Causal Deference

Books I Read in 2025

Fri, 02 Jan 2026 15:38:48 -0800

Books I read in 2025:

* The Secrets of Consulting by Gerald Weinberg.
Secondhand Time: The Last of the Soviets by Svetlana Alexievich.
A Fistful of Shells: West Africa from the Rise of the Slave Trade to the Age of Revolution by Toby Green.
The Way of Kings by Brandon Sanderson. Stormlight Archive 1. Recommended by Emmett Shear as an example of nobility.
Words of Radiance by Brandon Sanderson. Stormlight Archive 2.
* The Origins of Efficiency by Brian Potter.
Original Love: The Four Inns on the Path of Awakening by Henry Shukman. I also attended a retreat with the author at Mount Madonna.
* Stubborn Attachments by Tyler Cowen. It takes bravery to just confess your assumptions and moral frameworks in this manner.

My top recommendations noted with a *.

The best book here is Stubborn Attachments, which is also quite short and available on audiobook. You probably won’t agree with his specific framework but it really highlights a lot of important cruxes that we should all should be dealing with regardless of our specific commitments. For example, if you are consequentialist adjacent, what is your discount factor? If it’s low you should care about climate change but also you should want to reduce taxes on the rich, because in the long term they will create wealth that can be taxed to improve future welfare. If it’s high, you should tax the rich but also not care much about long term climate impacts. How many of us are consistent in the way we think about the value of the future?

The Brian Potter book was fun too, lots of wild industrial porn. The modern world is a miracle that’s made possible by endless small optimizations that add up to huge improvements. Inspired me to burn a lot of time playing Factorio.

Aside from those this was a disappointing year in my book choices. Hopefully I am starting the new year off on a better foot with Tokarczuk’s Books of Jacob.

ChatGPT on Bachs Mass in B Minor

Tue, 06 May 2025 23:17:21 -0700

ChatGPT is surprisingly useful or learning about and appreciating art. For the longest time I didn’t understand the fuss about Bach’s Mass in B Minor. It felt overlong and taking it literally as a Mass, I thought it was zany that it took about twenty minutes to say “Kyrie Eleison, Christe Eleison, Kyrie Eleison”. But now I’ve cracked it and I can’t stop listening to it.

One really useful way ChatGPT helped me was by explaining the musical ideas in each movement. Bach is contrasting different styles and moods to construct a prayer that touches on many aspects of spirituality. I’m reproducing the full deep dive here.

Kyrie

kyrie eleison

this is a full-on choral fugue, but soaked in grief. it’s written in stile antico, the “old style,” nodding to renaissance polyphony (think palestrina) but with bach’s brutal precision and depth. voices layer in slowly, each entering like someone stepping into a funeral procession. the counterpoint is tight, but the harmony is dark, dissonances rubbed raw. it’s less supplication than lamentation.

christe eleison

suddenly we’re in a bright galant duet for two sopranos, way more modern in style, almost italianate. it’s as if he’s saying: penitence has many masks—tears, grace, longing, light.

kyrie eleison

the second kyrie slams you back into minor-key choral counterpoint, tighter and more urgent than the first. less mournful, more pleading.

it’s like he’s setting up a triptych of theological mood swings. each one a different mode of crying out. none of them quite resolving.

Gloria

the gloria in the mass in b minor is like bach throwing the whole arsenal at the heavens. it’s exuberant, kaleidoscopic, stylistically omnivorous. here’s a quick character sketch of each movement:

gloria in excelsis deo

full choral + trumpet explosion. d major. this is joy as fireworks. rhythmically insistent, harmonically bold. it’s bach saying “look up.”

et in terra pax

abrupt mood shift—slower, more solemn. fugal textures, rich suspensions. this is peace as something hard-won, not soft. almost feels like a choral sigh after the opening blast.

laudamus te

soprano solo with violin—graceful, italianate, dance-adjacent. the praise here is personal, elegant, almost flirtatious. a galant interlude.

gratias agimus tibi

chorus again, and a fugue. more restrained than the opening—formal gratitude, not ecstatic. it’s a revision of a cantata chorus, bc of course it is.

domine deus

duet for tenor and soprano with flute. warm, lyrical. one of the most pastoral sections—god as light and beauty. feels like a miniature opera duet in a clearing.

qui tollis peccata mundi

chorus with plaintive oboe. minor key, slow, weighty. the lamb of god taking away sin feels crushing, like divine empathy under pressure.

qui sedes ad dexteram patris

alto aria with oboe d’amore. lilting, almost dance-like. a little odd, tonally ambiguous—heavenly throne room as courtly salon?

quoniam tu solus sanctus

bass aria with… horn and two bassoons? yeah. it’s weird, regal, and beastly. kind of an echo of the quoniam in the credo. like bach built a menagerie for the trinity.

cum sancto spiritu

closing choral fugue, full-on baroque banger. relentless, brilliant, engineered. like the heavens turned into clockwork and started to sing. feels earned after the journey.

Credo

the credo is the theological meat of the mass and bach sets it like a gothic vault: dense, structural, symbolic. this isn’t belief as feeling—it’s belief as architecture. eleven movements, each a rhetorical brick.

credo in unum deum

a choral creed-fugue—strict, stile antico, evoking ancient authority. monotone-ish intonation on “credo” over imitative counterpoint. feels eternal, pre-baroque, etched in stone.

patrem omnipotentem

switches to stile moderno—bright, orchestral, flowing lines. like the doctrine of the Father gets a fresher, more narrative setting. classic contrast play.

et in unum dominum

duet for soprano and alto. sweet and balanced—an intimate assertion of the Son’s divinity. like theology sung as a madrigal.

et incarnatus est

massive mood shift. slow, hushed, suspended harmonies. incarnation becomes mystery. the music almost stops breathing. bach’s handling here is reverent to the edge of fear.

crucifixus

passacaglia over a descending bass line = death spiral. choral writing is static, weary, beautiful in the way rot can be. baroque despair. one of the most devastating movements he ever wrote.

et resurrexit

whiplash—suddenly we’re in dazzling d major, trumpets screaming. resurrection = ecstasy. fugue built from leaping motifs, as if Christ vaults out of the tomb.

et in spiritum sanctum

bass aria with two oboes d’amore—gentle, lyrical, almost pastoral. holy spirit as warm breath, not fire. it’s serene. oddly human-scale for the third person of the trinity.

confiteor unum baptisma

BACK to stile antico. this is a double fugue. deeply austere. tightest counterpoint in the mass. if “i confess one baptism” were etched into granite, it would sound like this.

et expecto resurrectionem

this is where it snaps. the fugue collapses into suspended harmony—like the world holding its breath—and then erupts into joy. resurrection is no longer abstract, it’s here. total musical catharsis.

…the credo is structured like a theological narrative: from eternal god to the human christ to cosmic victory. bach uses style-switching (old/new, fugue/aria, minor/major) as his rhetorical toolkit.

each movement isn’t just a belief—it’s an experience of that belief. the whole credo feels like an act of construction: belief laid brick by counterpointed brick, until it forms a temple large enough to sing in.

Sanctus

sanctus

starts in 3/8, in a slow, stately dance—almost sarabande-like. but this isn’t human dance—it’s the cosmos in procession. the rhythm pulses like celestial breath.

he opens with massive block chords (“sanctus, sanctus, sanctus”)—not contrapuntal but architectural, as if each voice is a stone laid in the temple. harmonic rhythm is glacial. every shift feels like tectonic plates moving.

then the fugue: “dominvs devs sabaoth.” now it takes flight. swirling melismas, imitation, ecstatic layering. the mood turns from reverent to blazing. this isn’t just praise, it’s overwhelming.

you can hear the seraphim circling, like in isaiah’s vision—“holy holy holy” not whispered but thundered by entities made of fire.

pleni sunt caeli et terra gloria eius

modulation to e minor—darker, more earthy. the text says “heaven and earth are full of his glory” and bach responds by grounding it. rhythm picks up. counterpoint becomes thicker, more imitative. the architecture gets messy—in a good way. fullness = complexity.

hosanna in excelsis

back to d major and trumpet ecstasy. it’s the same music he’ll use later after the benedictus—clearly linked. double choir, fast compound meter, relentless rhythmic drive. this is exaltation as centrifuge. it spins and spins and refuses to resolve.

Benedictus

ah, the benedictus—that’s one of the strangest, most delicate moments in the whole mass.

where so much of the mass in b minor is cosmic, overwhelming, vertical, the benedictus is intimate, almost private. it’s scored for just tenor, flute, and continuo. no chorus. no grandeur. it feels like stepping out of a cathedral into a quiet grove.

the melody is sinuous, floating, almost improvisatory. the flute weaves around the tenor like breath around prayer. harmonically it’s gentle but unsettled—like hope tempered by sorrow.

and that text—“blessed is he who comes in the name of the lord”—usually gets triumphant settings. bach makes it tender, maybe even fragile. like the arrival isn’t a parade, but a whispered entrance.

it’s a moment of human-scale devotion in a work that otherwise feels built for angels. and then—boom—the hosanna returns in full choral glory, shattering the hush. you barely get to sit with the benedictus before being swept away again. it’s the eye of the storm.

Agnus Dei

agnus dei

the agnus dei is a plea for mercy and peace, set in g minor. bach adapted this movement from an earlier aria, “ach, bleibe doch,” from his ascension oratorio, which itself was derived from the lost wedding cantata auf, süß entzückende gewalt. the aria features a solo alto voice intertwined with an obbligato violin line, creating a texture of intimate supplication.

dona nobis pacem

the dona nobis pacem isn’t new music—it’s a reprise of the gratias agimus tibi fugue from the gloria. but bach recontextualizes it, and that changes everything.

in the gloria, the music says “thank you”; here, it begs for peace. same notes, different soul. this is baroque semiotics at full power—meaning arises not just from melody but from placement.

musically, it’s in d major, bright and triumphant—but the text is dona nobis pacem, and that plea feels deeper than the surface. it’s not soft or meditative—it’s monumental. this is peace as an act of will, not a gentle afterthought. like bach is insisting on peace the way a general might demand surrender.

also: ending the mass on a fugue—not on an aria, not on a chorus of resolution—is weird and brilliant. there’s no quiet fade, no “amen.” instead, a return to formal structure, to recursive logic, to music-as-machine-of-hope.

so what should you note?

it’s a quotation masquerading as a prayer
it’s circular—bringing the mass full-circle in theme and form
it’s peace not as balm but as burden, shouted into the void

this isn’t peace you fall into. it’s peace you build. fugally. together.

Recordings

I also asked ChatGPT for recommendations on recordings, mentioning that the one I’ve had on repeat is Pichon’s latest recording (which I strongly recommend btw). It googled and responded:

pichon’s new mass in b minor is a lush, emotionally charged take—more romantic than purist. his tempos swing wide: the second kyrie is glacial, the gloria bursts forward, and the crucifixus hits like a hammer. some critics love the drama; others find it overwrought and question the balance between voices and instruments.

for contrast, check out:

gardiner (1985): crisp, agile, with a lean period-instrument sound. his cum sancto spiritu is a masterclass in controlled exuberance.
herreweghe (1998): lyrical and restrained. the et incarnatus est here is all hush and inward glow—less theater, more prayer.
harnoncourt (1968): raw, edgy, and historically bold. it’s not pretty, but it’s alive. the quoniam sounds like a baroque fever dream.
richter (1961): massive, reverent, unapologetically romantic. not HIP, but if you want your dona nobis pacem to feel like the end of the world, this is it.

California Wildfire Updates

Thu, 09 Jan 2025 23:51:05 -0800

I updated a couple beliefs on the basis of these fires.

NEPA is worse than you think

Yup. When the Forest Service identifies high-risk forests needing prescribed burns, it takes an average of 4.7 YEARS just to get through environmental reviews. For complex projects, it's 7.2 years - longer than many fire cycles. https://t.co/nITtIe9bDd pic.twitter.com/p9i1hmeQne
— Tahra Hoops (@TahraHoops) January 8, 2025

It turns out that it takes years to get approval to do prescribed burns in California. Even mechanical treatment takes over a year to get a NEPA decision. With litigation, the average timeline goes up to 9.4 years. We know how to control fires, but we don’t allow our government to do it.

Some will blame activists, but the way the law works any project can be sued if it doesn’t proactively conduct a very expensive and time-consuming environmental review, which effectively makes it necessary to do so. Ezra Klein has written about this here. Environmental regulation is complex, but there must be a better way to achieve the goal of protecting the environment, while still enabling crucial ecological interventions.

The fire insurance market is broken in Caliofrnia

Guys.

We — as in Californian voters who voted in 1988 — voted for a system where every time an insurance company wants to raise rates for automobiles or homes, they have to have a public hearing and an elected insurance commissioner has to approve that rate increase. pic.twitter.com/7FjFZbN6iw
— Kim-Mai Cutler (@kimmaicutler) January 10, 2025

In 1988, California passed Proposition 103, which required public hearings to raise insurance rates. Naturally that makes it quite difficult to raise rates. Given that fire-prone areas are increasingly expensive to insure, insurers are dropping coverage from customers in these areas. In fact, in Pacific Palisades specifically, thousands of homes were dropped from fire insurance.

Because this is California, there’s even a non-profit grift incolving an organization called Consumer Watchdog that, ironically, profits from contesting rate increases through something called the “intervenor process”. But that’s another story.

There is an “insurer of last resort” called the California FAIR Plan, but it’s almost illiquid. It has about $385M in cash and $2.5B in reinsurance. Its exposure to the current LA fires is in the neighborhood of $24B. In the likely event that it can’t pay out, it will split the remaining costs among all insurers in the state, proportional to market share. Those insurers will then pass this on to their customers in the form of a surcharge, that could reach thousands of dollars. It’s possible this will also drive insurers out of the state (because they don’t want to risk being saddled with this liability), increasing insurance rates everywhere, regardless of fire risk.

Nobody seems concerned about moral hazard

The big problem with having homeowners throughout the state paying out to the wealthy homeowners in fire prone areas, aside from that being highly distributionally regressive, is that it creates a moral hazard. Fire prone areas are constantly being developed and redeveloped. Instead of taking on the risk of negative outcomes, homeowners know they will receive generous insurance payouts in the event of a fire. This means we’ll see more people building in these areas, more fires, more homes destroyed, and more payouts from the homeowners who did not choose dangerous locations to live.

The war on prices

All of this insurance business relates to people’s distrust of prices. Insurance premia could serve as a quantitative indicator of the risk of building and owning homes, if insurance premia were allowed reflected reality. Instead they’re not allowed to adjust, resulting in homes being uninsured or bailed out by the state with all the moral hazard that entails. Relevant book (that I still haven’t yet read) here.

Other things people are talking about.

There are reports of looting and arson, are people less prosocial than we think?
I’m seeing throwaway takes implicating climate change but have yet to see a more serious assessment of the role of climate change in these fires.

Books I Read in 2024

Wed, 01 Jan 2025 22:30:13 -0800

It was a much shorter list this year because I spent a lot of effort getting into classical music, so I’ve given it its own section.

Books

Standouts in bold.

Troubled by Rob Henderson. Autobiography depicting the author’s experience with the foster care system. Surprisingly this is a page turner and the topic is something we should all learn more about. [goodreads]
The Missing Billionaires: A Guide to Better Financial Decisions by Victor Haghani and James White. Cognitive biases cause systematic investment errors. A claim I hadn’t heard: There isn’t much generational wealth in the US because heirs tend to squander it and make bad investments. [goodreads]
Eve: How the Female Body Drove 200 Million Years of Human Evolution by Cat Bohannon. Cool topic, uneven epistemics. My review on Goodreads is here.
Dominion: How the Christian Revolution Remade the World by Tom Holland [goodreads]
Valuable Humans in Transit and Other Stories by qntm. Cool conceptual scifi short stories by the antimimetics guy. [goodreads]
Poor Charlie’s Almanack, talks by Charlie Munger. Avuncular.
The Secrets of Consulting by Gerald M Weinberg. There’s always a problem. It’s always a people problem. [goodreads]
The Universal Christ by Richard Rohr. I was not able to figure this one out and didn’t finish it. Weirdly though reading it did make me think about what it would mean to embody Christ in the world, and actually caused me to handle a few situations better.
Unsong by Scott Alexander. This novel asks, waht if Kabbalah was real? Wordplay and puns would matter a lot.
The Secrets of Our Success by Joseph Henrich [goodreads]
Crooked Plow by Itamar Vieira Junior. The life of two sisters in semi-feudal rural Brazil. [goodreads]

The clear winners were Holland’s Dominion and Henrich’s Secrets of our Success. They both tell the story of why we are the way we are.

Henrich’s Secrets of our Success traces the early origins of hominids and Homo sapiens. Why did humans take over the world? He argues forcefully and with a mountain of evidence that human intelligence is not the reason humans are dominant. Rather it is our ability to learn from one another and develop culturally mediated bodies of knowledge that accumulate over time that separates us from other animals. One scary implication: cultural improvement generally comes about via selection. So if our own culture “gets sick” the way that will be resolved is that we will lose out to other healthier cultures.

Tom Holland asks questions about more recent cultural evolution that happened in Western Europe during its Christian era. Why is our culture and morality the way it is? Why do we think people are “equal”? Why do we think there is a thing called “human rights”? Why do we ban slavery? The roots of all of these lie in the history of European moral thought, as it developed under Christianity and Roman Catholicism. Holland also documents how these ideas were intentionally packaged up as universal, rather than Christian, to make them more persuasive in the Ottoman Empire and beyond. Bigtime viewquakes in this book come from simply asking questions about things most people take for granted. This book pairs well with Henrich’s book The WEIRDest People in the World and Sarah Ruden’s Paul Among the People.

Honorable mention goes to Troubled, a touching and very readable memoir of one child’s experience in the foster care system.

Classical Music

It got started when I listened to Tyler Cowen’s podcast with Rick Rubin about music. I really enjoyed it and thought, it would be great to listen to books that have music in them. Strangely there aren’t a lot of actual audiobooks that do this, but Audible has many courses by Robert Greenberg, my neighbor in Oakland, that mix the art history of music with actual selections. I went a little overboard and listened to all of the following:

How to Listen to and Understand Great Music, 3rd Edition
Bach and the High Baroque
The 30 Greatest Orchestral Works
Great Masters: Haydn - His Life and Music
The Chamber Music of Mozart
Beethoven’s Piano Sonatas
The String Quartets of Beethoven
Great Masters: Mahler - His Life and Music

These were fantastic and I felt like they really shifted me from vaguely knowing stuff about classical music to feeling fairly confident about what’s going on with the great masters. Some things I learned:

General observations

I found generally I needed to listen to a fair amount of music through, in the background if necessary to get the ear for them. After that I really enjoy most pieces. I need to do this less now that I have a lot under my belt.
Spotify sucks at classical music because it doesn’t have a concept that the same piece could have multiple recording artists. I’ve been paying for Idagio because it’s a much better catalogue and it can be cool to see what recordings are more or less popular, but the actual streaming is buggy. I don’t think there’s an ideal solution to this.
Naming conventions are indefensibly bad. Everything is a weird numbering scheme with opus numbers plus pieces are literally just named after their tonal mode. Some of the value with Greenberg’s lectures is he explains this stuff.

Baroque Era

I was suprised to find the Baroque era easy to love. It’s all about someone playing an instrument virtuosically and the band getting into it. They go very hard. I’ve been spinning Jean Rondeau’s Dynastie album on repeat, first track on youtube here. But pretty much all of Bach’s concertos (e.g. BWV 1042, Brandenburg 5) are just a riot. Right now I’m also getting into Giardino Armonico’s Handel op 6.
Chaconne or Passicaglia is a very simple art form that results in gorgeous works, especially the solo violin piece by Bach. That might be one of my favorite pieces of music now but I did have to listen a few times to get it.
Greenberg’s explanation of the Goldberg Variations (also a type of Chaconne!) was extremely helpful, he dedicates 4 whole lectures to them! Again an absolutely amazing piece that took me quite a bit of exposure to warm up to.
Bach’s Cantata BWV 140 “Wachtet Auf” is almost like a sequence of super catchy pop songs. Video here.
I’m almost embarassed how much I love everything Bach did.
The French court of Louis XIV had a lot going on. French Overture was basically something Lully cooked up to be Louis’ theme song, see a depiction here. His successor Rameau’s opera pieces are really cinematic, for example Entrée de Polymnie.
Francesco Corti’s harpsichord recording of Handel’s suites is excellent for working, maybe the first time I’ve been addicted to harpsichord music.

Classical Era

Haydn was really jarring after listening to a lot of Baroque. I had to sort of detox. I ended up loving his string quartets the most, especially opus 76. For example here he makes the sun rise in Op 76 no 4 youtube.
Nothing can go harder than Mozart depicting the Don getting owned by the undead statue that he invited to his house youtube.
My favorite Mozart other than opera ended up being the late string quintets and his divertimento for string trio, but all his late piano concertos are great too and it’s fun to think of him rocking the house for a quick buck. Check out Uchida doing #20 here.
Beethoven’s piano concertos were a big discovery for me. #4 and #5 are desert island music, I’m honestly more partial to them than any of his symphonies. Trifonov playing #5 is here.
Beethoven’s piano sonatas are ridiculously good and it seemed to me actually explored a broader range of musical ideas than his symphonies or quartets. I think of his sonatas as his creative sketchbook. He was also one of the greatest pianists of his time. The famous ones are awesome like Apassionata and Waldstein, but I found myself listening to op 26 and op 54 on repeat. Listen to Lisitsa play the Apassionata here.
I found I had to work hard to get into quartets. Maybe this is the most alien format for 21st century ears. But Beethoven’s are truly awesome. I connected best to Razumovsky 1 (here), Harp, and the A Minor op 132.
It is astonishing that Haydn, Mozart, and Beethoven all lived in Vienna at roughly the same time. What we call “Classical Music” is really just the scene going on in Vienna in the 1780s and 90s.

Romantic Era

I did not connect to the romantic era strongly this year.
I did see Mahler’s 4th symphony in concert and it is awesome.
I also saw Salonen conduct Mahler’s 3rd and it is also awesome. It really requires you to sit and pay attention to it for 90 minutes, and being in person really helps maybe even partly just for the attention, although the dynamics is also really hard to reproduce in a recording.
I believe Trifonov’s Mazeppa (watch here) was his breakout performance. Listening to this next to other recordings can help you understand why not all performances of classical music are created equal.

Modern Era

Trifonov has an awesome CD of early Russian musicians called Silver Age. I especially love the recording of Prokofiev’s Piano Concerto #2, you can see him performing it live here.
Shostakovich is pretty awesome, I spent a bunch of time with Levit’s recording of his preludes and fugues, teaser here. Or for something with no chill whatsoever, check out quartet 8 mvt 2 for a head banger here.
There is still orchestral music being written! John Adams is easy to love as a contemporary composer. Listen to Yuja Wang playing his concerto “Must the Devil Have All the Good Tunes?” on youtube.

Effort and Ease

Wed, 27 Dec 2023 08:56:13 -0800

I’m on my yoga mat, and I’m struggling. I’m straining hard to do a bridge pose. My pelvis is off the floor and I’m grimacing. I feel tension through my whole body. I feel like I will collapse at any second, and I don’t know how long I can hold on.

I’m in a class taught by my favorite yoga teacher. She says “let’s meet this pose with effort and ease, effort and ease. Where is this pose asking us to work hard, and where is it asking us to relax?” I don’t get the hint, I continue to struggle. “Is your face helping you with this pose? If it’s not, it can relax.” Finally I understand. I relax my face, my pelvis stays rigid. Tension evacuates my head. It’s a struggle actually, to release the tension in my head while maintaining the pose in my core. I struggle with that. But it’s a new struggle, a better struggle. Now I’m struggling to make it easier. And I haven’t lost form. Muscles tense and relax throughout my head and neck as I try to figure out which ones I actually need. My body is learning, each muscle is learning where I need to make effort and where I need to ease up. I am getting good at bridge pose.

It’s popular in online advice these days to tell people they should ease up on themselves, stop “self-coercing”, lighten up. This can be good advice, but it’s not quite the right mindset. Working hard, maybe even 12 hour days and weekends, doing things your bodymind resists doing are essential to personal development. The trouble is finding the right combination of effort and ease. Often people strain in vain. It’s incredible how much energy you can spend worrying about whether something will work, or being mad about some personal drama. So lighten up on that. But double down on what’s important. Your life should have the right combination of effort and ease.

Nick Cammarata makes a similar point here:

a lot of TPOT people’s problems (meaning, money, impact etc) would be solved if they just worked (~40hrs+/wk) on a fun enough problem with an emotionally safe team. Lots of people realized they didn’t like work and so became anti job, but they were just working on the wrong thing
— Nick (@nickcammarata) August 19, 2023

Books I Read in 2023

Sun, 17 Dec 2023 19:54:46 -0800

My main theme for 2023 was religion, and Buddhism in particular, so I’ve moved those into their own section.

Fiction

Circe by Madeline Miller goodreads
Ficciones by Jorge Luis Borges goodreads
Cold Nights of Childhood by Tezer Özlü goodreads
The Death of Ivan Ilych by Leo Tolstoy, translated by Pevear and Volokhonsky goodreads
Cryptonomicon by Neal Stephenson goodreads my review
The Oedipus Trilogy by Sophocles, new translation by Bryan Doerries goodreads
God’s’ Dog: Monster a graphic novel by Jonathan Pageau and Matthieu Pageau goodreads
There Is No Antimemetics Division by qntm goodreads

In our ADHD TikTok world short fiction is a godsend. I read Cold Nights of Childhood almost in one sitting and proclaimed to myself that all books should be either 100 pages or 1000 pages with nothing in-between. Reaching for short story after, I inhaled Borges’ Ficciones. Now I’ve started on Michael R Katz’ promising new translation of Brothers Karamazov but plan to work in some shorter morsels on the side.

I didn’t directly seek to read religious fiction but the theme has enriched my reading. Tolstoy and Dostoevsky are obviously evangelists. But what is the theology of Miller’s Circe? Why is the source of Özlü’s heroine’s irrepressible zest for life?

My Favorite: Qntm’s There is no Antimemetics Division. Pulpy yet geniuinely novel. And we live in a time of intense conflict in both the memetic and antimemetic spheres. Likely the most underrated of the works here.

Nonfiction

The Sassoons: The Great Global Merchants and the Making of an Empire by Joseph Sassoon goodreads
The Inheritance of Rome: Illuminating the Dark Ages 400-1000 by Chris Wickham goodreads
Super-Infinite: The Transformations of John Donne by Katherine Rundell goodreads my review
How to Be a Founder: How Entrepreneurs Can Identify, Fund and Launch Their Best Ideas by Alice Bentinck and Matt Clifford goodreads
Six Easy Pieces: Essentials of Physics Explained by Its Most Brilliant Teacher by Richard Feynman goodreads
Building a Second Brain: A Proven Method to Organize Your Digital Life and Unlock Your Creative Potential by Tiago Forte goodreads

How did the catholic church get so successful? People take it for granted that the church was a middle ages thing when in fact operating a vast NGO in a time of anarchy is one of the strangest things that ever happened. Why did anyone outside of Italy listen to the popes? Wickham doesn’t focus on this question but can’t help answering it (in as few words as possible: their alliance with the Carolingians in the 8th century). This is important as context for Henrich’s WEIRDest People in the World, because this is ultimately how Europe got so WEIRD. What happens when a penniless yet devout Baghdadi jew strikes it rich in Mumbai? What made it possible? What motivates his descendants? Did anybody every stop to conisder it may have been wrong to be pushing opium on China?

The Sassoons and Super-Infinite were also excellent simply at providing slices of life that remind one how huge the world is.

My Favorite: Rundell’s Super-Infinite. Those were different times indeed and yet Rundell can paint them in vivid detail. Goodreads review here.

Religion

Strange Rites: New Religions for a Godless World by Tara Isabella Burton goodreads
Breathing Under Water: Spirituality and the Twelve Steps by Richard Rohr, OFM goodreads
The Life of St. Benedict by Gregory the Great, with translation and commentary by Terrence G. Kardong, OSB goodreads
The Rule of Saint Benedict by Benedict of Nursia goodreads
The Problem of God: Yesterday and Today by John Courtney Murray, SJ goodreads
The Language of Creation: Cosmic Symbolism in Genesis by Matthieu Pageau goodreads
Tattoos on the Heart: The Power of Boundless Compassion by Gregory Boyle, SJ goodreads
The Jesuit Guide to (Almost) Everything: A Spirituality for Real Life by James Martin, SJ goodreads

Tyler Cowen has predicted that the most important thinkers of the future will be religious ones. But what does that look like? Each book here proposes a set of values and spiritual epistemics. Although I leaned heavily on Catholicism due to my background, the catholic perspectives presented here were extremely diverse.

My Favorite: Tattoos on the Heart demonstrates a value system that is clearly superior to mainstream culture’s. If we imagine that the right wants to build a big wall around the center to protect from the periphery and the left wants to burn the center to the ground, Boyle wants to bring the periphery into the center to make a greater whole. And he is ready to devote his life to it with no hope of material reward. Boyle made me deeply question my own life choices and motivations.

…but if you are skepitcal of the importance of religion, it’s worth looking at Strange Rites by Tara Isabella Burton. Responding “No religion” on a survey doesn’t actually mean a person has “no religion.”

Buddhism

The Science of Enlightenment: Teachings and Meditations for Awakening Through Self-Investigation by Shinzen Young goodreads
The Mind Illuminated: A Complete Meditation Guide Integrating Buddhist Wisdom and Brain Science by Culadasa goodreads
Dependent Origination and Emptiness by Leigh Brasington goodreads
Spaciousness: The Radical Dzogchen of the Vajra-Heart by Longchenpa, translated by Keith Dowman goodreads
Seeing that Frees by Rob Burbea goodreads

Can we end suffering for ourselves and for all sentient beings? Buddhism says yes, and crucially, we can do it in a life-affirming way. Important if true!

In addition to these books I’ve been spending a lot of time learning both from Kati Devaney and Michael Taft at the Berkeley Alembic, and from Toby Sola via his digital sangha Brightmind. Toby and Michael were both students of Shinzen Young. Interestingly they’ve both published there own reading lists! Toby’s is here and Michael’s is here

My Favorite Young’s The Science of Enlightenment paints a compelling picture of the commonalities among the world’s mystical traditions, arguing that there’s a kind of “spiritual physics” that all traditions provide windows into.

Memoir

Pieces of the Action by Vannevar Bush goodreads
The Pathless Path: Imagining a New Story for Work and Life by Paul Millerd goodreads
Greenlights by Matthew McConaughey goodreads
Free Spirited: How My Daughter Healed Me From the Afterlife by Suzanne Falter (my neighbor) goodreads

My Favorite Greenlights was great entertainment and demonstrate how far you can get with a great attitude. I listened to McConaughey narrate his own audiobook which adds an additional layer of fun.

Review: Week of September 10

Wed, 20 Sep 2023 00:00:00 +0000

I’m trying to understand Friston’s free energy, and the predictive processing theory of neuroscience more generally. Very loosely, predictive processing states that instead of the usual model of “brain collects information from senses, analyzes it, and decides on actions” instead the brain constantly simulations, which it updates based on errors that arise when it predicts sensory data. Free energy serves as an objective function for this updating. Minimizing free energy punishes a) prediction inaccuracy and b) deviation from priors, providing an approximate solution to Bayesian learning. Although framed as learning, Friston also claims to explain allostatic behaviors as falling out of the predictions. I’m a little hazy on how that works but it seems to essentially include homeostasis in the prior. The theory explains important data from many domains of neuroscience, including why there are so many reverse connections in the brain. Reading materials:

Scott Alexander’s (2017) review of Surfing Uncertainty [goodreads]. The best short article on why this is interesting.
Scott Alexander (2018) “God help us let’s try to understand Friston on free energy” here.
Friston explaining it himself (2017). [youtube]
Shamil Chandaria, “The Bayesian Brain and Meditation.” (2022) [youtube]
Ruben E. Laukkonen and Heleen A. Slagter (2021). “From Many to (N)one: Meditation and the plasticity of the predictive mind.” [url]

Why both countercultures failed by Matt Arnold, part of David Chapman’s Meaningness project. I hadn’t realized how much the evangelical movement of the 1980s and 90s owed to the hippie movement and is arguably another offshoot. Arnold describes them here, basically claiming (along meaningness lines) that the hippies were monists and the evangelicals were dualists. In brief: “Identity is not community, although the countercultures often confused the two.” These countercultures gave way to the subcultural mode. John’s take, as of 2023 it sort of seems like there are again two major countercultures, “movement progressivism” (pejoratively called “wokeness”) and “MAGA.” Again they are mostly identitarian with community tacked on haphazardly. Movement progressivism seems a bit more coherent. My expectation is these won’t last forever but, like their predecessors, will be highly influential.

I’ve been doing Jenny Turbell’s Monday night series on the four immeasurables (also known as the brahmaviharas) at the Berkeley Alembic. Doing these practices makes me feel very warm and fuzzy inside, strongly recommended.

Currently Reading

Seeing that Frees by Rob Burbea [goodreads]. Michael Taft’s favorite meditation book, also recommended by Nick Cammaratta. “When you find yourself feeling somehow imprisoned in a situation, or grasping unhelpfully at something that is not present, see if you can look within and around you in a way that ‘deconstructs’ your world into its aggregate appearances of sights, sounds, smells, tastes, physical sensations, and mental and emotional experiences.”
Breathing Underwater by Richard Rohr [goodreads]. The 12 steps from a mystical Christian perspective. This is sort of a collection of musings, I think I needed more context about how 12 steps works in practice but much of it seemed like useful perspective. Also made me think about the vitality of Christianity. “We often given a bogus version of the Gospel, some fast-food religion, without any deep transformation of the self; and the result has been the spiritual disaster of “Christian” countries that tend to be as consumer-oriented, proud, warlike, racist, class conscious, and addictive as everybody else-and often more so, I’m afraid.”
Pieces of the Action by Vannevar Bush [goodreads]. How we set up our defense research to win WWII.
The Pathless Path: Imagining a New Story for Work and Life. Paul Millerd. I appreciate that Millerd was able to break out of one type of frame that was imprisoning him (careerism). But so far (I’m about halfway through) he still seems to be a victim of millenial narcissism. Our culture is failing badly in the meaning department.

Random links

Scott Alexander posted this book review contest entry for Zuozhuan. Society was in decline and people mostly knew it. This is a Chinese equivalent of post-Roman empire nostalgia.
Do progressives disparage one-parent families? MR post here. My hot take, children are much better off with two parents, much of the race gap is explained by single parenting rates. We can find compassionate ways to close this gap.
“The number of legal abortions has held steady, if not increased, nationwide since 2020, our colleagues Amy Schoenfeld Walker and Allison McCann reported today.” [MR post]
Shinzen Young’s equation for pain, suffering, and purification [twitter]

Updates

Gary’s stelfonta injections succeeded for the tumor on his leg but not for the one on his ear. We are going to try again tomorrow. The steroid he had to take during this treatment made him really miserable and caused him to pee on the floor many times which he’s almost never done before.
I finally decided to defeat Ganondorf in Legend of Zelda: TotK. It was fun but hopefully I can replace this compulsive activity with more accretive compulsive activities.

Review: Week of September 3

Thu, 14 Sep 2023 00:00:00 +0000

Late posting!

Robin Hanson claims “escalating signaling” cuts fertility, fairly pessimistic about restoring fertility.

Austin Vernon thinks brick thermal storage might actually already be competitive for industrial use cases, enabling 100% renewable energy in situations where you just need heat.

I did a retreat with Toby Sola’s Brightmind on Expansion and Contraction. It reminded me of a lot of the great insights Shinzen shared in his book The Science of Enlightenment. Basically the idea is that as you are able to see the nature of sensations arising, you can see closely that they are always moving (this is called flow). In fact, much of sensation can be thought of as expansion and contraction (waves of sensations that are arising and passing). This is key to maintaining equanimity even with very difficult sensations.

I looked at a bunch of literature on contemplative neuroscience. I was sort of just cruising through, there an example might be this one showing meditators perceived pain as less unpleasant.

Review: Week of August 27

Tue, 05 Sep 2023 00:00:00 +0000

Reading

Contemplative Neuroscience. I had a really cool IRL conversation with Kati Devaney at the Alembic about contemplative neuroscience and she sent me on a few reading rabbit holes. One was a Substack post she coauthered with Sasha Chapin about researching enlightenment. One big point they made was that the research is mostly either extremely light interventions on naïve target groups (mostly referring to Kabat-Zinn’s Mindfulness Based Stress Reduction, or MBSR), or research into deeply enlightened people. But what would be really cool is uncovering the journey from naïve to fully awakened! One banger:

Some interesting questions:

Which practices are most effective for which people?
What factors accelerate meditation learning?
Can we create useful reference states?
How can we prevent adverse effects of meditation?
Is there one kind of awakening or multiple kinds?
Are there predictable effects of awakening?

She also got me curious about the neuroscience of meditation, and what kind of biomarkers we can look at while trying to understand it. A few random links

Jhourney is a startup she’s involved in that I’ve tweeted about before, they are looking at EEG markers of the jhanas (among other things).
There is some evidence that mindfulness increases gray matter: paper
Kati has a paper showing that meditation induces lower default mode network relative to dorsal attention network

In general I find this area really interesting but a lot of brain signals have poor signal-noise ratio which can make it hard to study.

Claim: Social mobility is unchanged in England for 400 years. Gregory Clark wrote a paper about the inheritance of social status in England using a population of 422,374 English people through 422 years of history (1600–2022). He gets some really wild results. He finds that social mobility was unchanged during the entire period. Instead, social status seems to be strongly inherited. Even 4th cousins have correlated social status. This heritability is highly persistent, for example father-son correlation is not diminished when the father dies early in the son’s life. He has evidence to rule out direct wealth transmission as a mechanism; for instance having many children (meaning each child inherits less) does not affect heritability. Part of the reason the effect is so strong is that mating is highly assortative. Interestingly, there is no evidence of a “poverty trap” or “elite lock-in”: social mobility is symmetric and identical across the spectrum. See also podcast here.

Does China need more consumption? Paul Krugman seems to think so. As does the WSJ. When I heard this claim I was a bit puzzled: How would consuming scarce resources help an economy recover? I felt vindicated when Tyler Cowen posted similar skepticism with followup here. But the best explanation comes from Scott Sumner. In short, people are talking past each other. Obviously a better scenario is one in which Chinese consumers are able to consume more, but the more constructive conversation is to ask what sort of policy changes are under consideration and why China’s leadership is unable to easily chart a path through. Here’s how he frames it:

Here we can see what’s really going on: The CCP wants to allocate resources based on favoritism. Unfortunately, its favorite investments are not actually good investments, at least not in the sense of producing more than they cost. It is also unwilling to significantly devalue the RMB, which might be able to sweep at least some of the mispricing issues under the rug. As a result, it needs to allocate a lot of resources to its failing favorites and a lot of resources to propping up the RMB and there’s not much left either for good investments for consumption. You could say they need to allow more consumption or you could just say they need to stop misallocating massive amounts of capital. It’s all academic anyway because they are committed to digging in their heels. Something will have to give, but they may be able to stagnate and kick the can down the road for a long time.

Media

Interview between Tara Isabella Burton and Bishop Barron. I didn’t realize she was a practicing Episcopalian! I found her book on the religiously unaffiliated (Strange Rites) a bit uneven but it’s one of the only books I’ve seen about how people in the new “unaffiliated” bloc actually have a lot of quasi-religious beliefs and practices. The book made it seem like she was unaffiliated herself. Maybe her conversion is recent. One interesting exchange: Barron asked her how she’d start a religious conversation with a “none.” She says social justice people would be easy to talk to, there’s a lot in common! It’s the people who believe in “doing whatever I want” as the highest value that are really hard to engage.

Random Links

A Reconstruction of ancient Tenochtitlan here
Vitalik on Community Notes (they’re pretty neat!) here

Review: Week of August 20

Sun, 27 Aug 2023 00:00:00 +0000

Reading

The Comfort Crisis (Money Mustache Blog). Michael Easter is a recovering addict who believes the lack of hardship and abundance of material comfort is generating a psychological crisis. When nothing is wrong your brain will still look for stuff to be wrong. This post doesn’t investigate deeply but it seems like an important truth about the modern world. Venkatesh Rao has covered similar topics. Reminds me of this paper claiming that people’s sense of safety and abundance is not correlated to actual experiences:

https://twitter.com/DegenRolf/status/1694738097953505504

Austin Vernon (email to Tyler Cowen) on the extent of the Inflation Reduction Act’s subsidies. Some are completely uncapped. It would seem solar factories are not a long term investment: “Another wrinkle with solar factories is that they depreciate at an extreme rate. Usually they are obsolete after 2-3 years and need to shut down. So there is a real possibility at the end of the law that you see a massive drop in solar deployment because the inflated supply chain will have to rationalize to non-subsidy conditions and then our factories will not have the revenue to upgrade to the next generation. That will then kill the supplier ecosystem, etc”

Bluesky writes about their custom feeds. I think this is a really cool approach but I’m concerned that there’s just not much content on Bluesky. I would die to have this on Twitter.

Random Critical Analysis (2015) writes about homicide and race. Socioeconomics and demographics don’t explain group differences well, but single parenting does. This finding mirrors a fascinating thread by Alice Evans on the links between falling marriage and increasing inner city violence:

https://twitter.com/_alice_evans/status/1616804457043623943

I owe @nickcammaratta for helping me understand the Buddhist idea of “attachment.” I always thought, “How is it at all wholesome to not be attached to, e.g., my daughter?” But the problem isn’t loving your loved ones, it’s a specific mental clenching motion that gets translated as attachment.

https://twitter.com/nickcammarata/status/1695546470819127529

…incidentally attempts to popularize deeper meditative states are starting to get more attention, as in this Vox writeup by @OshanJarow. Apparently some are calling this “spirit tech.”

I enjoyed this critical review of Acemoglu and Robinson’s book Why Nations Fail. It excoriates the book for failing to produce quantitative evidence in favor of the thesis that differences in the wealth of nations result from differences in the extractiveness of institutions. Instead it focuses on anecdotes. Although this critique is valid, I nevertheless got a lot out of the book when I read it several years ago. For example, the section on the Kongo they explain how precolonial Kongo had such extractive elites that it wasn’t even worth farmers’ while to employ the plow in agriculture, since any surplus would be expropriated. Obviously this one case doesn’t prove generally why some nations succeed and fail. But before reading this book I had had a naïve view that central institutions generally favor and support progress. It’s clear that they can do that, but at many times and places they have likely worked against it.

Tyler on what Harvard can learn from the Olive Garden. “Wealthier institutions or establishments attract a mixed customer or user base only when they give up cultural control.”

Researchers are conducting a human challenge trial to fight malaria! I applaud the bravery of the participants.

Updates

I started using @malcolm_Ocean’s app intend.do. It seems useful so far. Good framing. Not a todo app (no concept of a todo that lives more than a day)
Our dog Gary has mast cell tumors :-(. We have scheduled a chemo approach called stelfonta that seems to have a high success rate at treatment.
This week on BrightMind I did the lesson pack on the “Concentration Algorithm.” The idea is to check in halfway through a sit, decide if you have a reasonable level of concentration, and if not identify what’s distracting you and shift your focus to that part of sensation-space. I found this useful but I kept being really interested in my original concentration object so rarely changed.

AI Alignment Is Already a Problem for Humanity

Tue, 30 Aug 2022 00:00:00 +0000

Stable Diffusion, Prompt: “Benevolent robot overlords creating a Golden Age on Earth. Soviet propaganda poster.”

At Stitch Fix, I lead the teams that built our in house recommender systems. These systems were incredibly powerful, and misunderstood by leadership. We were facing the AI alignment problem head on. That problem is getting larger everyday, engulfing the tech industry and society at large. AI alignment is not a problem for the future: It’s already here.

Recommender System Alignment

Three observations, taken together, explain why recommender system alignment is already a problem:

Well-tuned recommenders are extremely competent. When we tested replacing human expert stylists with mature algorithms, the algorithms typically outperformed. Often they outperformed stylists by a wide margin across many measures.
Recommender systems give you exactly what you ask for. When we tuned the system to maximize number of items sold per box, we sold a lot of items. When we changed it to maximize revenue, we could drive a lot of dollars. Decisions about the algorithm’s objective function were crucial to business operations, and changing configuration could create enormous operational headaches.
It is hard to align recommender systems with our true values. Due to the way our systems were configured, it was easy for us to maximize on outcomes that were tied to a single item, such as sales or revenue. It was hard to directly train on the thing we really cared about: Valuable client relationships. Sometimes humans disagreed with what the algorithm was doing and would add hard coded rules and constraints to solve for perceived problems. These typically made things worse. The cost to algorithm efficacy was large, while underlying problems weren’t actually fixed.

The combination of the above means that you must be careful what you wish for when you configure an algorithm. They are both extremely powerful, and not easy to align with human values. As a corollary, improving the alignment of these systems is the current frontier of making them more useful, even in the short term.

We might call this the Narrow alignment problem. Today’s algorithms don’t threaten to destroy the world the way a misaligned AGI might (see Scott Alexander’s review of Stuart Russel’s book Human Compatible here), but they can certainly make mischief: Selling us things we don’t need or addicting us to media that makes us emotionally worse off.™£

What to watch for

In the coming decade, this will play out across many spheres, so here are some trends to look out for:

Poorly aligned algorithms will continue to scale and cause mischief. Ham-fisted attempts to align algorithms might even make things worse. Don’t expect Twitter to wake up one day and realize their algorithm is making all its users stressed and anxious (see Scott Alexander post here, dire claims by Jonathan Haidt here and elsewhere).
Nevertheless, attempts to turn back the clock will fail. Algorithm-driven online experiences will outperform those that rely on self-curation and timelines, at least in the mass market. Setups where humans attempt to interfere aggressively with algorithmic output without improving the algorithms’ underlying objective functions will simply do a worse job.
Currently there are still specialists (scientists, journalists, intellectuals) relying on significant self curation of blogs, RSS feeds, and hand-curated twitter feeds. Look for these to be conquered by algorithms too. They may be too small for a mass market offering, but are highly influential, meaning better algorithms for them could produce high leverage in terms of intellectual output.
There will be a “Next generation” of algorithms that will be substantially more aligned. These will focus on getting better measurement of the things that matter to humans. Instruct-GPT is a great example of this (see OpenAI here). Imagine a TikTok algorithm that was specifically trained to make people feel calm and relaxed, trained on physiological sources such as EEG data.
Content will be algorithmically generated. We’re already seeing human-AI collaboration with Dall-E. It’s easy to imagine an AI reading tweets and creating stories as good or better than typical news articles. Ultimately AI can autonomously create content people want, and it will. Lessin predicts this here.

Ultimately I believe recommendation algorithms are the solution to what David Perell has dubbed the Paradox of Abundance. With so much information available, we have the tools at our fingertips to create an era of wisdom and enlightenment. But most people are drowning in the onslaught. Algorithms have proven they can tame the seas and command people’s focus. We need them to use that power for good.

Cultural Deep Dive: Scotland 🏴󠁧󠁢󠁳󠁣󠁴󠁿🏴󠁧󠁢󠁳󠁣󠁴󠁿🏴󠁧󠁢󠁳󠁣󠁴󠁿

Tue, 05 Jul 2022 00:00:00 +0000

Allan Ramsey’s “Lost portrait” of Charles Edward “Bonnie Prince Charlie” Stuart, currently on display in the Scottish Portrait Gallery.

File under: Context is that which is scarce.

I recently spent two weeks in Scotland. Inspired by Tyler Cowen’s “favorite things” posts (especially this one on Scotland), here are notes on what resonated with me:

Historical context: Scotland was prominently defined by efforts to remain independent. Wars of Scottish independence [wiki], establishment of Presbyterianism [wiki]. James VI and I [wiki]; he did not miss his homeland! Jacobite uprisings [wiki]. I did not study the recent independence referenda in detail.
Art: Henry Raeburn, Allan Ramsay, design and architecture by Charles Rennie MacIntosh, tartans. This book was helpful.
Authors: Classics (Robert Louis Stevenson, Walter Scott), Moderns (Alistair Gray). Scott has an enormous memorial in Edinburgh. Lots of crazy folktales too, like this one about the formation of the Hebrides and Iceland.
Music: Traditional (Loch Lomond, this album of folk songs about Bonnie Charlie with excellent oral commentary), Contemporary (Boards of Canada, Belle & Sebastian).
Philosophy: Scottish enlightenment especially David Hume, Adam Smith. Both have significant monuments in Edinburgh. Smith has an anti-slavery quotation on his tombstone.
Archaeology: Kilmartin Glen is a great day trip. Standing stones, cairns, and Dunadd.
Culinary: I ate and greatly enjoyed many pies and scones. I found the haggis unmemorable. Excellent Indian takeout is readily available, I’d recommend Slumdog Delivered and Tuk Tuk. For an upscale experience, my wife went to Kitchin with her family and raved about it.
Whisky: I barely drank any scotch, but am partial to Islays such as Lagavulin (peaty) or Bruichladdich “clasic laddy” (not peaty).

Fun fact: Kilts are not ancient attire in Scotland. Contra Braveheart, William Wallace’s men would have worn similar garb to their English adversaries. The “small kilt” widely known today was invented in the 1720s by a Quaker Englishman, Thomas Rawlinson.

On Cultural Immersion

The returns to researching culture compound. For example in the novel Lanark the protagonist attends the Glasgow art school where Macintosh designed the library, which is mentioned, and some sections are incomprehensible without awareness of the Scottish reformation. A whole section of the Scottish portrait gallery is dedicated to Jacobitism. The famous refrain “you take the high road and I’ll take the low road and I’ll be in Scotland before you” is about the soul of an executed Jacobite zipping home.

Is the university system doomed?

Mon, 10 Feb 2020 00:00:00 +0000

David Perell’s essay “What the Hell is Going On” provides a thought provoking review of how technology and the information age continues to disrupt the most time honored information markets: advertising, education, and politics. Much of his discussion is inspired by the hypothesis of Martin Gurri that the information age has permanently diminished the authority of existing elites. Perell is willing to go out on a limb a bit in its quest to “solve for the equilibrium” (as Tyler Cowen would put it).

This includes the bold claim that the university system is doomed to be replaced by online courses and new types of signaling. According to Perell, the proliferation of low-cost and high-quality online education spells the end of universities, or at least the middling ones.

In his recent book The Case Against Education, Bryan Caplan lays out the opposite case, and it’s worth examining the differences. Caplan begins with the point that the correlation between college graduation and later income appear substantial, clocking in at 30% (correcting for ability bias) even for “impractical” majors like Philosophy and Art History, and much higher for hard sciences and engineering.

He notes that this link between education and income can be ascribed to a mix of three sources:

Social capital added. The “improvement” of the individual via training and coursework.
Pre-existing skill. The tendency of the competent to enroll in college.
Signaling. The tendency of employers to prefer candidates with better credentials.

Caplan argues that signaling accounts for ~80% of the premium for college graduates. He provides research to back that number up, but he also asks a question that’s hard to ignore: “Would you rather have 4 years of world-class education at Princeton and no degree, or a Princeton degree without the coursework?” I for one am fairly convinced by the argument that the latter would make you better off financially than the former, even if you believe that the college experience itself is a priceless spiritual journey.

Bryan goes on to argue that the signal provided by a degree is valuable to employers, because it connotes a package of three traits: intelligence, conscientiousness, and conformity. A technology company hiring a philosophy major as a project manager isn’t counting on them to produce well-argued treatises on Locke, or make use of any information about philosophy whatsoever. Rather, the company is counting on them to be able to reason well about novel situations (intelligence), ensure that their work is produced on time and to a high standard (conscientiousness), and that they will “get with the program” and act in accordance with company values and incentives (conformity).

If college degrees are truly 80% signalling, then even the best online courses or MOOCs could not replace them. Under this hypothesis, universities do not owe their success in the modern world to their status as gatekeepers of information. Rather, owe it to their reputation for educational programs whose completion signals intelligence, conscientiousness, and conformity. As Bryan points out in later chapters, the conformity signal in particular provides a catch-22 to would-be disruptors: Doing anything other than going to a four year college would be non-conformist.

So what’s the prognosis right now? There are predictions circulating that the demand for 4-year colleges will diminish starting around 2026, but these have to do with a drop in the number of high school graduates rather than a reduction in the proportion of graduates attending. Perell claims that mid-sized liberal arts schools are already beginning to decline, but it’s not clear that the institution more broadly is suffering.

I don’t doubt that there is opportunity in the education space: college provides little in the way of useful job skills, and as a signaling device, it is both costly and inequitable. Delivering content and university-like interactions should be feasible in an online format. But delivering on the strong signal currently sent by attending a selective four-year institution, and especially signaling conformity, is an uphill battle for any disruptor. We’ve already seen some experiments that have produced quality content but failed to make a dent against 4-year degrees (e.g., Khan academy, edX) and a generation of boot-camps that have been hugely useful to a niche set of students, but may not be able to scale to the level of universities (e.g., Lambda School, App Academy). There’s still an opening for new entrants, and for an ensemble of these disruptors to coalesce into a better route to success for new high school graduates. But the university system is a hardy weed with deep pockets and a centuries-long track record. It won’t be uprooted overnight.

Correction 2019-02-10: I originally had listed Philosophy’s premium as ~20%, in fact Caplan lists it as closer to 30% after correcting for ability bias. I also added the graph of adjusted premia.

Bibliography

Caplan, Bryan (2018). The Case Against Education: Why the Education System Is a Waste of Time and Money. [amazon]

Fox, Justin (2019). “The Coming College Enrollment Bust.” Bloomberg Opinion. [link]

Gurri, Martin (2018). The Revolt of The Public and the Crisis of Authority in the New Millennium. [amazon]

Perell, David. “What the hell is going on?” [link]

Perell, David & Forte, Tiago. “Tiago Forte: The Future of Education.” [link]

David Perell on the Paradox of Abundance

Wed, 05 Feb 2020 00:00:00 +0000

David Perell recently published a long read article on the new age of democratized digital media we are entering. Although he gives a history of the last century of media, his ultimate message centers around how we should be consuming information in the digital era. The most important concept he shares is the Paradox of Abundance. Briefly, we have access to more information than ever before. Because there is so much information out there, information purveyors are in a desperate zero-sum competition for eyeballs. They respond by creating urgency and emotional hooks around articles and information that aren’t actually very important. This confronts consumers with a tsunami of clickbait and emotional appeals, which either sends us running from false crisis to false crisis, or makes us check out from media altogether. This means that people who can focus on truly useful and enlightening content are richer than ever before, while the median user is worse off, tugged in all directions by emotionally captivating but ultimately non-enriching content.

This feels most real to me whenever I log onto Twitter. The transcendent moments on Twitter are truly transcendent. Rarely do I spend 20 minutes on Twitter without encountering some new perspective or piece of information that I hadn’t considered before. But the tsunami can be overpowering: I often spend far longer than I had planned following trails of seemingly alarming scandals or political trends. Twitter’s algorithm accelerates this, bubbling the most alarming, shocking, or polarizing content to the top of the feed. My Feedly experience could not be more different: Since I mostly follow only academic blogs or writers I deeply respect, my feed is a bit more boring but also has a much higher signal-to-noise ratio. It’s also less addictive.

Perell draws an analogy to eating right that feels particularly apt. Ultimately this may be the sort of thing our culture needs to adapt to. Just as abundance in food resulted in the evolution of cultural norms around eating right, we may need to develop cultural norms around “informing” right. My feedly is more like a quinoa bowl, whereas a twitter binge can leave me feeling like I just ate a bagful of tootsie rolls. He even mentions writing as analogous to cooking at home, another regularizing function that can improve diet.

There is one question Perell leaves unanswered: What is it all about? Is there an individual duty around online engagement as a sort of civic service? Or would we all just be better off unplugging and reading books? Either way, it is clear that information abundance is here to stay, and passive consumption of whatever passes across our screens is no longer an acceptable strategy.

Is malinvestment in advertising killing D2C ecommerce?

Tue, 04 Feb 2020 00:00:00 +0000

Alex Taussig of Lightspeed writes that there is a typical playbook for vertical ecommerce brands: “(1) Launch a “hero” product that provides branded product innovation, (2) at high gross margins, (3) enabled by an owned & operated e-commerce experience, (4) with high marketing efficiency, usually driven by viral or word-of-mouth campaigns.” (Taussig 2019). The filings from Blue Apron and now CSPR Casper that item (4) on that list, marketing efficiency, can prove elusive as these brands attempt to scale to the level of the expectations they’ve set with investors.

When Blue Apron released their S-1, I was working on Stitch Fix’s own S-1, and so I immediately found their filing interesting. I looked into the numbers that I knew to be important for a subscription service: customer net present value and acquisition cost. From what I could glean, NPV on a customer basis actually looked fairly high; my guess was around $200 assuming a reasonable discount rate. But marketing costs were astounding: ~$144M in 2016, and with Q1 2017 marketing annualized to over $242M. We can’t say for sure how many new customers they paid for with that. If we use the difference in number of customers at the beginning of the period and the end (this is conservative due to the effect of churn), the implied blended CPAs are $320 in 2016 and $385 in Q1 2017. That already implies a money losing proposition, but it obscures how bad things really were. The marginal acquisition cost is typically much higher than the blended cost, because advertising does not scale up efficiently. It’s impossible to know exactly what the ratio is between blended and marginal. In my experience, 2x is a fairly conservative estimate. This means it is quite possible that at the margin, APRN was largely paying $700 or more to acquire a customer worth only around $200. When I saw that, I knew they were toast right away.

Now we have a new filing from Casper with a similar trend. Although Casper claims to establish deep customer relationships, they have a repeat rate of only ~14%, suggesting that we can safely treat marketing acquisitions as one-off. AOVs were $710 in the first nine months of 2019. Revenue plus discounts (which are included in AOV) were $392M, with COGS + refunds of ~$237M, for a profit margin of 39%, or $277 per order. Like APRN, this value per acquired customer is quite impressive, and may be slightly understated if customers are coming back unprompted to buy more big ticket items. Using the above numbers, we can assume 552k acquisitions on a sales cost of $114M, implying a blended CPA of $202. This may seem like a reasonable figure until, again, one considers that the marginal cost is likely to double the blended figure or more. In this case, again, this means that Casper may well be paying a $400 marginal CPA to acquire $277 of gross margin. This is not quite as dire as the Blue Apron situation, but far from comfortable. We can’t tell from this report how much value there is in those repeat orders, but they are absolutely relying on them to make their strategy work.

Direct to consumer brands succeed by providing good margin on product via vertical integration and efficiencies in customer acquisition via strong branding. Often marketing is easy in the early days, when word of mouth provides costless customer acquisition. But $APRN and $CSPR show that acquisition efficiency does not come for free for these brands, especially once they attempt to scale past the growth level supported by referrals. Once organic and word of mouth peter out, the temptation is strong to dump money into paid acquisition well past the point of efficiency. Casper has already had to reduce its offer price below its recent round; this week we’ll see the public market’s judgment.

Bibliography

Form S-1, Blue Apron, filing date 6/1/2017 [link].

Form S-1, Casper, filing date 1/10/2020 [link].

Taussig, Alex. “Drinking from the Firehose #154.” 1/13/2019.

How a dart-throwing monkey can beat the S&P 500.

Mon, 03 Feb 2020 00:00:00 +0000

Main points:

A monkey throwing darts can beat the S&P 500 on a risk adjusted basis.
This is because cap weighting has historically been a poor portfolio construction strategy.
Passive portfolios aren’t fragile. Random perturbations are generally benign as long as they tilt toward the so-called “smart beta” factors such as value and small cap.

I recently read a paper that changed the way I think about passive indexing.

Cap weighting (short for market capitalization weighting) has long been upheld as the gold standard of passive investing, from Bogle’s Common Sense on Mutual Funds to Warren Buffet’s famous bet that hedge funds could not beat the S&P 500. Market capitalization is the implied value of all of a given company’s shares if you added them up at the market price. Indexes like the S&P 500 weight the importance of stocks based on market capitalization. So at ~1.3T, Apple, would receive over twice as much weight as Facebook, whose capitalization is ~600M. In theory, if markets are efficient, a cap weighted portfolio is perfectly efficient because it reflects the allocation of market capital as a whole. If some other weighting of assets could produce higher returns, one would imagine that the market as a whole would shift into those assets, restoring equilibrium.

As it turns out, however, it is easy to outperform the index. It is so easy that a monkey can do it. Arnott et al (2014) investigate several alternative portfolio allocation strategies that have been proposed in the literature and appear to outperform cap weighting. Shockingly, they find that these strategies work just as well, or often even better, when they are inverted. This is strange enough that it’s worth spelling out clearly. In the investment literature, researchers commonly identify some set of characteristics that might indicate that a stock will produce outsized returns. It is common practice to test these hypotheses by looking at the simulated performance of portfolios constructed using those characteristics at different points in the past. For example, one might imagine that stocks with high 5 year average earnings would be strong performers. In fact, these stocks do have higher returns: Arnott et al. estimate 11.18%, as opposed to 9.66% for a U.S. cap weighted index.

The authors make an astonishing revelation by flipping the strategies on their head. For example, they invert 5 year average earnings strategy, simulating portfolios filled with the stocks that earned the least. Common sense implies that the inverse of a winning strategy should be a losing strategy. Instead, the inverse index actually performs better, with returns of 14.38% and an improved Sharpe ratio (the ratio of expected return to variance in return). The authors show similar results for about a dozen other strategies, using two different techniques for inversion. This is almost spooky: If a strategy is good, how can its opposite also be good?

Superficially, the answer is clear. What do the 5 year high earner portfolio and 5 year low earner portfolio have in common? They are both not cap weighted. Cap weighting is such a bad way of selecting stocks that it is beaten by almost any alternative strategy! To drive this point home, the authors simulate a dart-throwing monkey, picking portfolios of 30 stocks completely at random, and show that these portfolios also beat the cap-weighted index: The monkey portfolios produced returns of 1.6 percent in excess of the index, with a better Sharpe ratio (0.33 vs. 0.29). Looking more deeply at why this is the case, the authors show that most of the excess returns they observed across all these strategies are explainable by an enriched Fama-French model. This is a widely discussed model in the finance world that assumes that investors are compensated for certain types of exposure, which are known as factors. In this case, the portfolios were typically weighted toward value and small-cap stocks. These anomalies are often referred to as Smart Beta, because the Fama-French equation uses the symbol beta to denote the coefficients for the factors.

I’ve been concerned in the past that smart beta indexing is inimical to the philosophy of passive investing. Trying to outsmart the index exposes your portfolio to the precise stock-picking scheme used by a bespoke ETF. Because of this, I had always been suspicious of Betterment’s decision to tilt the portfolio toward value, and the recent underperformance of value seemed to bear out my bias. But if even a monkey can pick a good portfolio, it implies that the effects are quite robust to the precise method of stock picking so long as you’re not actively trying to weight against the smart beta factors.

The literature is not clear on exactly how or why smart beta outperforms the cap weighted index. This outperformance is enigmatic since it seems contrary to EMH. Some argue that they result from investor irrationality or the perverse incentives of fund managers (e.g., Baker et al. 2011), while Fama and French themselves seem to believe that this reflects efficient pricing of hidden risk factors (Fama & French 1993). There is good reason to doubt that the past performance of these factors might not be consistent in the long run, since the explanation is not clear. Nevertheless, this paper made me stop worrying and love the tilt.

Resources & Bibliography

Arnott et al (2014) “The Surprising Alpha from Malkiel’s Monkey and Upside-Down Strategies.” The Journal of Portfolio Management. https://thereformedbroker.com/wp-content/uploads/2014/11/jpm_summer2013_rallc.pdf

Arnott et al. (2016) “How Can ‘Smart Beta’ Go Horribly Wrong?” https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3040949

Baker, Bradley, and Wurgler (2011) “Benchmarks as Limits to Arbitrage: Understanding the Low-Volatility Anomaly” Financial Analysts Journal. https://doi.org/10.2469/faj.v67.n1.4

Bogle, John (1999). Common Sense on Mutual Funds.

Fama and French (1993), “Common risk factors in the returns on stocks and bonds.” Journal of Financial Economics.

Fama and Thaler discuss efficient markets, including the Fama-French 3-factor model. https://www.youtube.com/watch?v=bM9bYOBuKF4

What is a tech worker worth?

Fri, 31 Jan 2020 00:00:00 +0000

As a tech worker I’ve always been bewildered by the high compensation on offer for technical workers in the Bay Area. If you’re not in tech and you want to see the gory details, see this site, which is quite accurate as far as I can tell.

I’ve sometimes wondered if the whole thing is some sort of Ponzi scheme or bubble: Employers don’t want to feel like they’re getting “second best” labor, so the compensation rises past a level that might be reasonably justified on an ROI basis. The whole song and dance is funded by VC money and a few very lucky big ticket exits that keep the party going.

Academic economists would typically point to an efficient market perspective: Employers act in rational self interest, overpaying for labor would be a very expensive mistake, and so the prices probably simply reflect workers capturing a larger portion of their contribution. But could we really be worth so much?

It turns out we can actually answer this question. Dimmock et al (2019, pdf) were able to do a causal analysis using random assignment provided by the H-1B lottery (h/t Alex Tabarrok). When you hire a worker on an H-1B visa. There’s a lottery in Spring, and you have a 50/50 chance of winning. If you win, the visa takes effect in September. Otherwise you’re out of luck.

Maybe I shouldn’t have been surprised, but the results are extreme. Winning a single H-1B lottery raises the likelihood of a successful IPO dramatically. Each standard deviation increase in win rate (which was roughly equivalent to a single incremental win) increased the rate of successful IPOs within 5 years by 23% (rising from 6.6% to 8.1%). How much does that make each H-1B win worth? If the marginal benefit of a lottery win is approximately 1.5% of an IPO, and we assume an IPO within 5 years is worth $100M relative to alternative outcomes, then that worker is worth $1.5M. And what’s especially amazing is that this is simply value over replacement player: Typically, when a company loses the lottery, they re-open their hiring req and hire someone else. So this is actually the value of the immigrant worker over and above the domestic alternative.

I see two main takeaways here. One is, immigration is underrated. As Bryan Caplan has been shouting from the rooftops, immigrants don’t just “take our jobs,” they contribute in ways that are irreplaceable. Second, as heady as the market for tech workers may appear, it is not a bubble: At least in aggregate, tech workers are writing their own check.