Month: March 2021

Dr. Seuss as a policy issue

That is a Substack essay from Matt Yglesias, and open source at that.  Excerpt, using quotation marks rather than forcing further indents on the segment:

“To me, there’s something attractive about the “constitutional copyright” idea of returning to the 1790 Copyright Act rule. But there’s also something attractive about the idea of an author retaining control over their works during their lifetime. There’s also something to be said for the idea that if you publish something and then get hit by a bus the next day, maybe that happenstance shouldn’t cut your heirs out of the downside. Mashing that all together might leave you with life of the author OR 28 years, whichever is longer.

I think it’s hard to specify the exact right number (Rufus Pollock tries with some fancy math and comes up with 15 to 38 years), but these two points from Hal Varian’s paper on copyright terms seem relevant:

  • “Fewer than 11 percent of the copyrights registered between 1883 and 1964 were renewed after 28 years.”
  • “Of the 10,027 books published in 1930, only 174 were still in print in 2001.”

It is just super-rare for old works to have large commercial value. But Xing Li, Megan MacGarvie, and Petra Moser show that copyright extensions have a big impact on consumer prices. And I would argue the cultural cost is higher.”

There is much more at the link.

Jiaolong, China’s Private City

Shruti Rajagopalan and I wrote about India’s private cities, Guragaon and Jamshedpur in Lessons from Gurgaon, India’s Private City. One thing we discovered was that transaction costs prevented private developers from coordinating on infrastructure such as sewage and electricity even though that was clearly the efficient solution and there was plenty of time to bargain. We suggested larger purchases–such as at Disneyland–were necessary to internalize the externalities.

In The Contractual Nature of the City, Qian Lu looks at Jiaolong, an unusual private city in China. Jiaolong is small, about 4.3 square km and a population of 100,000 but all the infrastructure including electricity, sewage, roads, apartments, shopping malls, aquarium, office buildings, and hotels have been built privately. The firm in charge, Jiaolong Co., has planning rights and crucially it collects 25% of all property tax which means it internalizes part of the increase in value generated by its investments.

Jiaolong is a city built and operated by a business corporation. This is rare in China because in most cases the local city government is in charge of urbanization. In almost all cities, government makes land and city planning, takes farmersland, builds city infrastructure, sells land to housing developers and manufacturers, operates police stations, hospitals, schools and universities. By holding the monopoly power of coercion, the government is able to pool together resources by fiat and hold transaction costs low.

The urbanization of Jiaolong is not based on coercive power, but by a series of contracts with Shuangliu government, firms, farmers, residents and other relevant parties. As the central contractor, Jiaolong Co. is able to simplify the contractual web and reduce coordination cost. The essential contracts are the investment contract with the county government to transfer planning rights, and a series of contracts with the government and firms to share tax. Tax sharing contracts define the income rights for Jiaolong so that Jiaolong could share the surplus of urban development and infrastructure construction. Sharing contracts also motivate Shuangliu government to provide public services including protection of property rights. A series of contracts transfer planning rights, land use rights, and income rights to Jiaolong Co., and thereby endogenize the externality of infrastructure building and urban development. From the perspective of institutional change, Jiaolong offers a case of contract-based rather than coercion-based urbanization, the latter being the typical approach in China.

Ross Douthat on Substack on decadence

Meanwhile in the non-ideological regions of the culture the deepening of decadence seems assured. The pandemic has (further) weakened every cultural institution that relies on physical presence, spontaneity and localized or mid-sized audiences, which means basically all of them except the “content” industry, the ever-expanding realm of Peak TV. The spirit of Mustapha Mond presides over the Covid era: He waved his hand; and it was as though, with an invisible feather wisk, he had brushed away a little dust, and the dust was museums, was symphonies; some spider-webs, and they were ballets and bookstores and Broadway theatricals. Whisk. Whisk—and where was the mid-size daily newspaper, the regional university, the local Protestant congregation, the urban Catholic school. Whisk—the place where the local movie theater had been empty. Whisk, the touring pop music acts, whisk, the minor league baseball teams, whisk, whisk ..

I know, I know: We can make art on the blockchain now, and do journalism on Substack, and host salons on Clubhouse.

But if these are the seeds of renaissance, I expect things to get worse before they get better.

Here is the full post, ungated, Ross will be doing free Substack for a limited time.  Ross’s The Decadent Society is coming out soon in paperback, and it a new subtitle and Ross says plenty of new and original material.

Noah Substack interviews Patrick Collison

Here goes, here is one good excerpt of many:

Isaac Asimov’s New Guide to Science. I read that when I was 13 or 14 and thought it was just amazing. (I was an exchange student in Germany at the time. I didn’t learn much German but I did have my eyes opened to many aspects of science that I previously knew nothing about!) Some of John Gribbin’s books, like In Search of Schrödinger’s Cat, really inspired me. Douglas Hofstadter — especially Metamagical Themas. (I read GEB when I was a teenager but found it a bit of a slog.) But, honestly, I think I was always interested in creating technology to some extent. I spent hours and hours playing with Lego when I was young and then transitioned pretty quickly to programming. I remember being pretty certain that I’d love programming before I’d ever written a line of code and, sure enough, I did. So, maybe it’s just something about how my mind is wired.


Overall, my single biggest science policy suggestion would be to pursue far greater structural diversity in our mechanisms. More different kinds of grant making institutions, more different kinds of research organizations, more different career paths for participants, etc. That’s not easy to do — bureaucracies by their nature seek to standardize which this fosters homogeneity. So, to the extent that the Endless Frontier Act can bring us closer to a more structurally varied world, I’m probably supportive relative to the status quo. My biggest qualm would probably be that it combines regional development policy with scientific policy. While the political merit is easy to see, I’m not sure that that’s a good idea. Talent clusters are real and I think it probably makes more sense to think about how best to improve those clusters than it does to foster underdog competitors.

Recommended, interesting throughout.

Dose Stretching Policies Probably *Reduce* Mutation Risk

One objection to dose-stretching policies, such as delaying the second dose or using half-doses, is that this might increase the risk of mutation. While possible, some immunologists and evolution experts are now arguing that dose-stretching will probably reduce mutation risk which is what Tyler and I concluded. Here’s Tyler:

One counter argument is that letting “half-vaccinated” people walk around will induce additional virus mutations.  Florian Kramer raises this issue, as do a number of others.

Maybe, but again I wish to see your expected value calculations.  And in doing these calculations, keep the following points in mind:

a. It is hard to find vaccines where there is a recommendation of “must give the second dose within 21 days” — are there any?

b. The 21-day (or 28-day) interval between doses was chosen to accelerate the completion of the trial, not because it has magical medical properties.

c. Way back when people were thrilled at the idea of Covid vaccines with possible 60% efficacy, few if any painted that scenario as a nightmare of mutations and otherwise giant monster swarms.

d. You get feedback along the way, including from the UK: “If it turns out that immunity wanes quickly with 1 dose, switch policies!”  It is easy enough to apply serological testing to a control group to learn along the way.  Yes I know this means egg on the face for public health types and the regulators.

e. Under the status quo, with basically p = 1 we have seen two mutations — the English and the South African — from currently unvaccinated populations.  Those mutations are here, and they are likely to overwhelm U.S. health care systems within two months.  That not only increases the need for a speedy response, it also indicates the chance of regular mutations from the currently “totally unvaccinated” population is really quite high and the results are really quite dire!  If you are so worried about hypothetical mutations from the “half vaccinated” we do need a numerical, expected value calculation comparing it to something we already know has happened and may happen yet again.  When doing your comparison, the hurdle you will have to clear here is very high.

(See my Washington Post piece for similar arguments and additional references.).

Now here are evolutionary theorists, immunologists and viral experts Sarah Cobey, Daniel B. Larremore, Yonatan H. Grad, and Marc Lipsitch in an excellent paper that first reviews the case for first doses first and then addresses the escape argument. They make several interrelated arguments that a one-dose strategy will reduce transmission, reduce prevalence, and reduce severity and that all of these effects reduce mutation risk.

The arguments above suggest that, thanks to at least some effect on transmission from one dose, widespread use of a single dose of mRNA vaccines will likely reduce infection prevalence…

The reduced transmission and lower prevalence have several effects that individually and together tend to reduce the probability that variants with a fitness advantage such as immune escape will arise and spread (Wen, Malani, and Cobey 2020). The first is that with fewer infected hosts, there are fewer opportunities for new mutations to arise—reducing available genetic variation on which selection can act. Although substitutions that reduce antibody binding were documented before vaccine rollout and are thus relatively common, adaptive evolution is facilitated by the appearance of mutations and other rearrangements that increase the fitness benefit of other mutations (Gong, Suchard, and Bloom 2013; N. C. Wu et al. 2013; Starr and Thornton 2016). The global population size of SARS-CoV-2 is enormous, but the space of possible mutations is larger, and lowering prevalence helps constrain this exploration. Other benefits arise when a small fraction of hosts drives most transmission and the effective reproductive number is low. Selection operates less effectively under these conditions: beneficial mutations will more often be lost by chance, and variants with beneficial mutations are less certain to rise to high frequencies in the population (Desai, Fisher, and Murray 2007; Patwa and Wahl 2008; Otto and Whitlock 1997; Desai and Fisher 2007; Kimura 1957). More research is clearly needed to understand the precise impact of vaccination on SARS-CoV-2 evolution, but multiple lines of evidence suggest that vaccination strategies that reduce prevalence would reduce rather than accelerate the rate of adaptation, including antigenic evolution, and thus incidence over the long term.

In evaluating the potential impact of expanded coverage from dose sparing on the transmission of escape variants, it is necessary to compare the alternative scenario, where fewer individuals are vaccinated (but a larger proportion receive two doses) and more people recover from natural infection. Immunity developing during the course of natural infection, and the immune response that inhibits repeat infection, also impose selection pressure. Although natural infection involves immune responses to a broader set of antibody and T cell targets compared to vaccination, antibodies to the spike protein are likely a major component of protection after either kind of exposure (Addetia et al. 2020; Zost et al. 2020; Steffen et al. 2020), and genetic variants that escape polyclonal sera after natural infection have already been identified (Weisblum et al. 2020; Andreano et al. 2020). Studies comparing the effectiveness of past infection and vaccination on protection and transmission are ongoing. If protective immunity, and specifically protection against transmission, from natural infection is weaker than that from one dose of vaccination, the rate of spread of escape variants in individuals with infection-induced immunity could be higher than in those with vaccine-induced immunity. In this case, an additional advantage of increasing coverage through dose sparing might be a reduction in the selective pressure from infection-induced immunity.

…In the simplest terms, the concern that dose-sparing strategies will enhance the spread of immune escape mutants postulates that individuals with a single dose of vaccine are those with the intermediate, “just right” level of immunity, more likely to evolve escape variants than those with zero or two doses (Bieniasz 2021; Saad-Roy et al. 2021)….There is no particular reason to believe this is the case. Strong immune responses arising from past infection or vaccination will clearly inhibit viral replication, preventing infection and thus within-host adaptation…. Past work on influenza has found no evidence of selection for escape variants during infection in vaccinated hosts (Debbink et al. 2017). Instead, evidence suggests that it is immunocompromised hosts with prolonged influenza infections and high viral loads whose viral populations show high diversity and potentially adaptation (Xue et al. 2017, 2018), a phenomenon also seen with SARS-CoV-2 (Choi et al. 2020; Kemp et al. 2020; Ko et al. 2021). It seems likely, given its impact on disease, that vaccination could shorten such infections, and there is limited evidence already that vaccination reduces the amount of virus present in those who do become infected post-vaccination (Levine-Tiefenbrun et al. 2021).

I also very much agree with these more general points:

The pandemic forces difficult choices under scientific uncertainty. There is a risk that appeals to improve the scientific basis of decision-making will inadvertently equate the absence of precise information about a particular scenario with complete ignorance, and thereby dismiss decades of accumulated and relevant scientific knowledge. Concerns about vaccine-induced evolution are often associated with worry about departing from the precise dosing intervals used in clinical trials. Although other intervals were investigated in earlier immunogenicity studies, for mRNA vaccines, these intervals were partly chosen for speed and have not been completely optimized. They are not the only information on immune responses. Indeed, arguments that vaccine efficacy below 95% would be unacceptable under dose sparing of mRNA vaccines imply that campaigns with the other vaccines estimated to have a lower efficacy pose similar problems. Yet few would advocate these vaccines should be withheld in the thick of a pandemic, or roll outs slowed to increase the number of doses that can be given to a smaller group of people. We urge careful consideration of scientific evidence to minimize lives lost.

Emergent Ventures winners, 13th cohort

Kenny Workman, building tools for computational biology.

Brianna GoPaul, “17 y/o learning fusion energy.”

Justin Glibert, from Belgium near Liege, nanotechnology and cryptography and space manufacturing.

Andrew Tate Young, custom audio from blogs, and to create audiobooks from science information in the public domain.

Rasheed Griffith, Barbados, podcast on China in the Caribbean, and Substack on the same.

Michael Trinh of Toronto, synthetic biology and immunology, general career development.

Austin Diamond, general career development.

Trevor Chow, from Hong Kong now at Cambridge studying economics and monetary policy, here is his blog.

Lea Degen, from southern Germany now in San Francisco, podcasting and general career support, more here.

A splendid cohort, we are honored to have you as winners, and here are previous Emergent Ventures cohorts.

The Covid-19 relief bill

Is that what they should call it?  In any case, for all the bickering over inflation, the real news to me is that the Republicans just didn’t try very hard to fight it.  Partly they are left with few good arguments after their own fiscal profligacy.  Partly they are consumed with their own internal squabbles.  And partly their own pollsters/advisors told them the thing is going to be pretty popular, at least initially and perhaps always.

In my view, this is the watershed event for entering a new era of politics.  Polarization in the old sense peaked in 2011 or so.  I call the new regime “Democrats can get a lot done if they soft pedal it, veer away from the mood affiliation, pretend they do not control the presidency, and stick to ideas that are popular.”

We’ll see how long that lasts, but I think for at least another year.

The new version of “throwing away the key” — our prison regulatory state is failing us

According to Arizona Department of Corrections whistleblowers, hundreds of incarcerated people who should be eligible for release are being held in prison because the inmate management software cannot interpret current sentencing laws.

KJZZ is not naming the whistleblowers because they fear retaliation. The employees said they have been raising the issue internally for more than a year, but prison administrators have not acted to fix the software bug. The sources said Chief Information Officer Holly Greene and Deputy Director Joe Profiri have been aware of the problem since 2019.

The Arizona Department of Corrections confirmed there is a problem with the software.

As of 2019, the department had spent more than $24 million contracting with IT company Business & Decision, North America to build and maintain the software program, known as ACIS, that is used to manage the inmate population in state prisons.

One of the software modules within ACIS, designed to calculate release dates for inmates, is presently unable to account for an amendment to state law that was passed in 2019.

Senate Bill 1310, authored by former Sen. Eddie Farnsworth, amended the Arizona Revised Statutes so that certain inmates convicted of nonviolent offenses could earn additional release credits upon the completion of programming in state prisons. Gov. Ducey signed the bill in June of 2019.

But department sources say the ACIS software is not still able to identify inmates who qualify for SB 1310 programming, nor can it calculate their new release dates upon completion of the programming.

“We knew from day one this wasn’t going to work” a department source said. “When they approved that bill, we looked at it and said ‘Oh, s—.’”

Here is the full story, via Zach Valenta.

The new phrenology?

Ubiquitous facial recognition technology can expose individuals’ political orientation, as faces of liberals and conservatives consistently differ. A facial recognition algorithm was applied to naturalistic images of 1,085,795 individuals to predict their political orientation by comparing their similarity to faces of liberal and conservative others. Political orientation was correctly classified in 72% of liberal–conservative face pairs, remarkably better than chance (50%), human accuracy (55%), or one afforded by a 100-item personality questionnaire (66%). Accuracy was similar across countries (the U.S., Canada, and the UK), environments (Facebook and dating websites), and when comparing faces across samples. Accuracy remained high (69%) even when controlling for age, gender, and ethnicity. Given the widespread use of facial recognition, our findings have critical implications for the protection of privacy and civil liberties.

Here is more from Michal Kosinki, published in a journal which is an offshoot of Nature.

Saturday assorted links

1. Recreating tarot cards in Port-au-Prince.

2. Julia Galef interviews Vitalik, with transcript.

3. For not entirely transparent reasons, China censors Nomadland (the director is ethnic Chinese, but American).  And the real America has reemerged, in Texas.

4. Keynes was more “buy and hold” for his institutional trading for King’s College than for his personal account.

5. Don’t underrate the chemistry and the lipids.

6. Maybe absentee voting didn’t matter so much?

Detroit Fauci

The mayor of Detroit has turned down an allocation of the J&J vaccine.

Detroit Mayor Mike Duggan declined an initial allocation of the newly authorized Johnson & Johnson Covid-19 vaccine….”So, Johnson & Johnson is a very good vaccine. Moderna and Pfizer are the best. And I am going to do everything I can to make sure the residents of the city of Detroit get the best,” Duggan said during a news conference Thursday.

Sigh. What an error. Note, however, that the Detroit Mayor rejecting the J&J vaccine is exactly what the FDA has done with the AstraZeneca vaccine. Moreover none other than Anthony Fauci made exactly the same argument about AstraZeneca (an argument I criticized at the time):

But even if the vaccine ends up being approved, it will probably only have an efficacy of 60 to 70 percent. “What are you going to do with the 70 percent when you’ve got two (vaccines) that are 95 percent? Who are you going to give a vaccine like that to?” Anthony Fauci, the leading American expert on vaccines, recently wondered.

To be clear, I don’t blame Fauci for the actions of Detroit Mayor Mike Duggan. Duggan would probably have said the same had Fauci never made his error. Indeed, perhaps you might even read this as excusing Duggan (if even Fauci, “the leading American expert on vaccines”, could make this error then…).

Still, Fauci’s error has been much more costly for the United States.

Hat tip: JF.