Tag Archives: Heterogeneity

The Decline and Fall of a Virus

19 Tuesday May 2020

Posted by in Uncategorized

≈ Leave a comment

Tags

,

Asymptomatic cases of coronavirus have some important implications, both good and bad. Of course, it’s great that so many people are asymptomatic. It demonstrates an innate immunity or some other kind of acquired immunity to the virus. On the other hand, these individuals can still spread the virus while infected, and they are hard to identify.

Estimates of the share of asymptomatic cases vary tremendously, some reaching almost 90%. But being asymptomatic is a matter of degree: in some cases there might be no symptoms whatsoever, from initial infection to complete suppression. In others, the symptoms are mild and may not raise any alarm in one’s mind. That distinction implies that testing criteria should be broadened, especially as the cost of testing declines.

Here I show two simple examples of viral spread to demonstrate that some level of asymptomatic “pre-immunity” in the population reduces the threshold at which the impact of the virus reverses. Both examples involve a population of 100 people. In both cases, social interactions are such that an infected person infects an average of two others. That is, the initial reproduction rate (known as R0) is equal to two. In both examples, the process starts with one infected individual:

Example #1:

Everyone is susceptible, meaning that the virus will cause symptoms and illness in anyone who catches it. Condensing the timeline, let’s just say we go from the first infection to three infections; then #2 and #3 each infect two more, and we have a total of seven infections; then the extra four pass the virus along to another eight victims and we’re up to 15; and so on. This is the exponential growth that is characteristic of the early stage of an epidemic. But then other dynamics start to kick in: most of the infected people recover with adaptive immunity, though a few may die. By now, however, only 85 susceptible people remain in the population, so each infected person infects an average of less than two more. The reproduction rate R must fall from it’s initial value of R0 as the susceptible population shrinks. By the time 50 people are infected and 50 susceptible people remain, the value of R is halved. In this example, that’s where herd immunity is achieved: when 50% of the population has been infected.

For those who enjoy math, here is a useful relation:

Herd Immunity Threshold (HIT) = 1 – 1/R0.

The higher is R0, the initial reproduction rate, the more people must be infected to achieve herd immunity. The coronavirus is said to have an R0 somewhere in the mid-2s. If it’s 2.5, then 60% of the population must be infected to achieve herd immunity under the assumption of universal susceptibility. When 60% are infected, R is equal to one. More people will be infected beyond that time, but fewer and fewer. R continues to fall, and the contagion wanes.

While I’ve abstracted from the time dimension, the total number of people who will be infected depends on factors like the duration of an infection. It takes time for an infected individual to come into contact with new, susceptible hosts for the virus, and fewer hosts will be available as time passes. That means the virus will die out well before the full population has been infected.

Example #2:

Let’s say 40 of the 100 people are not susceptible to the virus, meaning they will experience few if any symptoms if they catch it. Those 40 are innately immune, or perhaps they retain some adaptive immunity from previous exposure to a non-novel coronavirus. Strictly speaking, the entire population can catch the virus and can transmit it to others, but only 60% the population is susceptible to illness. It’s still true that each infected person would infect two others at the start. However, only 1.2 of those newly infected people would get sick on average. I will call that value the effective R0, which is net of the immune cohort. By the time 17 people have been infected, and about 10 of them get sick, there are only 50 susceptible people remaining. The effective R is already down to one. Herd immunity is effectively achieved after less than 20 infections. The HIT is just 17% (rounded)! That means the number of symptomatic infections will begin declining beyond that point. In this case, again depending on the average duration of an infection, it’s likely that much less than half of the population is ultimately sickened by the virus.

To summarize thus far, what example #2 demonstrates is that the existence of prior immunity in some individuals reduces the effective HIT. We know that sub-groups have differing levels of prior immunity / susceptibility to the coronavirus. In fact, for the coronavirus, we know the non-susceptible share of the population is substantial, given the large number of individuals who have been exposed but were asymptomatic.

Other Impacts on Reproduction Rate

Other influences can inhibit the spread of a virus. Weather, for example (see the nice interactive tool in “Weather and Transmission Rates“). Social distancing, including avoidance of “super-spreader events“, reduces the average number of people anyone can come into contact with. Masks might reduce the spread to others as well. Quarantining infected individuals obviously eliminates contacts with other individuals. Quarantining susceptible individuals prevents them from being exposed. In all of these cases, R is reduced more drastically over time from it’s initial value R0. This reduces the effective HIT and the ultimate number of individuals infected. Those effects are incremental to the impact of a large, non-susceptible sub-group, as in example #2, And there are variations on the appeal to heterogeneity that are equally convincing, as described below.

New HIT Literature

So herd immunity is not as far out of reach as many believe. That question is now being addressed more intensively in the academic world. Herd immunity occurs in the context of a virus’s ability to spread from host to host, which is summarized by R. In my limited review, most of the articles addressing a lower HIT emphasize distancing or other practices that reduce R. However, herd immunity really means that given a set of social conditions, enough of the population has either an innate or an acquired immunity to cause the impact of a contagion to recede. Both the level of immunity and the social conditions can alter the effective HIT.

Jacob Sullum offers a nice summary of some of this work. One paper describing the impact of heterogeneity emphasizes the order in which individuals become infected. Here is Sullum’s description with a link to the paper:

“A couple of new reports speculatively lower the possible herd immunity threshold for the coronavirus to just 10 to 20 percent of the population. This conjecture depends chiefly on assumptions about just how susceptible and connected members of the herd are. In their preprint, a team of European epidemiologists led by the Liverpool School of Tropical Medicine mathematical bioscientist Gabriela Gomes explains how this might work.

If highly susceptible herd members become infected and thus immune first, the preprint says, their subsequent interactions with the still-uninfected will not result in additional cases. Basically, the virus stymies itself by disproportionately removing those most useful to it from contributing to its future transmission. In addition, if herd members are very loosely connected and interact with one another rarely, the virus will have a much harder time jumping to its next victims. Sustained social distancing aimed at flattening the curve of coronavirus infections and cases mimics this effect.”

The sequential explanation is of obvious importance, But don’t it’s not the fundamental mechanism at play in example #2, which is strictly the heterogeneity of the population.

Nick Spyropoulas of the Alma Economics Group describes reductions in the herd immunity threshold in “Notes on the Dynamics of Subsequent Epidemic Waves“. It’s a very nice write-up, but it only emphasizes social distancing.

Judith Curry provides an excellent and well-referenced exposition of some herd immunity experiments. They are based on an even more extended approach to heterogeneity introducing: 1) variation in susceptibility across individuals; and 2) variation in the dispersion of transmission. The latter means, “… the extent to which infection happens through many spreaders or just a few“. She uses these mechanisms to modify a standard epidemiological model using prior estimates of variability to calibrate the model. Both experiments arrive at drastically lower HITs and total infections than her baseline experiment, which uses the standard model. The chart below shows her results with moderate heterogeneity. Her results with more extreme (though realistic) values of the heterogeneity metrics are even more remarkable. See the link above.

Check Against Real World

How does all this square with our experience to-date with the coronavirus? It’s difficult to tell with case counts, as the volume of testing keeps increasing and so many infected individuals are asymptomatic and remain undiagnosed. Estimates of R vary, but most states appear to have an R currently less than one. That means the virus is receding almost everywhere in the U.S. The same is true in much of the developed world, where the virus was most prevalent. Even Sweden, where achieving herd immunity is policy, diagnosed cases and deaths have been largely confined to vulnerable groups, and in total are less than many other (though not all) European countries.

Does that mean many areas in the U.S. and elsewhere have reached herd immunity? Locales that have had serological testing have thus far shown infection rates of anywhere from 2% to 10%, though New York City, where the outbreak was most severe, may have had more than 20% of its population infected as of a month ago. Different regions may have different HITs, so there is a chance that some areas, including NYC, are close to herd immunity.

Unfortunately, some of the reductions in R and in the effective HIT were won by social distancing, which will be reversed to some extent as the economy reopens. That’s the flip side of the “flat curve” we’ve managed to experience. The value of R may drift back toward or above one for a time. Diminished sunlight and humidity in the fall might have a similar effect. A second wave is not likely to be as bad as the first, however. That’s because: 1) we’ll now have more adaptive immunity in the population; 2) the most susceptible people are among those who already have acquired immunity or, more sadly, have died; 3) we’ll be better at coping with an outbreak in multiple ways; and 4) more speculatively, we’ll have identified the most effective treatments and, with less likelihood, a vaccine for those who want it.

Policy Lessons

In any outbreak, keeping R below one at least-cost is the objective. Given the alternatives, that rules out full-scale lockdowns because we know a large share of the population already has innate or acquired immunity. Forced shutdowns are unnecessarily costly relative to a targeted approach. But what form does that take?

Infected individuals must be quarantined until they recover, and their close contacts should be quarantined for up to a full incubation period. Large gatherings must be suspended temporarily. Testing capacity must be such that anyone with a fever or any symptom, mild or otherwise, can be tested. Regular testing of certain individuals like health care workers, teachers, and other first responders should take place. Simple screenings using infrared thermometers will be useful in high-traffic establishments. Precautions must be targeted at the most susceptible, and it’s pretty easy to identify them: the elderly and those with co-morbidities such as heart disease, diabetes, and lung conditions.

There are questions of civil liberties that must be addressed as well. Many high-risk individuals can live independently, so their freedoms must be weighed against their safety. Keeping this cohort quarantined is out of the question unless it’s voluntary. Regular testing should take place, and a subset of this group might already have the markers of immunity. Another question of civil liberties involves detailed contact tracing, which requires the establishment of an apparatus capable of great intrusion and abuse. I believe identification of close contacts should be an adequate precaution, though there may be degrees of tracing that I would find acceptable. Finally, a vaccine would be welcome, but it should not be mandatory

 

 

 

On the Meaning of Herd Immunity

09 Saturday May 2020

Posted by in Pandemic, Public Health, Risk

≈ Leave a comment

Tags

, , , , , , , , , , , , , , , , , , , , , , ,

Immunity doesn’t mean you won’t catch the virus. It means you aren’t terribly susceptible to its effects if you do catch it. There is great variation in the population with respect to susceptibility. This simple point may help to sweep away confusion over the meaning of “herd immunity” and what share of the population must be infected to achieve it.

Philippe Lemoine discusses this point in his call for an “honest debate about herd immunity“. He reproduces the following chart, which appeared in this NY Times piece by Carl T. Bergstrom and Natalie Dean:

Herd immunity, as defined by Bergstrom and Dean, occurs when there are sufficiently few susceptible individuals remaining in the population to whom the actively-infected can pass the virus. The number of susceptible individuals shrinks over time as more individuals are infected. The chart indicates that new infections will continue after herd immunity is achieved, but the contagion recedes because fewer additional infections are possible.

We tend to think of the immune population as those having already been exposed to the virus, and who have recovered. Those individuals have antibodies specifically targeted at the antigens produced by the virus. But many others have a natural immunity. That is, their immune systems have a natural ability to adapt to the virus.

Heterogeneity

At any point in a pandemic, the uninfected population covers a spectrum of individuals ranging from the highly susceptible to the hardly and non-susceptible. Immunity, in that sense, is a matter of degree. The point is that the number of susceptible individuals doesn’t start at 100%, as most discussions of herd immunity imply, but something much smaller. If a relatively high share of the population has low susceptibility, the virus won’t have to infect such a large share of the population to achieve effective herd immunity.

The apparent differences in susceptibility across segments of the population may be the key to early herd immunity. We’ve known for a while that the elderly and those with pre-existing conditions are highly vulnerable. Otherwise, youth and good health are associated with low vulnerability.

Lemoine references a paper written by several epidemiologists showing that “variation in susceptibility” to Covid-19 “lowers the herd immunity threshold”:

“Although estimates vary, it is currently believed that herd immunity to SARS-CoV-2 requires 60-70% of the population to be immune. Here we show that variation in susceptibility or exposure to infection can reduce these estimates. Achieving accurate estimates of heterogeneity for SARS-CoV-2 is therefore of paramount importance in controlling the COVID-19 pandemic.”

The chart below is from that paper. It shows a measure of this variation on the horizontal axis. The colored, vertical lines show estimates of historical variation in susceptibility to historical viral episodes. The dashed line shows the required exposure for herd immunity as a function of this measure of heterogeneity.

Their models show that under reasonable assumptions about heterogeneity, the reduction in the herd immunity threshold (in terms of the percent infected) may be dramatic, to perhaps less than 20%.

Then there are these tweets from Marc Lipsitch, who links to this study:

“As an illustration we show that if R0=2.5 in an age-structured community with mixing rates fitted to social activity studies, and also categorizing individuals into three categories: low active, average active and high active, and where preventive measures affect all mixing rates proportionally, then the disease-induced herd immunity level is hD=43% rather than hC=1−1/2.5=60%.”

Even the celebrated Dr. Bergstrom now admits, somewhat grudgingly, that hereogeniety reduces the herd immunity threshold, though he doesn’t think the difference is large enough to change the policy conversation. Lipsitch also is cautious about the implications.

Augmented Heterogeneity

Theoretically, social distancing reduces the herd immunity threshold. That’s because infected but “distanced” people are less likely to come into close contact with the susceptible. However, that holds only so long as distancing lasts. John Cochrane discusses this at length here. Social distancing compounds the mitigating effect of heterogeneity, reducing the infected share of the population required for herd immunity.

Another compounding effect on heterogeneity arises from the variability of initial viral load on infection (IVL), basically the amount of the virus transmitted to a new host. Zvi Mowshowitz discusses its potential importance and what it might imply about distancing, lockdowns, and the course of the pandemic. In any particular case, a weak IVL can turn into a severe infection and vice versa. In large numbers, however, IVL is likely to bear a positive relationship to severity. Mowshowitz explains that a low IVL can give one’s immune system a head start on the virus. Nursing home infections, taking place in enclosed, relatively cold and dry environments, are likely to involve heavy IVLs. In fact, so-called household infections tend to involve heavier IVLs than infections contracted outside of households. And, of course, you are very unlikely to catch Covid outdoors at all.

Further Discussion

How close are we to herd immunity? Perhaps much closer than we thought, but maybe not close enough to let down our guard. Almost 80% of the population is less than 60 years of age. However, according to this analysis, about 45% of the adult population (excluding nursing home residents) have any of six conditions indicating elevated risk of susceptibility to Covid-19 relative to young individuals with no co-morbidities. The absolute level of risk might not be “high” in many of those cases, but it is elevated. Again, children have extremely low susceptibility based on what we’ve seen so far.

This is supported by the transmission dynamics discussed in this Twitter thread by Dr. Muge Cevik. She concludes:

“In summary: While the infectious inoculum required for infection is unknown, these studies indicate that close & prolonged contact is required for #COVID19 transmission. The risk is highest in enclosed environments; household, long-term care facilities and public transport. …

Although limited, these studies so far indicate that susceptibility to infection increases with age (highest >60y) and growing evidence suggests children are less susceptible, are infrequently responsible for household transmission, are not the main drivers of this epidemic.”

Targeted isolation of the highly susceptible in nursing homes, as well as various forms of public “distancing aid” to the independent elderly or those with co-morbidities, is likely to achieve large reductions in the effective herd immunity ratio at low cost relative to general lockdowns.

The existence of so-called super-spreaders is another source of heterogeneity, and one that lends itself to targeting with limitations or cancellations of public events and large gatherings. What’s amazing about this is how the super-spreader phenomenon can lead to the combustion of large “hot spots” in infections even when the average reproduction rate of the virus is low (R0 < 1). This is nicely illustrated by Christopher Moore of the Santa Fe Institute. Super-spreading also implies, however, that while herd immunity signals a reduction in new infections and declines in the actively infected population, “hot spots” may continue to flare up in a seemingly random fashion. The consequences will depend on how susceptible individuals are protected, or on how they choose to mitigate risks themselves.

Conclusion

I’ve heard too many casual references to herd immunity requiring something like 70% of the population to be infected. It’s not that high. Many individuals already have a sort of natural immunity. Recognition of this heterogeneity has driven a shift in the emphasis of policy discussions to the idea of targeted lockdowns, rather than the kind of indiscriminate “dumb” lockdowns we’ve seen. The economic consequences of shifting from broad to targeted lockdowns would be massive. And why not? The health care system has loads of excess capacity, and Covid infection fatality risk (IFR) is turning out to be much lower than the early, naive estimates we were told to expect, which were based on confirmed case fatality rates (CFRs).

Private Social Distancing, Private Reversal

04 Monday May 2020

Posted by in Liberty, Pandemic, Uncategorized

≈ Leave a comment

Tags

, , , , , , , , , , , , , , , , , , , , , ,

My original post on the dominance of voluntary social distancing over the mandated variety appears below. That dominance is qualified by the greater difficulty of engaging in certain activities when they are outlawed by government, or when the natural locations of activities are declared off-limits. Nevertheless, as with almost all regulation, people make certain “adjustments” to suit themselves (sometimes involving kickbacks to authorities, because regulation does nothing so well as creating opportunities for graft). Those “adjustments” often lead to much less desirable outcomes than the original, unregulated state. In the case of a pandemic, however, it’s tempting to view such unavoidable actions as a matter of compromise.

I say this now because the voluntary social distancing preceding most government lockdown orders in March (discussed in the post below) is subject to a degree of self-reversal. Apple Mobility Data suggests that something like that was happening throughout much of April, as shown in the chart at the top of this post. Now, in early May, the trend is likely to continue as some of the government lockdown mandates are being lifted, or at least loosened.

An earlier version of the chart above appeared in a Forbes article entitled, “Apple Data Shows Shelter-In-Place Is Ending, Whether Governments Want It To Or Not“. The author, John Koetsier, noted the Apple data are taken from map searches, so they may not be reliable indicators of actual movement. But he also featured some charts from Foursquare, which showed actual visits to various kinds of destinations, and some of theoe demonstrate the upward trend in activity.

In the original post below, I used SafeGraph charts lifted from a paper I described there. The four charts below are available on the SafeGraph website, which offered the services of the friendly little robot in the lower right-hand corner, but I demurred. You’ll probably need to click on the image to read the detail. They show more granular information by industry, brand, region, and restaurant categories. The upward trends are evident in quite a few of the series.

I should qualify my interpretation of the charts above and those in my original post: First, nine states did not have stay-at-home orders, though a few of those had varying restrictions on individuals and on the operation of “non-essential” businesses. The five having no orders of any kind (that I can tell) are lightly-populated, very low-density states, so the vast majority of the U.S. population was subject to some sort of lockdown measure. Second, eight states began to ease or lift orders in the last few days of April, Georgia and Colorado being the largest. Therefore, at the tail end, a small part of the increase in activity could be related to those liberalizations. Then again, it might have happened anyway.

The authoritarian impulse to shut everything down was largely unnecessary, and it did not accomplish much that voluntary distancing hadn’t accomplished already (again, see below). Healthy people need to stop cowering and take action. That includes the non-elderly and those free of underlying health conditions. Sure, take precautions, keep your distance, but get out of your home if you can. Get some sunny Vitamin D.

Committing yourself to the existence of a shut-in is not healthy, not wise, and it might destroy whatever wealth you possess if you are a working person. The data above show that people are recognizing that fact. As much as the Left wishes it were so, government seldom “knows better”. It is least effective when it uses force to suppress voluntary behavior; it is most effective when it follows consensus, and especially when it protects the rights of individuals to make their own choices where no consensus exists.

Last week’s post follows:

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

How much did state and local governments accomplish when they decided to issue stay-at-home orders? Perhaps not much. That’s the implication of data presented by the authors of “Internal and external effects of social distancing in a pandemic” (starts on page 22 in the linked PDF). Social distancing began in the U.S. in a series of voluntary, private actions. Government orders merely followed and, at best, reinforced those actions, but often in ham-handed ways.

The paper has a broader purpose than the finding that social distancing is often a matter of private initiative. I’ll say a bit more about it, but you can probably skip the rest of this paragraph without loss of continuity. The paper explores theoretical relationships between key parameters (including a social distancing construct) and the dynamics of a pandemic over time in a social welfare context. The authors study several alternatives: a baseline in which behavior doesn’t change in any way; a “laissez faire” path in which actions are all voluntary; and a “socially optimal” path imposed by a benevolent and all-knowing central authority (say what???). I’d offer more details, but I’ll await the coming extension promised by the authors to a world in which susceptible populations are heterogenous (e.g., like Covid-19, where children are virtually unaffected, healthy working age adults are roughly as at-risk as they are to the flu, and a population of the elderly and health-compromised individuals for which the virus is much more dangerous than the flu). In general, the paper seems to support a more liberalized approach to dealing with the pandemic, but that’s a matter of interpretation. Tyler Cowen, who deserves a hat-tip, believes that reading is correct “at the margin”.

Let’s look at some of the charts the authors present early in the paper. The data on social distancing behavior comes from Safegraph, a vendor of mobility data taken from cell phone location information. This data can be used to construct various proxies for aggregate social activity. The first chart below shows traffic at “points of interest” (POI) in the U.S. from March 8 to April 12, 2020. That’s the blue line. The red line is the percentage of the U.S. population subject to lockdown orders on each date. The authors explain the details in the notes below the chart:

Clearly POI visits were declining sharply before any governments imposed their own orders. The next two charts show similar declines in the percent of mobile devices that leave “home” each day (“home” being the device’s dominant location during nighttime hours) and the duration over which devices were away from “home”, on average.

So all of these measures of social activity began declining well ahead of the government orders. The authors say private social distancing preceded government action in all 50 states. POI traffic was down almost 40% by the time 10% of the U.S. population was subject to government orders, and those early declines accounted for the bulk of the total decline through April 12. The early drops in the two away-from-home measures were 15-20%, again accounting for well over half of the total decline.

The additional declines beyond that time, to the extent they can be discerned, could be either trends that would have continued even in the absence of government orders or reinforcing effects the orders themselves. This does not imply that lockdown orders have no effects on specific activities. Rather, it means that those orders have minor incremental effects on measures of aggregate social activity than the voluntary actions already taken. In other words, the government lockdowns are largely a matter of rearranging the deck chairs, or, that is to say, their distribution.

Many private individuals and institutions acted early in response to information about the virus, motivated by concerns about their own safety and the safety of family and friends. The public sector in the U.S. was not especially effective in providing information, with such politicos as President Donald Trump, Nancy Pelosi, Andrew Cuomo, Bill De Blasio, and the mayor of New Orleans minimizing the dangers into the month of March, and some among them encouraging people to get out and celebrate at public events. Even Anthony Fauci minimized the danger in late February (not to mention the World Health Organization). In fact, “the scientists” were as negligent in their guidance as anyone in the early stages of the pandemic.

When lockdown orders were issued, they were often arbitrary and nonsensical. Grocery stores, liquor stores, and Wal Mart were allowed to remain open, but department stores and gun shops were not. Beaches and parks were ordered closed, though there is little if any chance of infection outdoors. Lawn care services, another outdoor activity, were classified as non-essential in some jurisdictions and therefore prohibited. And certain personal services seem to be available to public officials, but not to private citizens. The lists of things one can and can’t buy truly defies logic.

In March, John W. Whitehead wrote:

“We’re talking about lockdown powers (at both the federal and state level): the ability to suspend the Constitution, indefinitely detain American citizens, bypass the courts, quarantine whole communities or segments of the population, override the First Amendment by outlawing religious gatherings and assemblies of more than a few people, shut down entire industries and manipulate the economy, muzzle dissidents, ‘stop and seize any plane, train or automobile to stymie the spread of contagious disease,’…”

That is fearsome indeed, and individuals can accomplish distancing without it. If you are extremely risk averse, you can distance yourself or take other precautions to remain protected. You can either take action to isolate yourself or you can decide to be in proximity to others. The more risk averse among us will internalize most of the cost of voluntary social distancing. The less risk averse will avoid that cost but face greater exposure to the virus. Of course, this raises questions of public support for vulnerable segments of the population for whom risk aversion will be quite rational. That would certainly be a more enlightened form of intervention than lockdowns, though support should be offered only to those highly at-risk individuals who can’t support themselves.

Christopher Phelan writes of three rationales for the lockdowns: buying time for development of a vaccine or treatments; reducing the number of infected individuals; and to avoid overwhelming the health care system. Phelan thinks all three are of questionable validity at this point. A vaccine might never arrive, and Phelan is pessimistic about treatments (I have more hope in that regard). Ultimately a large share of the population will be infected, lockdowns or not. And of course the health care system is not overwhelmed at this point. Yes, those caring for Covid patients are under a great stress, but the health care system as a whole, and patients with other maladies, are currently suffering from massive under-utilization.

If you wish to be socially distant, you are free to do so on your very own. Individuals are quite capable of voluntary risk mitigation without authoritarian fiat, as the charts above show. While private actors might not internalize all of the external costs of their activities, government is seldom capable of making the appropriate corrections. Coercion to enforce the kinds of crazy rules that have been imposed during this pandemic is the kind of abuse of power the nation’s founders intended to prevent. Reversing those orders can be difficult, and the precedent itself becomes a threat to future liberty. Nevertheless, we see mounting efforts to resist by those who are harmed by these orders, and by those who recognize the short-sighted nature of the orders. Private incentives for risk reduction, and private evaluation of the benefits of social and economic activity, offer superior governance to the draconian realities of lockdowns.

Social Distancing Largely a Private Matter

26 Sunday Apr 2020

Posted by in Liberty, Pandemic, Uncategorized

≈ Leave a comment

Tags

, , , , , , , , , , , , , , , , ,

How much did state and local governments accomplish when they decided to issue stay-at-home orders? Perhaps not much. That’s the implication of data presented by the authors of “Internal and external effects of social distancing in a pandemic” (starts on page 22 in the linked PDF). Social distancing began in the U.S. in a series of voluntary, private actions. Government orders merely followed and, at best, reinforced those actions, but often in ham-handed ways.

The paper has a broader purpose than the finding that social distancing is often a matter of private initiative. I’ll say a bit more about it, but you can probably skip the rest of this paragraph without loss of continuity. The paper explores theoretical relationships between key parameters (including a social distancing construct) and the dynamics of a pandemic over time in a social welfare context. The authors study several alternatives: a baseline in which behavior doesn’t change in any way; a “laissez faire” path in which actions are all voluntary; and a “socially optimal” path imposed by a benevolent and all-knowing central authority (say what???). I’d offer more details, but I’ll await the coming extension promised by the authors to a world in which susceptible populations are heterogenous (e.g., like Covid-19, where children are virtually unaffected, healthy working age adults are roughly as at-risk as they are to the flu, and a population of the elderly and health-compromised individuals for which the virus is much more dangerous than the flu). In general, the paper seems to support a more liberalized approach to dealing with the pandemic, but that’s a matter of interpretation. Tyler Cowen, who deserves a hat-tip, believes that reading is correct “at the margin”.

Let’s look at some of the charts the authors present early in the paper. The data on social distancing behavior comes from Safegraph, a vendor of mobility data taken from cell phone location information. This data can be used to construct various proxies for aggregate social activity. The first chart below shows traffic at “points of interest” (POI) in the U.S. from March 8 to April 12, 2020. That’s the blue line. The red line is the percentage of the U.S. population subject to lockdown orders on each date. The authors explain the details in the notes below the chart:

Clearly POI visits were declining sharply before any governments imposed their own orders. The next two charts show similar declines in the percent of mobile devices that leave “home” each day (“home” being the device’s dominant location during nighttime hours) and the duration over which devices were away from “home”, on average.

So all of these measures of social activity began declining well ahead of the government orders. The authors say private social distancing preceded government action in all 50 states. POI traffic was down almost 40% by the time 10% of the U.S. population was subject to government orders, and those early declines accounted for the bulk of the total decline through April 12. The early drops in the two away-from-home measures were 15-20%, again accounting for well over half of the total decline.

The additional declines beyond that time, to the extent they can be discerned, could be either trends that would have continued even in the absence of government orders or reinforcing effects the orders themselves. This does not imply that lockdown orders have no effects on specific activities. Rather, it means that those orders have minor incremental effects on measures of aggregate social activity than the voluntary actions already taken. In other words, the government lockdowns are largely a matter of rearranging the deck chairs, or, that is to say, their distribution.

Many private individuals and institutions acted early in response to information about the virus, motivated by concerns about their own safety and the safety of family and friends. The public sector in the U.S. was not especially effective in providing information, with such politicos as President Donald Trump, Nancy Pelosi, Andrew Cuomo, Bill De Blasio, and the mayor of New Orleans minimizing the dangers into the month of March, and some among them encouraging people to get out and celebrate at public events. Even Anthony Fauci minimized the danger in late February (not to mention the World Health Organization). In fact, “the scientists” were as negligent in their guidance as anyone in the early stages of the pandemic.

When lockdown orders were issued, they were often arbitrary and nonsensical. Grocery stores, liquor stores, and Wal Mart were allowed to remain open, but department stores and gun shops were not. Beaches and parks were ordered closed, though there is little if any chance of infection outdoors. Lawn care services, another outdoor activity, were classified as non-essential in some jurisdictions and therefore prohibited. And certain personal services seem to be available to public officials, but not to private citizens. The lists of things one can and can’t buy truly defies logic.

In March, John W. Whitehead wrote:

“We’re talking about lockdown powers (at both the federal and state level): the ability to suspend the Constitution, indefinitely detain American citizens, bypass the courts, quarantine whole communities or segments of the population, override the First Amendment by outlawing religious gatherings and assemblies of more than a few people, shut down entire industries and manipulate the economy, muzzle dissidents, ‘stop and seize any plane, train or automobile to stymie the spread of contagious disease,’…”

That is fearsome indeed, and individuals can accomplish distancing without it. If you are extremely risk averse, you can distance yourself or take other precautions to remain protected. You can either take action to isolate yourself or you can decide to be in proximity to others. The more risk averse among us will internalize most of the cost of voluntary social distancing. The less risk averse will avoid that cost but face greater exposure to the virus. Of course, this raises questions of public support for vulnerable segments of the population for whom risk aversion will be quite rational. That would certainly be a more enlightened form of intervention than lockdowns, though support should be offered only to those highly at-risk individuals who can’t support themselves.

Christopher Phelan writes of three rationales for the lockdowns: buying time for development of a vaccine or treatments; reducing the number of infected individuals; and to avoid overwhelming the health care system. Phelan thinks all three are of questionable validity at this point. A vaccine might never arrive, and Phelan is pessimistic about treatments (I have more hope in that regard). Ultimately a large share of the population will be infected, lockdowns or not. And of course the health care system is not overwhelmed at this point. Yes, those caring for Covid patients are under a great stress, but the health care system as a whole, and patients with other maladies, are currently suffering from massive under-utilization.

If you wish to be socially distant, you are free to do so on your very own. Individuals are quite capable of voluntary risk mitigation without authoritarian fiat, as the charts above show. While private actors might not internalize all of the external costs of their activities, government is seldom capable of making the appropriate corrections. Coercion to enforce the kinds of crazy rules that have been imposed during this pandemic is the kind of abuse of power the nation’s founders intended to prevent. Reversing those orders can be difficult, and the precedent itself becomes a threat to future liberty. Nevertheless, we see mounting efforts to resist by those who are harmed by these orders, and by those who recognize the short-sighted nature of the orders. Private incentives for risk reduction, and private evaluation of the benefits of social and economic activity, offer superior governance to the draconian realities of lockdowns.

American Homicide Rates: Which America?

12 Thursday Oct 2017

Posted by in Discrimination, Gun Control, Immigration

≈ 1 Comment

Tags

, , , , , , , , , , , , , ,

A heterogenious society and the successful assimilation of minorities are two very different things, as much as we might wish otherwise. Two populations within a region will come into contact, but conditions promoting real assimilation are complex. (I’m avoiding use of the term “diversity” because it has come to imply the successful assimilation of distinct groups.) While cultural differences can enrich the lives of both populations, sharp economic gaps between minority and majority populations (and even some cultural differences) will tend to slow the process of assimilation. This is often associated with social dysfunction, such as high crime and homicide rates, especially among the minority group. This is a fairly common phenomenon in countries with racial and ethnic minority or immigrant populations, as Ryan McMaken writes in a recent piece on international differences in heterogeneity and homicide rates.

Heterogeneity In the West

Countries in the Western Hemisphere tend to have relatively high immigrant and minority populations, as McMaken describes:

“… when considering the Americas, … nation-states are in most cases frontier states with populations heavily affected by immigration, a history of conflict with indigenous populations, and institutionalized chattel slavery that lasted until the 19th century. The factors are significant through the region, and the United States cannot be held apart in this regard from the Caribbean, Brazil, Colombia, and other states impacted by all these factors. 

Importantly, these factors also make the Americas significantly different from Western Europe and other areas — Japan and Korea, for example — where the present situation is marked by much higher levels of cultural uniformity and quite different recent histories and current demographic trends.“

Homicides

McMaken questions popular theories of cross-country differences in homicide rates based on the degree of gun control and gun ownership rates. Homicides and violent victimization have been declining in the U.S. for many years even as gun ownership has soared. Furthermore, international comparisons are traditionally plagued by arbitrary country classifications and exclusions, as well as inconsistent definitions of homocide and gun ownership. However, McMaken points to other explanations for violent crime found to be fairly robust in the academic literature: poverty and population heterogeneity:

“… these factors contribute to lower levels of social cohesion, and thus higher levels of criminality and other socially-undesirable behaviors.“

McMaken cites research involving ethnic minority populations of Slavs in Germany, Italians in Argentina and the U.S., and Arabs in Europe, all of whom had crime rates far exceeding those in their countries of origin. The connection between heterogeneity and crime might have nothing to do with particular ethnic groups, though it seems all too easy for observers within individual countries to blame specific “others” for crime. It is a symptom of alienation from the majority as well as economic desperation and vulnerability to opportunities and threats arising from the underground economy. Illegal activities might truly provide the best alternatives available to low-skilled, minority job seekers. Needless to say, underground economic activity, such as the drug trade, involves high risk and often violence among users and between competing factions. This is an important source of the high crime and victimization that typifies many minority communities.

Despite declines since the 1970s, the U.S. still has a higher homicide rate than many other industrialized countries. Beyond the weakness cited above, such comparisons fail to control for other confounding effects, including the degree of heterogeneity across countries.

Policies

Heterogeneity poses a problem in the context of involuntary and often voluntary segregation of sub-cultures. If you don’t believe the “voluntary” part, take a close look at the different clusters of individuals in the cafeteria at almost any “diverse” university or corporate office. Judge for yourself. Differences in language, fertility, demographics, religion and cultural traditions may be noteworthy, but where crime is associated with effectively segregated minorities, there is usually a gap in economic status and mobility relative to society at large.

What policies can mitigate these conditions and their impact on crime? It would be nice to approach this question strictly from the perspective that heterogeneity is a given, but the degree of heterogeneity is, to some extent, an endogenous outcome. Restrictive immigration policies might leap to mind as a way of restraining heterogeneity, and there is little doubt that illegal immigrants are less likely to assimilate (many contend that their crime rate is low). Policies allowing less restricted flows of legal immigrants tend to be salutary if they are based on domestic economic need, economic potential, or compassion for those seeking asylum or a haven from political oppression. A legal immigrant receiving a welcome on new shores is more likely to assimilate successfully than an illegal immigrant, all else equal. Citizenship and language education are avenues through which assimilation might be encouraged. And there could be ways to improve sponsorships and even temporary visa programs so as to encourage assimilation.

What can be done to encourage more effective assimilation of all minorities? And what can be done to reduce the crime associated with unassimilated populations? One major corrective is a strong economy. Policies that encourage economic growth will lead to greater participation in markets and society, with consequent interaction and mixing of sub-cultures. Growth policies include low and non-distortionary taxes and light regulation.

The war on drugs also accounts for a major share of homicides, and that war interacts with non-assimilation in perverse ways. It is crippling to disadvantaged communities precisely because it creates risky “opportunities” in the underground economy. It also produces high levels of incarceration and dangerous forms of “cut” contraband. As I’ll discuss in my next post, ending the war on drugs would reduce violent crime and lead to safer drugs in relatively short order.

A short list of other policies that would foster assimilation and economic mobility would include: improved education: school choice and apprenticeship programs; better labor market outcomes: reduce the minimum wage or create sub-minimum wage categories to enhance opportunities to gain experience and skills; better housing: eliminate rent controls.

Assimilation is always more effective when it occurs “organically”. Affirmative action and forced diversity initiatives often fail to achieve effective assimilation. Beyond the obvious infringement on liberty, these policies may sow resentment among those who suffer reverse discrimination, and among those who witness it, to the probable detriment of efforts to eliminate bias. Even worse, these policies often put their intended beneficiaries into vulnerable, un-winnable situations: jobs or programs for which their skills are not adequate. There are undoubtedly excellent candidates among those placed in positions under quotas, but there is a likelihood that many will be unsuccessful in their roles.

Conclusion

The anti-gun left is eager to attribute differences in homicide rates to the impact of gun control policies, but a close examination of the facts reveals better explanations. A prominent factor contributing to differences in homicide rates is the degree of heterogeneity across countries. Those with more homogeneous populations tend to have lower homicide rates and vice versa. But the problem is not merely heterogeneity, but the difficulty of economic and cultural assimilation of minority populations. These factors appear to lead to greater crime within many minority populations. The U.S. is not unique in its experience with high minority crime rates, but it is a relatively heterogenous nation. This is an important factor in explaining why the homicide rate tends to be higher in the U.S. than in other industrialized countries. To close, I’ll offer something cogent from Bretigne Shaffer’s On the Banks blog, in which she offers this quote from an individual named Michael Owen (the soccer player?):

“... we don’t really have a single America with a moderately high rate of gun deaths. Instead, we have two Americas, one of which has very high rates of gun ownership but very low murder rates, very comparable to the rest of the First World democracies such as those in western & northern Europe, Australia, New Zealand, Canada, Japan, South Korea. The other America has much lower rates of gun ownership but much, much higher murder rates, akin to violent third world countries.“