, especially if the relevant vaccine appears to be generally safe.

The incarcerated are effectively in quarantine to begin with and are trivial to relocate within a prison compound.

Such policies can lead to other negative effects, such as rises in crime rates or drug use.

Viewing a list of states by median age shows that there is no clear pattern of stereotypically rural states being older or younger than stereotypically urban ones. The top 5 are Maine, New Hampshire, Vermont, West Virginia, and Florida. 4 of these are mostly rural while the remaining one is North America's #1 Retirement Destination. But the bottom is similar - 5 youngest states are Utah, Alaska, Texas, North Dakota, and Nebraska.

This (non-peer-reviewed) study shows only that a small minority of the non-specific antigens they tested cause lymphocytes from people who got the vaccine to respond differently from those of people who did not receive the vaccine. The differences are statistically significant but the magnitude is fairly small. They did not compare it to other vaccines. In short, it is not a surprise that an injection known to cause inflammation (which is desirable so that immune cells are recruited to the site of injection) causes a short-term change in how the body reacts to other things that might also cause inflammation.