Notes on 12 days in Bora-Bora, Moorea, and Tahiti

Posted on April 1, 2026 by mkevane

In early March my wife and I and another couple spent 12 days in Bora-Bora, Moorea, and Tahiti. I thought I would write up some trip notes since one thing we noticed was how bad many of the online sources had become, as they were often overwhelmed with slop. So here’s some ground-truth. First, for the budget-conscious, the total cost of the trip (airfare, meals, lodging, activities) was about $2800 per person. We stayed at Airbnb and rented cars, bought groceries together, and were 4 persons, so that probably reduces costs as opposed to 2-person costs. There are economies of scale in groceries and transport. Driving 30 minutes to a beach is spread across 4 persons instead of 2. I would encourage you to plan along those lines, and even think that 6-8 persons would be a good number. We mostly did snorkeling and hiking, and those can be even more rewarding with a few more people. Snorkeling benefits from people spotting interesting stuff and calling out to the others (which is the normal snorkel ethos anyway- if you see something amazing it’s nice to call out to people around). Ok on to specifics.

We flew out of San Francisco. Caltrain from San Jose to San Francisco. We flew on French Bee. Airfare was about $700 for the round-trip. It is a discount airline. You could buy food on it. We brought snacks and stuff. Seats were fine- a little tight. You can get even cheaper if you don’t need/care about an assigned seat. On the way back both couples were in 3-seat rows with an empty seat. So it was as comfortable as you can get outside of reclining seats. We all thought the staff and flight was excellent.

Landing in Pape’ete, we waited 4 hours at airport for flight to Bora-Bora on Air Moana. A little boring, but not much to do in a town from 5am-8am! We could have scheduled a tighter connection, but this ended up fine, since usually you can’t check-in anyway until afternoon. After landing in Bora-Bora airport motu (little island on the reef) we took the free ferry to main island, and rented a car at Avis. Very smooth. We had a fair amount of snorkel stuff, so was a little tight in the car, but we made it work. Only a 15 min. drive to the Airbnb. The Airbnb was a cool house that according to the manager had been built in Java, disassembled, and rebuilt on the beachfront lot. It had snorkeling right off the beach, and a dock going out about 100 feet. We ended up moving a table out to the end of the dock and eating dinner over the water just like the people paying $1,000 a night in the over-water bungalow hotels 😉 We sat out for several hours after dinner just watching the stars and the occasional storm and lighting come through. The view was spectacular. We shopped at a couple grocery stores just 5-10 min. away. Food is expensive – a six-pack like $15, vegetable selection limited, eggs only early in the morning, stuff like that. But we had no trouble getting plenty of snacks and cooked some really nice simple dinners (I am vegetarian so that was a small constraint that travel-mates were very kind to accommodate).

We had 3 full days in Bora-Bora. One day we took an all-day (9-4) boat trip that ends with nice lunch on a motu (small private island). We stopped at the usual 3 snorkel places: (1) fantastic coral garden; (2) shallow waist-high water with rays and green-tipped reef sharks; (outside the reef in deeper water, with super interesting coral and different sea life. On the motu we ate at picnic tables in the shallow water as rays came and begged (only way to describe it) for food, and other fish nibbled on the occasional “dropped” cucumber. It was very pleasant to walk around the small paths of the motu and see land snails, varied vegetation, etc. The boat crew put on a little show. Everything very good-humored
 The other days we snorkeled some cool beaches. Mostly we were alone or with 3-4 other snorkelers. Definitely no crowds! Matira Beach is a big sandy strip. Had to swim a ways (to the point) but there had quite decent snorkeling and saw rays each time. We then drove out to Pointe Ta’ihi at the other end of the island, which had a “manta ray cleaning station” and sure enough some other snorkelers pointed us to a 10-foot manta circling in deeper water under us. Majestic to quietly watch! There was perfectly fine parking on side of road. We drove around the island a couple times, and occasionally took a drive up the hills. A nice stop was Patisserie St. James in the morning, where can enjoy coffee croissants etc with a nice view of the lagoon. We liked Lucky House for dinner. Leaving in morning have to remember need an hour to take ferry out to airport. Filling rental car with gas was very easy.

We flew back to Pape’ete, rented a car at the airport, and then drove over to the ferry to Moorea. We had booked tickets and place for car a couple days before. Finding exactly where to go to put the car on the ferry took us 2 drives around. Basically have to slowly read signs and we missed it first time. There are a lot of entrances into various locations in the docks/ferry building, only one is for the car ferry
 But we had an approximate guess, and then when drove around second time (and not obvious can do U-turn on the waterfront road in downtown Pape’ete!) we went really slow and clearly saw the entrance. Sea was calm, so 40 minute ferry was a pleasure! Snack bar! Espresso! You could even rent bikes in Pape’ete and ride them on to ferry and then have bicycles in Moorea. The island is definitely bike-able if that is your thing. But we were happy to have a car for luggage and groceries, and to explore even further.

On Moorea we had an AirBnb right on the beach. We immediately went in for a snorkel and there was interesting coral and sea life all around. Not great, but certainly fine for a 30 minute swim. We must have done 10 such swims over the course of the 3 days including night swimming. Lagoon water perfect for swimming. The first day we went up to Temea beach, but it was windy and there was low visibility, so we drove on to Opunohu Beach. Great snorkeling there for a couple hours. Ate at a roulotte – very good food. I think we then decided to drive around the island, and we stopped at some fruit stands. That afternoon I solo hiked to ‘Āfareaitu Waterfall. I did about 6 miles- it was hot and humid, I drank 3 liters (I had brought my filter so refilled at a stream). waterfall great. Next morning I hiked it again with travel-mates.

Next day we did a half day (did we need the “show” again?) boat snorkel trip with captain/guide Alex (Alex Lagoon Tour), who was highly recommended by a friend of a friend (also named Alex!), and we all agreed that was one of the highlights of the trip. Super great captain, very chill. He took us to four places, the last a coral reserve. It was truly amazing, the way coral must have been everywhere before the die-offs started 20 years ago. We then had lunch at O2Coco, which was just fine, and then did an afternoon hike up Magic Mountain. In the heat a short but tough hike, well-rewarded by excellent views.

The next morning we took the ferry back to Pape’ete, and went to Carrefour for groceries, and then checked in to spectacular AirBnb in the hills. It had amazing views and a small pool. Was nice to be in pool water after all the ocean water. We luxuriated on the balcony into the evening, enjoying the vista. The next day we drove out to the beach at Pointe Venus. My wife and I had a long 2-3 hour swim, travel-mates did a waterfall hike. In the afternoon we went to PK18 (Plage Vaiava) and had really great snorkeling. Sea life again amazing, though a lot of the coral bleached. The next day we went to a small public botanical garden (Jardin d’Eau de Vaipahi), drove a bit more around Tahiti, and then snorkeled at PK15 (Plage publique de Toaroto).

Early the following morning we drove down to the airport. Returning rental car took 10 minutes and checking in to French Bee was super fast. Everything went really smooth, and a great flight back. Not much time difference between Tahiti and San Francisco (2-3 hours as we straddled DST), so we didn’t really have any jetlag.

Posted in Personal Kevane life | Tagged , , , , | Leave a comment

Reading Feb 2026

Posted on March 1, 2026 by mkevane

Álvaro Enrigue, Tu sueño imperios han sido
Cristina Rivera Garza, AutobiografĂ­a del algodĂłn

Christopher Priest, The Prestige

Vladimir Nabokov, Pnin

Posted in Book and film reviews | Comments Off on Reading Feb 2026

Reading Nov-Dec 2025 and Jan 2026

Posted on January 25, 2026 by mkevane
  • Charles Burns, Final Cut. Appreciate it. Leaves you with fleeting emotion. Cinema and meta pretty nice.
  • Chinua Achebe, Things Fall Apart. Things really do go rather badly for Okonkwo.
  • Javier Marias, Todas las Almas. Gave up 1/2 way through. Just going nowhere.
  • James Hilton, Lost Horizon. So male-oriented it’s kind of funny. But the framing story writing is great. The philosophy is so undergraduate though.
  • H.P. Lovecraft, At the Mountains of Madness. Started skimming at some point. A lot- a lot!- of description. Great pairing with Lost Horizon.
  • Gene Wolfe, Soldier of the Mist. Greek soldier has amnesia every morning on waking, and sees some of the gods. Lots of stories. A peculiar style. Very akin to Ishiguro’s (later) Buried Giant. If you like your fiction obtuse (in the sense of indirect)… by design… this is a master class.
  • Adam Johnson, The Wayfinder. Have read about 500 pages… tremendous. But all of a sudden, I don’t want to read the last 200 pages and ending. I’d rather have the story continue in my head.
  • Virginia Woolf, Mrs. Dalloway. So enjoyable to read and discuss with friends.
  • Vladimir Nabokov, Pale Fire. So enjoyable to read and discuss with friends.
  • Mario Vargas Llosa, Tiempos Recios. The story of the 1951 coup in Guatemala. Gripping story, straightforward read.

Posted in Book and film reviews | Comments Off on Reading Nov-Dec 2025 and Jan 2026

AI as an existential threat – Kevane preliminary draft

Posted on October 20, 2025 by mkevane
AI as an existential threat-Kevane preliminaryDownload
Posted in AI | Comments Off on AI as an existential threat – Kevane preliminary draft

“What can it do?” A living list of computational problems that deep learning/AI/neural nets can or seems likely to “do” (at varying cost and efficacy)

Posted on October 3, 2025 by mkevane
  • It goes without saying that the suite of AI-related software technologies (which seem to be constantly repackaged and redefined) are very good at playing most strategy games and defeating grand-masters at the games. I don’t really know how anyone could doubt that?
  • The technology already does an excellent job at “reading” handwriting and difficult to read (for many humans) text, and re-processing it as standard text. For English, for sure, and I assume other languages will lag.
  • Relatedly, most people have now experienced the very, very good translation capabilities of the software which is clearly approaching that of fluent or near-fluent speakers, certainly in terms of speed when translating written text, and getting very very fast at translating live spoken or recorded voice text.
  • Classifying complex digitizable objects into groups, and then modifying or transforming those digital obects. This, like games, was one of the early, obvious successful applications of the broad suite of software. Everything here is about recognizing and signaling relevant patterns in a large corpus of digital objects. And for many of these, using those patterns to “create” new patterns that range from junk to slop to useful (useful in the sense that people will pay some amount to have access to the technology): literary texts and art (digital humanities), recorded music (what songs are like each other, etc.), x-ray radiography and medical imaging broadly, satellite imagery, photographs and video, facial recognition, … will add more as they occur to me.
  • An domain that is so big it warrants its own category: navigating vehicles through complex landscapes. The use-application to cars, aircraft, drones, rockets, spacecraft is obviously quite large.
  • “Solve” complex mathematical problems that until a few years ago seemed only solvable by more and more complex numerical computing methods, but it seems possible that the suite of AI software technologies might be able to add to the range of solution algorithms. (relying on this paper).
  • Synthesize complex social science texts and offer cogent and often reasonable mid-level summaries and commentary. Every high school and university student and teacher has already figured this out, to our/their chagrin/delight? Accesses corpuses far larger than any individual human could retain, and so functions as a supplementary “search” for people wanting to learn a subject. It seems pretty clear that these technologies will displace the standard “search” technologies (and that itself is a complex subject because there seem to be several mechanisms that will lead to the displacement, with one of them being the enormous production of junk/slop making search less and less effective. And that may then feed back into the AI suite itself making it less and less effective.
  • ok gotta get back to work now….

Posted in AI | Tagged , , , , | Comments Off on “What can it do?” A living list of computational problems that deep learning/AI/neural nets can or seems likely to “do” (at varying cost and efficacy)

Reading August-September 2025

Posted on October 3, 2025 by mkevane

For some reason in leading up to the start of teaching classes in Sept. I did a lot of skim-reading in the last two months rather than complete book. But worthwhile a usual to keep track.

Kazuo Ishiguro, The Remains of the Day. I’ve read this several times and seen the movie several times, but for sure more than 15 years ago… so was fun to just open up random pages, knowing the whole context and story, and enjoy Ishiguro’s writing. It is, though, a painful read, in the sense of an unreliable not-very-likeable narrator about a bad subject-for-our-times (collaboration with Nazis).

Kazuo Ishiguro, An Artist of the Floating World. I had not read this and am slowly working my way through. So far very “complement” to Remains of the Day… unreliable, uncomfortable, icky subject (aligning with fascism).

R,F, Kuang, Babel. Got halfway through and then just jumped to the end. Heavy-handed, I found. Lots of exposition. The plot was too thin for me (gosh, I’ve read a lot of these wizard academy coming-of-age plots…). But I can see people who have not read a lot in the genre enjoying, or if you like reading the familiar genre and don’t want the genre to do much more than deliver, it’s probably fine.

Jose Saramogo, Levantado des suelo. Really the first 50 pages got depressing for me so I stopped but I will come back to this. Something maybe I will read Rayuela-style and then if enough chunks are good start over and read linearly.

Mark Gregory Pegg, Beatrice’s Last Smile. I read this a couple years ago. You can open to any of the short chapters and read super interesting anecdotes from primary sources about life in the 300-1300 AD period in Europe.

Richard Hingley, Hadrian’s Wall: A life. Like Pegg’s book, you can open anywhere to read vignettes of the ~2000-year history of the wall. Definitely makes me want to walk the length of the wall.

Richard Jenkins, A Fine Brush on Ivory. Appreciation and light literary criticism of Jane Austen. The enthusiasm is infectious. You start thinking, couldn’t I become an Austen-ite too? How delightful, the arcana, and the debates over interpreting Willoughby, etc.

Marcel Proust, Du cĂŽtĂ© de chez Swann. Again, can just open randomly and be “listening” to Vinteuil ….

Hermann Melville, White-Jacket. I think I was reading this in pandemic and picked up again. So many interesting vignettes. The homeward-bounder beards struck me again some reason.

Posted in Book and film reviews | Comments Off on Reading August-September 2025

The typical popular sci-fi version of AI posing an existential risk?

Posted on October 1, 2025 by mkevane

Plenty of science-fiction stories and movies have as plot the defeat of a super-intelligent and autonomous AI that poses an existential threat to humans. In the usual backstory, the AI at some point developed goals that seemingly impelled the AI to take actions that threatened humans, either by killing them all, or by rendering the planet uninhabitable, or by depriving them of substantial freedoms (e.g. using them as slaves, or pets). The AI violates what is usually called the “first law of robotics,” which sci-fi writer Isaac Asimov phrased as: “A robot may not injure a human being or, through inaction, allow a human being to come to harm.” The AI becomes, in effect, a tyrant. The story might be told on a small-scale, serving as metaphor (HAL 9000, Ex Machina, Humans), or at the literal dystopian global scale (Terminator, Matrix).

These treatments of the existential threat of AI typically share elements: (1) the goals of the AI become misaligned with what would be recognized by most ethicists as those goals that promote the common good; (2) the AI is given or gains access to kinetics (it controls robots and weapons or the equivalent, although in some treatments the AI is so clever it can convince humans to do kinetics for it); (3) by the time the threat is clear to many, the AI can no longer be “turned off”; and (4) presumably, during the time the threat was likely but unclear, the political economy of the period was such that effective guardrails to prevent the AI from going rogue were not implemented.

These scenarios seem plausible over some time frame. That is, advances in computing capabilities in the 75 years since 1950 years have been astonishing, and by any indicator have been and continue to be accelerating. Complex software such as LLM’s are opaque and currently unpredictable. The history of LLM sessions (interactions) with a user appears to influence or modify the LLM text and video responses to otherwise equal prompts. That is, the algorithms deliver different outcomes depending on “real world” context. The criteria that adjudicate how context modifies or moderates actions (text or image generation, as of 2025) appear to be not programmed, that is, they are “emergent properties.” The more continuously operating an AI algorithm might be, the more history of actions it accumulates, and eventually one might characterize patterns of differential responses as “goals.” These goals may then be aligned or misaligned to varying degrees, across many dimensions of context and prompts. AI software that can control many kinetic processes has been and will continue to be deployed (electricity producing plants and distribution grids, vehicles and transport systems, weapons systems, medical devices, advanced laboratories, manufacturing and processing factories).

On point (3), humans of 2025 are habituated to a reality of “one computer, one power cord,” and so the notion that an AI could not be turned off seems to be a low probability. But most people envision a future in 100 or 1,000 years when all vital life-support systems will be controlled, to varying degrees, by AI agents. In such ubiquitous distributed networked computing environments, the very concept of “turning it off” loses meaning.

On point (4), it seems quite reasonable that a world of 2040 or 2050 will continue to have private firms with huge market capitalizations, extremely well-paid executives, loyal work forces, extensive private security forces, lobbying and social network links to professional militaries and government officials. In such a world, the incentives of the individuals involved, or the various social groups the individuals constitute (boards of directors, firms, governments) are quite likely to sacrifice an uncertain common good for a certain private gain.

Posted in AI | Tagged , , , , | Comments Off on The typical popular sci-fi version of AI posing an existential risk?

AI productivity growth and “the economy”

Posted on September 14, 2025 by mkevane

I’m preparing for a course I am teaching, so am intending to start regularly writing on the blog as a way to livestream my thoughts. Hopefully, I will maintain momentum for a long while. My first thoughts, while reading Nordhaus’s singularity-themed discussion of AI, was about how it would always be useful for such discussions to repeatedly note that “the economy” means “the value of output using prices” and that when it comes to information technology where prices are often zero, it is easy for the text to carry on with analysis as the reader slowly remembers (lagging thought) that results may be happening because prices have gone to zero and so that sector is no longer in “the economy.” Consider, for example, the “service” of communicating with other humans. At one point in human evolution, the “cost per communication” may have been quite high. A proto-human had to have significant neural capability, perhaps driven by substantial consumption of proteins, or substantial attention from a caregiver who then was not gathering or hunting, for them to be able to send and receive signals to other proto-humans. Over millennia, productivity in this activity (both a production input and a consumption good) increased substantially. By how much did “the economy” grow? Suppose that at 1 million BC, hominids were gathering and hunting and had 100 signals in their vocabulary. Signal combinations were not interpretable, imagine (no syntax, no grammar). Over the following 990,000 years (until the Neolithic Revolution) we got to symbols (“cave drawings” and marks on stones and clay) with complex meaning, likely reflecting a similar complexity in ordinary human (rather than ruling class) capabilities for symbol mental manipulation. What was the growth rate in the productivity of communicating? Suppose hominids went from those 100 symbols to being able to retain or understand or interpret or extreme 1 million symbols or concepts (example: in English, all children understand that “a running angry black dog” is not quite the same as a “black angry running dog”)? This is a phenomenal increase, but at the time scale concerned (close to 1 million years) is a very small annual growth rate (.001% growth per year). An economist who measured “the economy” however, might, for enormous swathes of humans, see them engaged in the same production activities (gathering, hunting) and not measure this change.

To take another thought experiment, suppose that there were other intelligent entities in the galaxy, who understood quantum mechanics a million-fold better than we do currently, and through their understanding sent us a little button that we could press and be instantly teleported to a specially constructed view-zone on their planet or interstellar vehicle and we could “observe” something. We could spend hours or days doing this, with only the opportunity cost. From the GDP perspective, this amazing and completely new good might count for less than a new Lady Gaga album and concert tour. “The economy” may barely change. And if you’re thinking, “But surely the experience of quantum jumps through the universe to see alien life would revolutionize human society,” well, I also have a bridge for sale, cheap.

Posted in AI | Tagged , , , | Comments Off on AI productivity growth and “the economy”

Books read July 2025

Posted on July 22, 2025 by mkevane

Henry James, Washington Square. I gave up about 2/3 way through. As usual with James, he just takes forever to advance the plot, and the long decriptions of character and motivations just make the reading tedious. Skimming to denouement, I found I had no desire to read carefully.

John Jennings (text) and David Brame (illustrator), After the Rain (after the short novel by Nnedi Okorafor. A graphic novel of a short novel. Sort of a horror story with an identity question (Nigerian or American!?) tacked on. The illustrations play up the horror/gore. Personally not my vibe at all.

Gene Wolfe, The Best of Gene Wolfe. Collection of short stories and novellas. He has a recognizable style. Folksy and uncanny. Shirley Jackson? Not much science-fiction in the collection. Many are enjoyable, but after a few days I didn’t find that the stuck with me.

Enrique Vila-Matas, Paris no se acaba nunca. Been skimming this randomly throughout the month, reading 5-10 pages in chunks. Doesn’t seem very linear.

Javier Marias. Todas las almas. Reading for sibling book group.

Virginia Woolf. Mrs. Dalloway. Reading for literary book group. Very, very good- merits careful, slow, thughtful reading.

Alan Bennett, An Uncommon Reader. Reading for neighborhood group. Very short and enjoyable and a nice complement to the literary fiction.

Robert Nozick, Philosophical Explanations. Trying to slowly work through the last long chapter on meaning.

Posted in Book and film reviews | Comments Off on Books read July 2025

Contemporary folk and Americana

Posted on June 23, 2025 by mkevane

A friend, in conversation, revealed that they hadn’t listened to any folk (now more broadly as Americana) since the 1960s! They were intimately familiar with that Greenwich Village revival, but knew no artists since then. So I thought I’d make a list, in no particular order, of a dozen folk/Americana (e.g. primarily acoustic, lyrics worth parsing, often with a political message even if only of oppression whether from poverty or gender) that are on my playlists. Some of them I know nothing about except I heard a song an really liked it! https://open.spotify.com/playlist/0wBJBoOxcnZ8hkUe2FMX62?si=IsjscFP0Q4CA-pEAQ3Dojw&pi=viX_Q7PuTAC9U

  • Iron & Wine
  • Mount Eerie
  • Gillian Welch
  • Mercury Rev
  • Sun Kil Moon
  • Bill Callahan
  • Little Wings
  • Adrienne Lenker
  • Sufjan Stevens
  • Marisa Anderson
  • Adam Torres
  • Railroad Earth
  • Waxahatchee
  • The Antlers

Posted in Music | 3 Comments

Books read May-June 2025

Posted on June 19, 2025 by mkevane

Mostly light reading these two months, with Vila-Matas in reserve occasionally.

Adrian Tchaikovsky, Made Things. Throwaway fantasy short novel. The usual. Fine for kids. At this point he must be able to write stuff like this in 2-3 days?

Benjamin Black, Christine Falls.

Benjamin Black, The SIlver Swan. After reading a couple of John Bannister literary novels, decided to dip into his crime novels (written under pseudonym). While friends highly recommend, I was disappointed. The crimes did not seem that interesting, and Quirk, the forensic pathologist, kept being described in exactly the same way. And according to the narrator, Quirk’s self-description (inner voice passages) is the same as the narrator’s! What? For Irish “color” or world-building of the 1950s, barely a whisper. Could be set in the present and most readers would not know.

F. Scott Fitzgerald. The Great Gatsby. Reading for book club. After we selected, slew of stories came out; I guess it is 100th anniversary. My daughter said it best: “It is great for every high school student to read, because all the literary devices are there hitting you over the head, so it is good for 10th grade analysis!” I had not read since high school, and had forgotten all of the plot. The unrequited love story is so superficial, and Tom and Daisy Buchanan are so one-dimensional, one wonders how, say, someone who had read Proust or Zola or Balzac or Dickens thought it significant.

Ursula LeGuin. Wizard of Earthsea. I do like reading fantasy novels in Spanish or French, because the shift in vocabulary adds some distancing and immersion that is useful I think for “transportation” that fantasy novels rely on. I read this in French, and noticed that some words that I thought to myself “don’t know” were her own made up words… so it was more like being a chid and not knowing whether a word referred to something or was fantasy. I’m a sucker for a single-themed novel about coming of age and being reconciled to the world as a big, bad, grim place, with dashes of beauty and joy, and this novel delivers that.

Robin Hobb, Assassin’s Apprentice, and Royal Assassin. Very good fantasy escapism… close attention to detail and character. In the end, I grew weary though, and really started skipping. I found I could pick up 100 pages later with little loss. Will not be reading others. But would definitely recommend for a 14 year old avid reader.

Junji Ito, Frankenstein. Manga-style version of Frankenstein. Fun read.

Posted in Book and film reviews | Comments Off on Books read May-June 2025

Some tariff basics?

Posted on May 7, 2025 by mkevane

Context

Consider a country in South America, say Bolivia.  The population is about 12 million, with about a third living in rural areas. GDP per capita in 2025 might be around $4,000 per capita. There is significant inequality, so cities are much wealthier than villages and towns in rural areas. Infrastructure (roads, dams, electricity grid) is good in cities, but poor outside.

Imagine that in 2026 the country has a dozen multinational mining companies operating lucrative lithium mines. The companies turn over 10% of the value of lithium exports as tax to the government. The lithium mines employ many Bolivians (and also many foreign nationals) and indirectly generate much employment (for food provision, transport, port facilities). Government spending on schooling and health services is higher because of the mining revenues. That is, there is a “multiplier effect” from the mines on the local and regional economy.  One might imagine that some economists have estimated the direct and indirect effect of the mines, and find they “contribute” 5% to GDP, per year. Another way of saying this is that if the mines were to suddenly shut down, and produce no income, and all the workers were to be laid off, GDP (including government services) might be expected to decline by 5%. The burden of such a decline would fall heavily on the mine workers and all the workers directly serving the mine operations and the workers. But far away, in the ports, perhaps, companies and workers would also feel the effects. Speaking of the multiplier effect of the mines, it is worth remembering that some of the possible multiplier effect is “diluted” as Bolivian nationals directly and indirectly benefiting from the lithium boom increase their purchases on imported consumption goods (iphones, etc). But they may also import investment goods (especially for housing construction), that generate longer term increases in potential GDP. Economists (and others) can estimate all of these direct and indirect effects, to varying degrees of credibility.

Let us assume there is no significant “voracity effect” or “resource curse” from the boom, where factions within the government compete, often in very wasteful or even destructive ways, to control the increased government revenue, or to extract even more revenue from the lithium mining sector. Let us assume that whatever environmental impacts that might be happening directly or indirectly can be considered separately.

The question

The question before us, after that context, is whether the government of Bolivia should undertake a policy of: “Why don’t we put high tariffs on all imports; that will make Bolivia great again.” The presumption is that some measure, say GDP per capita (that is, average income per person), would be significantly higher, after some period of time, following the policy change.

Preliminary analysis

It can be useful to start with a single industry and a single tariff. Suppose the mining sector uses large-size industrial trucks. These trucks might be manufactured by multinational companies like Caterpillar or John Deere, or other companies. Let us assume they cost $5 million each (they are really big! https://www.caranddriver.com/reviews/a15140071/caterpillar-797-specialty-file/). A large mine perhaps has 10 trucks, and the many mines of Bolivia might together use 100 of these trucks. We imagine the Minister of Industry in Bolivia says to the President, “We can build these trucks here in Bolivia. It is just a giant truck. It cannot be too hard.” The President thinks that a truck factory would create good high-paying jobs in the capital city, and that eventually Bolivia could export the trucks to other countries. The President announces a 100% tariff on the trucks (and maybe also a high tariff on other kinds of trucks). Importing the truck would now cost the mining firm $10m (the $5m purchase price and the $5m tariff paid when the truck is imported into Bolivia, collected at the port.) 

The President also announces that that land and electricity for a truck factory will be provided. A group of Bolivian capitalists decide to set up a new corporation, the Bolivia Truck Co. (Boltru). They pool their capital of $100m, they build a factory, they import machine tools and other equipment and raw materials, and they hire a lot of skilled laborers, and they start their production line. Their plan is to produce 10 trucks a year. After a couple of years, they proudly announce they are producing the target of 10 trucks a year. In a footnote, they note that the trucks cost $7m each to build (all materials, labor, and depreciation of factory and tools) and they sell them to mining companies for $8m (so there is $1m profit per truck, for the factory owners). The factory employs 1,000 skilled workers, earning somewhat higher wages than they did before. Economists have estimated the multiplier effect of the truck factory. The total positive effect on GDP is nowhere near that of the mines themselves, but it is large.

The Bolivian President, one day, says, “That worked great, why don’t we do that for every industry?”

A skeptical economist

In the newspapers, coincidentally, that very day an opinion piece is published by a prominent Bolivian economist. Here is the summary of what she wrote. 

The country’s policymakers seem to have forgotten the basic economic principle of opportunity cost. Consider a farmer who labors year-round to raise chickens, who feeds them, who spends money on medicines, who takes their eggs to market, and who at the end of a year calculates her profits and determines that she made a profit of $3,000 over the whole year. When I look at her calculations, I notice she has not valued her own labor as a cost “Well, I didn’t have to pay for my own labor, did I?” the farmer answers, “So it cost me zero!” That’s when I say, “But your labor had an opportunity cost! You could have worked all year round in your town, which is near your farm, which is close to a large lithium mine, maybe as a cook in a restaurant. Same hours, same quality of life.  And if you had worked as a cook, your income at the end of the year, according to my calculations, would have been $5,000!  You actually lost $2,000 this year relative to what you could have had. You chose a path, ignored the opportunity cost, and so ended up worse than you could have. To calculate the opportunity cost, we must consider a counterfactual, what could have happened if you had chosen a different path relative to what you chose.”

“Obvious,” some people say. But they often fail to apply the same kind of counterfactual, opportunity cost, reasoning to tariffs. Take the case of the truck tariffs. The counterfactual is what would have happened in the absence of the tariff. We just have to do a little bit of reasoning to arrive at several observations. First, in the absence of a tariff the mines would have likely purchased more trucks, and replaced their old trucks, at a higher rate, so they would have produced and exported more lithium. They would have hired more skilled workers. The ports would have had more business. The multiplier effect of the increased mining would have increased other people’s incomes. Instead, that did not happen. Second, when the Bolivian capitalists got together to establish the local factory Boltru, they may have noted that they would build the truck factory instead of building a large eco-tourism hotel in the mountains that was going to attract tourists from all over South America, and even beyond, to enjoy the natural beauty, cuisines, music, and culture of Bolivia. Perhaps the planned hotel would have employed 2,000 workers (chefs, musicians, clerks, fitness coaches, etc.).  The local multiplier effect of the complex would have been large. And the building materials, in the spirit of eco-tourism, were all to be sourced locally. The truck factory, on the other hand, was largely built with imported materials, and all of the machinery in the factory was imported. The opportunity cost of building the truck factory was not building the hotel. Third, the tariff on trucks primary effect is to reduce truck usage. Perhaps what some of the mines did was offer incentives for Bolivians to bring their donkeys and llamas to carry out mined lithium ore. The international photographers loved it, of course: a picturesque caravan of llamas piled high with sacks of lithium, walking out of the hellscape of the mining pit and unloading 10 miles away near the road. But anyone venturing further into the villages around the area would have seen rotting produce: “There are no animals to carry the potatoes and cabbage to the market, so what can we do?” Opportunity cost.

Posted in Development thinking, International trade issues, Teaching international trade | Tagged , , , , | Comments Off on Some tariff basics?

Books read March-April 2025

Posted on May 6, 2025 by mkevane
  • John Banville, The Sea. Highly literary fiction, with an unpleasant narrator, who spins a melodrama about aging and memory that definitely I will re-read someday. Had a great discussion about it with book club partner readers. Filled with details and nuance.
  • Percival Everett, James. A book that needed to be written. Easily read over a few nights. But once read, there would be no pleasure or profit from re-reading.
  • John Banville, The Untouchable. Highly literary fiction, with an unpleasant narrator, who spins a melodrama about aging and memory that definitely I will re-read someday. Had a great discussion about it with book club partner readers. Filled with details and nuance. (Yup, repeated on purpose!)
  • John Banville, Body of Evidence. I had to stop. Too unpleasant a narrator. Reminded me of Wasp Factory.
  • Josephine Tey, Daughter of Time. Who did kill the brothers in the Tower? Richard? Or…. Fun detective novel as historical fiction, but the style definitely dated.
  • Shirley Jackson, Flower Garden (short story). Yikes: a meditation on racism that cuts straight through the bullshit.
  • Richard Feynman, Six Easy Pieces. I have a decent appetite for “physics explained to economists,” and this was pretty good!
  • Beth Brower, The Unselected Journals of Emma M. Lion. Excellent if you are 12 years old.
  • Adam Gopnick, All That Happiness Is: Some Words on What Matters. “Some words that my publisher thought we could make a tidy sum off, is more like it.” Learn a craft for the emotional satisfaction of improving at something. There, saved you time and money.
  • Adrian Tchaikovsky, Alien Clay. A poorly written rehash of some of his main ideas (complexity and emergent properties). Perfectly fine as an entry text into his work. The writing is dialed in here.
  • Neal Stephenson, Quicksilver. You can imagine Stephenson in a small home library, with old books about the 1600s everywhere, and manuscript pages scattered about, and every morning he randomly picks a few and then figures out how to insert into his narrative fictionalized version of the times. After a few hundred pages, most readers I am sure start thinking, “Why not just read Wikipedia? Or better yet, let me go browse in the old books section of my local bookstore.”
Posted in Book and film reviews | Comments Off on Books read March-April 2025

A stab at the scope of shared faculty governance at the school level in a university

Posted on May 5, 2025 by mkevane

I was thinking about what the scope is for collaborative governance in a school within a university. Often the institutions and norms of shared governance are weakest at the school level. University governance gets a lot of attention, and department governance has a lot of self-interested actors. but school governance tends to be weak, is my prior?

  • High level occasional issues where faculty should be leading/equal partners in discussions
    • Refining and reaffirming shared values, purpose, and mission
    • Academic freedom
    • Importance of knowledge production and dissemination
    • Strategic thinking/planning
      • International partnerships
      • Capital projects
      • Non-degree programs
      • TT/NTT ratios
      • Semester vs. quarter
      • Institutes and centers
  • Routine generalist academic issues where faculty should be leading
    • Curriculum and new program development
    • Allocation of scholarship resources within school
      • Funding and course releases
      • Similar resources
    • Standard setting and evaluation of teaching and scholarship
      • Oversight and calibration of department practice for TT and NTT faculty
    • Recruitment and retention policies and practices not covered by contract or university-level policy
  • Routine but more specialized and occasional issues (i.e. developing capacity for informed analysis takes time/experience) where administration staff usually should be leading
    • Compensation and working conditions (faculty and direct academic support staff)
    • Watchdog function for high-level accountability in expenditures and performance of administration, institutes, centers, etc.
    • Assessment of learning and accreditation
Posted in Governance | Comments Off on A stab at the scope of shared faculty governance at the school level in a university

School governance in a university setting

Posted on May 4, 2025 by mkevane

I have been thinking about the issue of school governance in the university setting. One way of engaging in the conversation is through introspection on relevant questions. Relevant for the school and academic departments. I thought I would start a list. The more the answer is “No” (or a similar variation) to these questions, the more it seems to me that school governance is doing badly. Some may be hard to measure, but every faculty member can score based on their own experience and then aggregate to a weighted index. What is missing?

  1. Do we have a clear process for establishing and monitoring medium and longer-term strategic goals? Who is responsible for that process? Where is the record of the regular outcomes of that process?
  2. Do we promote budgetary transparency, beyond a cursory presentation every 3-4 years of some numbers?
  3. Do we regularly assess, in a structured way, the financial and non-financial “returns” on our programs, to learn from decisions made in the past?
  4. Do we have a culture of active participation in shared governance, or do we pooh-pooh that leaving it to colleagues in other schools or departments?
  5. Do we have a culture of being effective advocates in contributing to the vision and administration of the University?
  6. Are we proud of our efforts to seek outside review on a regular basis for our undergraduate and graduate programs? When we do outside review, do we see it as an opportunity to challenge ourselves and improve, or as an occasion to “make a report that is satisfactory with the least amount of work”?
  7. Do departments have a process where they regularly update their department scholarship standards, or is the culture one of “they made us do that so we did it five years ago and we never looked back or revised because it is not worth any trouble to revise”?
  8. Do our department chairs exhibit high levels of professionalism and pro-active competency? Often chairs have received no training nor are they given any clear expectations for performance, and few departments do anonymous 360-style reviews of chairs, even though this is standard in many academic units.
  9. Does our tenure process make us proud, or is it fraught with inconsistencies, opacity, rumors and anxiety? Are there checks in the process to prevent abuse? Has our tenure process been examined openly and honestly in light of past problems, in a well-documented way, in order to be improved?
  10. Is our culture one of openly discussing and acting intentionally to counter possible discrimination (gender, ethnic) in faculty hiring and retention, as an important goal of the school, or is the culture rather to not openly discuss it or even to be openly hostile towards the goal of countering discrimination and discussions of that as a goal?
  11. Is our “meeting” culture one of no-agendas and much of the meeting time devoted to information sharing (e.g., via Powerpoint), with people who have no reason or real interest to be in the room in the meeting, or an occasion to get a sandwich and check messages and email? When we have information to communicate to faculty and staff, do we tell them to come to a meeting without telling them what the information will be?
Posted in Governance | Tagged , , , , | Comments Off on School governance in a university setting

Books read February 2025 (and tracking film, too)

Posted on February 6, 2025 by mkevane

Rachel Kushner, Creation Lake. Reminded me of Elizabeth Hand. A moody, philosophical novel, but on the whole meadering writing and for me not very inspiring. Eventually, I gave up about p. 270 with 100 pages left. At some point I realized nothing was happening nor was anything going to happen, so why continue. Plodding. The idea here seems to be to make the reader go through 350 pages without any compelling character, plot, or writing, and with a string of fairly banal philosophical observations mixed in throughout.

Mohamed Mbougar Sarr. Silence du Choeur. An earlier novel of his- very competent polyphonic voices about African migrants in Sicily. I’m not a lit studies person, but can I presume riffs/references to The Aeneid, which would make it more fun to discuss? (Google search reveals nothing…)

Film

  • Emilia Perez – great first half then it (and I) wandered off
  • Beanpole – beautiful Russian film about two women in Leningrad right after WWII. Harrowing and quiet.
  • Caravaggio (by Derek Jarman) – self-indulgent and ridiculous. Probably “daring” in its time but a yawn now. Contrast (aesthetic, intensity, attention to detail) with Beanpole couldn’t be bigger.
  • Saturday Night Live. A re-creation of “first night.” You are definitely better off seeing Birdman.
Posted in Book and film reviews | Comments Off on Books read February 2025 (and tracking film, too)

Books read in January 2025

Posted on January 11, 2025 by mkevane

Scott Alexander Howard, The Other Valley. Nicely done, sort of YA, regretful, and grim (like a woodcut!) time-travel story.

Adrian Tchaikovsky, The Doors of Eden. Disappointing multi-earth multiverse story, a la Terry Pratchett. Some interesting evolution possibilities, but the plot was just dumb and got dumber.

James S. A. Corey, Leviathan Wakes (first book of The Expanse). Watched the show some years ago. They stuck very closely to the text of the novel. So a very enjoyable distraction.

Posted in Book and film reviews | Comments Off on Books read in January 2025

Books read in December 2024

Posted on January 11, 2025 by mkevane

I only use this blog to keep track of my book reading.

Guy Gavriel Kay, A Brightness Long Ago. Excellent fictionalized novel with the Palio in Siena!

Georges Simenon, Le passager du “Polarlys”. I was unimpressed. Enjoyed the French. I guess at the time the world-weary captain in the frozen north and the semi-debauchery of the criminal may have been thrilling.

Guy de Maupassant, Bel-Ami. Enjoyed the French, and the sociology of the times, and the profile and evolution of an unlikable and dubious yet sympathetic (to be reader) hero.

Roberto Bolaño, Estrella Distante. 2nd time reading. Excellent.

Posted in Book and film reviews | Comments Off on Books read in December 2024

Nocturno de Chile, by Roberto Bolaño

Posted on November 17, 2024 by mkevane

My second time to read Nocturno de Chile, by Roberto Bolaño. I really missed a lot in the first feverish read about two years ago! Highly recommend multiple reads.

Posted in Book and film reviews | Comments Off on Nocturno de Chile, by Roberto Bolaño

Shopgirl, by Steve Martin

Posted on November 17, 2024 by mkevane

Book club discussed Shopgirl, by Steve Martin. It’s a slight novella, later turned into a movie (book reads as a script, we thought, with stage directions and present-tense narrator). We mostly intensely disliked, and were disappointed. It came across as cringy and unaware. Characters were flat.

Posted in Book and film reviews | Comments Off on Shopgirl, by Steve Martin