His Dark Adaptations

Like any YA-novel heroine worth the name, Philip Pullman’s His Dark Materials series has repeatedly been tasked with filling the shoes of bigger heroes in order to save the world – scratch that: worlds – from peril. First came the world of bookselling, as Pullman’s novels were used to fill the lulls between releases of the Harry Potter books and, eventually, help bookstores in their attempts to stave off Amazon.com, which entered the book-selling business the same week The Golden Compass was published in July 1995. Then came the world of film, when New Line was looking for its next big hit after The Lord of the Rings wrapped in 2003, and so gave The Golden Compass the biggest budget in its history (which was still ≈70 million less than what Amazon later paid for just the rights to LOTR’s backstory). Finally, in 2019, began the finale of this trilogy: the Great Streaming Wars. HBO, in trying to ward off spectres like Amazon, partnered with the BBC to adapt Pullman for TV, part of its search for new materials to follow up Game of Thrones.

Matching the success of the Harry Potter books, Lord of the Rings films, and Game of Thrones show would have been daunting enough even if His Dark Materials was a comparable product. But, although perhaps equivalent in terms of quality, it does not possess the immense quantity of those other series. The full His Dark Materials series clocks in at 390,000 words, compared to more than half a million for Lord of the Rings, over a million for Harry Potter, and soon-to-be two million for Thrones. Pullman’s first two books combined are shorter than the shortest of George RR Martin’s – the book off of which GOT’s Season 1 was based. In trying to create a similar hit, HBO had to stretch the material in His Dark Materials dangerously, even hobbitishly thin:

*The stats for the Harry Potter films and for the Peter Jackson Middle Earth movies were taken from the blog Overthinking It.  For the GOT show, only the first 2 seasons are shown here – after season 2 the book/tv timelines no longer overlap neatly at all.

Pullman’s brevity might have actually been an advantage in terms of making a film adaptation without having to cut key scenes or characters, as the Harry Potter and Lord of the Rings movies and later seasons of Game of Thrones all had to do. The Golden Compass movie, however, was forced to carry out this same type of cinematic intercision anyway, only for completely different reasons.

Fearing that advocacy groups like the Catholic League, and also the Vatican, would suppress the film’s ticket sales, New Line decided to cut any scenes that might broadly offend the Catholic Church. This was way back in the 2000s, a time when even the inoffensive and pro-Christian Harry Potter was still criticized from certain religious corners as being too pro-witchcraft, Baruch and Balthamos couldn’t yet have legally married even in Los Angeles, and New Line itself had just produced a prequel of Mel Gibson’s The Passion of the Christ the year before releasing Compass. So, it was no surprise that there would be serious pushback against a series of children’s books in which (spoiler alert) sexual awakening between young teenagers is the fruit of original sin and can save the world, and the heroine of the entire story is, in effect, the antichrist.

But New Line’s decision meant bungling the adaptation, despite an excellent casting job that included Nicole Kidman, Daniel Craig, Sam Elliot, and Derek Jacobi. The movie was forced to remove the entirety of the book’s final chapters, including the book’s devastating climax which mingles and twists a number of Genesis myths (the tempting of Eve, binding of Isaac, and ladder of Jacob). The result was akin to ending Empire Strikes Back before the Luke, I am Your Father scene.

An essay in The Atlantic in 2007, How Hollywood Saved God, describes how this came to pass:

…[Director Chris] Weitz told me he tried to keep in a line where Asriel says, “Dust is sin,” but “that didn’t make it. What can I say?” Hollywood “is just terrified that anything that brings up religion or anything controversial will be disastrous.” But after three years of working on the movie, he’d come up with a solid explanation for why he’s not selling out: In the ’80s and ’90s, Hollywood was “scornful in a very intellectually unsound way about religion. Any priest or nun was a dogmatic idiot. So I think there’s something valid in the way the Christian community has responded.” There will be some religious imagery in the movie, Weitz said, but it will be blended so unobtrusively into the production design that it will take a “DVD player and working knowledge of Latin to decipher the symbols.”

SPECTRE: Lord Asriel Will Return in “Quantum of Soulless”

That was then. Thirteen impressively sacrilegious years later, the idea that entertainment should have to bend the knee to the likes of Bill Donohue and the Catholic League seems quaint. And so, HBO, in partnership with the BBC, had the chance to adapt things faithfully – that is, without faith getting in the way – this time. Yet the pressure to replicate Game of Thrones’ success may have been too strong, forcing producers towards the opposite extreme instead. Whereas the 2007 adaptation had a runtime of 1 hr 45 minutes, for a book that probably needed 2 or 3 hours to do justice to – maybe 4, at a stretch – the 2019 version expanded the story into 7 and a half hours of TV.

The result of this was a too-drawn-out storytelling pace, and so much new material being written for side characters that Lyra, whose point of view dominates the book, is off-screen for much of the show. The show also had to move key parts of the story forward from the second book into the first season, and then from the third book to the second, in the process diluting the power of both books’ revelatory final chapters. (This book-shuffling also spoiled perhaps the best minor twist in the books, but in that case there was the silver lining of getting to watch more of Ariyon Bakare). The need to fill the extended runtime also led to quite a lot of thematic exposition, with characters discussing philosophical aspects of the narrative which might otherwise have unfolded gradually. All of this undermined what, in the books, is one of the series’ most coveted qualities: subtlety.

There is another even deeper danger lurking, which could be glimpsed in the very first scene of Materials’ first season: spinoff creep. Rather than begin with the books’ iconic decanter of tokay opening scene, the show instead starts with events pulled from Pullman’s The Book of Dust, the prequel/sequel trilogy Pullman began publishing in 2017 and is currently still writing the finale of. Obviously — as can also be seen in HBO’s move to set a handful of new shows in Westeros — there is pressure on producers to engage in extended world-building in order to compete with the megadeals that have landed Amazon Middle Earth, Netflix Narnia, and Disney galaxies near and far.

Killing Eve: the sinister Magisterium seeks to pre-empt original sin in the second season of His Dark Materials

(As an example of just how crowded this competition is, His Dark Materials was not even the only beloved ‘90s fantasy book with the antichrist as its hero to debut on streaming services in 2019. There was also Amazon Studio’s adaptation of the Terry Pratchett and Neil Gaiman novel Good Omens. Like Materials, it too was co-produced with the BBC and featured a star-studded cast. It too may return for a second and third season, despite having been adapted from a single book).

In the end, the attempts to adapt and market Pullman’s fantasy series may serve to illustrate how, like a teenager’s daemon, the more things change the more they stay the same. Books become movies, which later beget television shows. Standalones become trilogies and then grow into cinematic multiverses. Popes and Catholic Leagues wax and wane; Hollywood’s religious mores can swing like the needle of a compass. But throughout all of these transformations, film and television studios can always be relied upon to try to duplicate their past hits. They worship a different God, and fear different devils, than the ones who will do battle at the end of His Dark Materials.

The Shootout Problem’s Simple Solution:

The NHL is back, thank god. But so is the shootout.

There is a problem with the shootout: it’s anticlimactic, especially after the exciting 3-on-3 OT that precedes it. And it does not resemble the sport of hockey enough to warrant the significant impact over the league standings that it often has.

Most hockey fans want the shootout to be replaced by more sudden death OT, especially since 3-on-3 hockey tends not to go on for too long before a goal is scored anyway. The problem is, the stars who have to play the most in those OT periods are understandably wary of extending their duration and further increasing their intensity.

The solution is simple: for the 2nd OT period, do not let any players from the 1st OT play. And in the 3rd OT period, don’t let any players from the 1st or 2nd OTs play.

If the game is still not over after these 15 minutes of 3-on-3 hockey, then you can either have a shootout (shootouts will actually become exciting again, by virtue of being rare!), or else go on to a 4th OT period in which the 1st-OT players can play again (just like how, in a shootout, a player can go for a second turn once everyone else on his team has taken a turn themselves).

So, NHL, make this change. And make it now. It’s a hell of a lot less radical then breaking your league into Canadian and American halves, after all…

On Big Z

Even in the most congenial divorce, the division of assets can hurt. When the Czechoslovakian men’s hockey team split after the country’s Velvet divorce in 1993, it was the Czech who ended up with most of the family fortune. With stars like Jaromir Jagr (second only to Gretzky in NHL career scoring) and Dominik Hasek (arguably the best goalie ever), the Czech would go on to win the gold at the 1998 Olympics, something that even the unified Czechoslovakian team had never been able to accomplish. 

Slovakia, however, still managed to keep the house: Zdeno Chara. 

Chara was already 16 years old when the country split. He was just about to begin his hockey career playing for his hometown team in Trenčín, a small city five miles from the newly established Czech border. Unlike his father, Zdenek Chara, who had been an Olympic champion in Graeco-Roman wrestling, Zdeno Chara never got to represent Czechoslovakia in international competition. He did, however, inherit his father’s Olympian physique: At 6”9 (7 ft. on skates), weighing 250 pounds, Chara has long been the largest player in the NHL, dominating defensive zones for 21 seasons since 1998. Though he has only won one James Norris award for Best Defenseman—Nik Lindstrom often stood in his way, winning seven, leaving Chara with five runner-ups—he remains the perpetual winner of the Chuck Norris award for deadliest man with a blade.  

As of today, Chara is the only European player, other than that same Nik Lindstrom, to captain a Stanley Cup-winning team, having led the Boston Bruins to end a nearly four-decades-long Cup drought in 2011. Chara has been the Bruins captain for 14 years, during which time they have made it to the playoffs 11 times, and to the Cup finals three times. He would easily be the franchise’s greatest defenseman, except that, this being the Bruins, Chara will always rank far behind Ray Bourque and Bobby Orr. Nor is Chara even the greatest 43-year-old athlete in recent Boston memory — that honor goes to Tom Brady of course, six months Chara’s junior. Like Brady, Chara is finally leaving New England this season, but not yet retiring. He signed with the Washington Capitals last week, where he will join Alexander Ovechkin as both seek to drink from a second Cup. 

Much attention has gone to Chara’s flashier records, such as the speed of his slap shot (108.8 miles per hour, the NHL’s fastest), his size (the league average is 6”1, 199 lbs), his career plus-minus (the highest plus of any current player), or the fact that he played one of the longest ever shifts (4 min and 18 seconds) when he was already 40 years old. It is easier though to overlook the most impressive record Chara is building, because the NHL has not historically kept track of ice time played over the course of a career. The league keeps track of average ice time per game, a stat Chara has led the league in during past years, and even led the Bruins in last year at 42 years old. But look a bit deeper and you can see Chara is racking up an almost Lebron-esque resume for total minutes played, across regular seasons and playoffs. 

With 21 full seasons thus far, 15 playoff runs, 3 finals runs, and almost no time lost to injuries, Chara is the leader in both regular season and playoff minutes played, and shows little sign of slowing down just yet. He has racked up far more ice time than the player with the second most career minutes, iron man Patrick Marleau (who, at age 41, takes ice baths to reenergize himself during every intermission between second and third periods). By the end of this season Chara will become the fourth oldest skater ever to play in the NHL, passing hockey legends Teemu Selanne, Tim Horton, and Doug Harvey. That will leave Chara chasing only Jagr (who retired last year at 46 years old, and still plays pro hockey back in the Czech league), Chris Chelios (48 years old, retired in 2010), and the great Gordie Howe (52 years old!). Howe actually pulled off what Lebron hopes to do one day, ending his career playing alongside his son

This brings us to Chara’s other Lebron-ish attribute: making his teammates, and specifically his goaltenders, look amazing. Admittedly some of this may be coincidence; trying to measure an individual player’s impact in a sport like hockey can be a somewhat fuzzy pursuit.  But it is probably not a complete coincidence that Chara’s teammates in Boston have won the Vezina award for Best Goalie three times — Tim Thomas twice and Tukka Rask (who was also Vezina runner-up this past season) once. Both goalies have had exceptional seasons with the Bruins: Thomas had a season with the 2nd best save percentage in modern NHL history, and Rask a season with the 8th best. (Dominik Hasek, who had the 3rd best save percentage season, also played with Chara for a year in Ottawa, and played well, but he had already won 6 Vezinas and 2 MVPs by that point in his career, without Chara). Chara’s long reach and ability to remove opponents from the front of the net have made him particularly useful at limiting goals during penalty kills, at which the Bruins have usually excelled.  

I say all of this with a respect that is truly grudging. Chara’s shadow has loomed over my hometown team, the Toronto Maple Leafs, for two decades now. The Leafs have faced Chara in 6 out of their past 7 playoff berths, going back all the way to 2001 when Chara was still playing for Toronto’s provincial rival the Ottawa Senators. On the Bruins, the Leafs have faced Chara in three Game 7s — and are 0-3 against him. That record includes this Bruins comeback in 2013, one of the more devastating losses in Leaf history. Watch as Chara helps Boston score consecutive goals with the Bruins’ own net empty with about a minute left in the game, first with his slap shot and then by using his giant body to screen the Leafs goalie. 

It is easy, of course, to overlook hockey stars in America, especially in Boston where all three of the other big sports teams have been so great at the same time. Most sports fans probably have not caught on to what Chara has done in his career, or, even more impressively, what Connor McDavid, barely more than half Chara’s age, is now doing (McDavid has been sparking hockey’s own GOAT debate — if you haven’t watched any Oilers hockey since Gretzky, try watching a period this year and you will immediately see why). It’s good, then, that Chara is going to a team where another physically imposing hockey great, Alex Ovechkin, has helped put hockey on the map. Going into this season, which starts next week, both Washington and Boston are near the top of the betting odds to win the Cup. 

Chara may also have one more run left in him on the international stage, with the 2022 Beijing Olympics now a year away. Thus far Chara has won two silvers with Slovakia at the World Championships (a tournament held annually), but has not yet won a medal at the Olympics, where the best that the Slovaks have finished is fourth place. 

This, of course, begs the question that often follows divorce: what if they had just stayed together? A post-Cold War Czechoslovakian team would have paired Chara — and other Slovak stars, such as Marian Hossa — with the Czech greats like Jagr and Hasek. This is one of the big “what if’s” of Eastern European sports, along with the much uglier breakup that put an end to Yugoslavia’s basketball teams. (Somewhere, in a more peaceful alternate reality, Doncic and Jokic are building an Olympic juggernaut together…). With no Czechoslovakia in play, the road to gold has been much easier for the likes of Canada (which, unlike Czechoslovakia, is actually a country divided by language), or for Olympic Athletes From Russia, than would otherwise have been the case. 

Indeed, all this is not just a eulogizer’s praise – Chara’s career’s not done yet.  Boston, it seems, wanted Chara to transition into becoming a penalty-kill specialist, but Chara still wants a bigger role than that, and Washington is prepared to give him one.  And if he can help his new team to one more Cup run, Chara will leave behind a hockey legacy that is nearly as unassailable as he is himself. 

T3: Trailer Trolleybus-Trolleytruck?

1942 Ford Semi-Trailer Bus - Other Truck Makes - BigMackTrucks.com
A 1942 Ford trailer bus

Trailer buses used to be fairly common. They were mostly abandoned by the early 1960s, having had a number of disadvantages when compared to conventional vehicles.  They often needed to employ both a driver (in the tractor) and conductor (in the trailer), for example, which greatly increased their operating costs.

München, Leopoldstraße 14.08.2017 | Münchner Verkehrsgesells… | Flickr
Munich trailer bus

A few cities, such as Munich, have recently started using some trailer buses again. Today’s trailer buses have the ability to automate away some of their previous disadvantages. They use digital payment systems and camera-monitors in order to do away with the need to employ a conductor, for instance. They also use rear-view and side-view cameras to help the driver keep the trailer from moving out of sync with tractor — and in the future they might use other driver-assist technologies to do so even more smoothly. The process of coupling and uncoupling the tractor and trailer from one another may also be automated, making it easier to add the trailer segment for use during rush hour and then decouple it from the tractor during less busy times of day when it is not needed.

In Switzerland, where hydro-electric dams have made electricity cheap and clean, trailer buses also come in trolleybus form:

Trailer trolleybus in Lausanne

Trailer buses can also be used to transport cargo, either by having a bus tow a cargo trailer, or by having a tractor unit alternate between towing a passenger trailer and a cargo trailer.  

bus trailer

This got me thinking again, in a roundabout way, about what might happen as companies automate the process of receiving deliveries of cargo, for instance by automating the unloading of trucks and the warehousing of their contents.  This would presumably allow more deliveries to take place at night, when warehouse labour would otherwise be expensive. It would therefore allow trucks to avoid daytime traffic.

Electric trucks would seem to be especially well-suited for nighttime deliveries, because they are much quieter than diesel trucks and because range-anxiety would be lessened when there are no traffic jams to get stuck in.

Best of all might be trolley-trucks, which, unlike battery-powered trucks, can operate 24 hours a day without needing to have their batteries charged. (Batteries, especially the big ones that  cargo trucks need, have many other economic and environmental costs to worry about too). Trolley-trucks would also benefit from avoiding daytime traffic, especially if they were sharing their wired routes with daytime trolleybuses, or if they were using small batteries to travel short distances off-wire. They would also benefit from the fact that electricity prices tend to be cheaper overnight than during the day.

Electric Trolley Trucks may come soon to North American highways ...

Perhaps most intriguing of all is the possibility of a trailer trolleybus-trolleytruck hybrid, where the trolley tractor unit pulls a passenger trailer during the day and a cargo trailer at night. Such a system could allow a city to buy fewer trolley tractor units, which tend to be more expensive than the trailers (especially if they have new, smart-pantograph wire-gripping technology) and then maximize their use by operating them close to 24 hours a day.   It might also help to address the first-mile/last-mile issues associated with trolley vehicles; namely, the question of how to use them off-wire. The cargo trailers could be pulled by diesel or battery-powered tractor units in off-wire areas, then decoupled and attached to trolley tractor units for the the wired section of the route.

This process might be useful for the passenger trailers too, if there are times when non-wired bus routes become in higher demand (a sporting event, for example), or if electricity is unavailable (if there is an extended blackout, for example), or if extreme weather puts the trolley wires out of service.

Countries’ True Heights

If for some reason you are trying to look up the average heights of people in various countries, you’ll quickly find that the tallest countries are in northern Europe or the Balkans, where men are just over 6 feet on average and women are around 5″7. The shortest are certain poor countries, such as Bolivia, where men average 5″3 and women between 4″8 and 5″0 feet.

I don’t know if these statistics are correct (the 4″8-5″0 range given for Bolivian women is suspiciously imprecise, for example). Even if they are accurate, you may have noticed that they do not really answer the question. These statistics are giving the heights of adult men and women. But, of course, not everyone is an adult, and not all countries are split equally between men and women.

If (again, for some reason) you are interested in finding the true heights of countries, you have to adjust for children and gender. Once you do this, you realize that the country with the shortest people on average is probably somewhere like Niger, where 50% of the population are under 15 years old. The country with the tallest people on average is probably somewhere like Qatar, where men outnumber women 3.4 to 1 and where only 14% of the population are under 15 years old. The true average height of Niger may be something like 4 feet; the true average height of Qatar may be something like 5″6.

[Actually, these estimates are probably way off too, since they do not take into account babies. The tallest places are presumably those which have had the fewest babies in the past year or so, and the shortest places those which have had the most].

This same adjustment can be applied to cities: the megacity of Lagos (in Nigeria, not Niger) is 50% children, whereas Tokyo is 12% children. Las Vegas, especially on the weekend, is perhaps the tallest — the most male and the most adult — big city in the United States, somewhat similar to the glitzy desert cities on the Persian Gulf coast.

Now, does knowing this have any real-life utility? No. But it might be cool if a city like Lagos were to build, say, a low-ceiling double-decker bus that only kids could comfortably use.


A Pan-Canadian NFL Team

Canada does not have a single stadium that would be suitable for an NFL franchise.

Canada does, however, have four stadiums that are not too unsuitable: Montreal’s Olympic Stadium, Edmonton’s Commonwealth Stadium, Vancouver’s BC Place, and Toronto’s Rogers Centre.

Instead of a city like Toronto spending $1 billion or so on a new stadium in order to get an NFL team of its own, Canada should aim for an NFL team that splits its eight home games per season between four existing stadiums in Toronto, Montreal, Vancouver, and Edmonton. The team – the Canada Moose? – would play in Montreal and Edmonton during September and October, then in Toronto and Vancouver during November and December.

Montreal’s stadium can seat 66,000 people, but cannot be used in the winter. Edmonton’s can sit 56,000 people, but cannot easily be used in the winter either. Vancouver’s can sit 54,500 people, plus it has a retractable roof (and a much warmer climate) that allows for winter use. Toronto’s can only sit 49,000 people, but also has a retractable roof. No other stadiums in Canada are close to as big as these. The next largest, in Calgary, seats 36,000 people.

By way of comparison, the average NFL stadium can sit 69,500 people. Unlike in Montreal, Toronto, or Vancouver, most NFL stadiums were purpose-built for football. Still, this should not necessarily be a deal-breaker for Canada, because the NFL does not actually derive a large share of its revenues from ticket sales. Rather, revenues come from TV (and increasingly, Internet) deals and merchandising — which could make a Canadian franchise profitable, as it might have one of the biggest markets in the entire league.

What about playoff home games, where would they be played? Well, to begin with, no team ever plays more than 2 home playoff games per year, so the lack of a good-quality playoff stadium is probably not as big a problem as one might think. (Canada’s current favourite team, the Bills, have not played a single home playoff game in the past 24 years). When a home playoff game does need to be played, some or all of the following options could work:

  • Get Montreal to shell out the estimated $200 million needed to repair its retractable roof, thus allowing it to host some playoff games. There are already plans for such repairs to take place in advance of the FIFA World Cup in 2026
  • Make the venue conditional on the opponent. If playing a home game against Seattle, for example, play in Vancouver. If playing against Buffalo or Detroit, play in Toronto. If playing against New England, play in Montreal (if Montreal has fixed its roof).
  • Play a Superbowl-style Conference Final: if Canada’s team makes it to the semi-finals with home field advantage, play the game in a massive college football venue like Ben Hill Griffin Stadium (home of the Florida Gators; capacity 88,500), to which Canadian snowbirds could flock in January. Give all Canadian citizens first dibs on tickets.

Of course, the idea of a pan-Canadian team is not without risk. Such a team might have difficulty attracting good players. Canadian patriotism might not be sufficient to win new fans to the sport. It might just be too weird to have American cities competing against an entire country (although that has not limited the Raptors’ or Blue Jays’ aggressively national marketing strategies). Combined with the lack of a good stadium, these risks are not insignificant.

On the other hand, look at the possible rewards. You get a huge market in Canada, no wasting taxpayer dollars on a new football stadium, and no direct competition with the CFL or other pro sports franchises. The NFL also gains a lot of flexibility in scheduling, since it can have west coast matchups played in Vancouver or Edmonton and east coast matchups in Toronto or Montreal.

It would also create great local matchups. Canadian games against Seattle, Buffalo, Detroit, Cleveland, and New England would be especially fun. So would Canadian games in snowbird states like Florida, Arizona, Nevada, or California around Christmas-time. Even isolated cities like Green Bay, Minneapolis, Denver, and Kansas City would be able to pull in some tailgaters from the Prairie provinces.

I think the rewards outweigh the risks. For Canada to get an NFL team, without having to build any huge new football stadiums, would be as good as Thanksgiving coming twice a year.


Back-of-the-Envelope World City


1000 square feet per person, for an apartment with a large balcony, is very comfortable.

For 10 billion people to have 1000 square feet is 10 trillion square feet.

With 10-story buildings, 10 trillion square feet would require 1 trillion square feet of land.

1 trillion square feet is 92,900 square km…let’s round that up and say 100,000 square km.

Throw in another 60% more land for roads, parks, etc. – also a very comfortable amount, especially if you don’t waste most of it on cars –  and you get 160,000 sq. km.

160,000 square km is a square with sides of 400 km. It is about the size of Wisconsin, or Tunisia.

In Canada alone, there is already an estimated 126,000 square km of urban land.

If you don’t like living in a spacious apartment 10 stories high, but prefer instead that your city have 5-story buildings, you would need 2 trillion square feet of land for apartments. That’s 186,000 square km of land. Add another 60 percent for public space and you get a bit under 300,000 square km. That’s a square with sides of about 550 km. That’s the size of Arizona, or Italy.

Whatever else we may lack, space is not it.








This article is about earthquakes and other disasters. I hope it’s informative, rather than just morbid or ill-timed or pointless..

With an estimated 144,000 deaths so far as a direct result of the Covid-19 virus, this is the first disaster in the past decade to have killed at least 50,000 people. However, it is the seventh to have done so in the past 15 years. There was the Indian Ocean tsunami in December 2004, which caused an estimated 230,000 deaths, the 2005 Kashmir earthquake (~87,000 deaths), the 2008 Burma cyclone (~138,000), the 2008 Sichuan earthquake (~88,000), the 2010 Haiti earthquake (~220,000-316,000) and the 2010 Russian heat wave (~56,000). Covid-19 may prove to be by far the worst of these disasters, but for now at least it has not been the deadliest.

One obvious lesson here is the destructiveness of earthquakes and earthquake-triggered tsunamis. They  caused 4 out of these 7 disasters, including the two deadliest to have occurred so far.

Financially speaking too, earthquakes have usually been the most devastating disasters. According to Wikipedia, the most expensive disaster was the Japan earthquake and tsunami in 2011, which caused approximately 16,000 deaths (2,203 of which were related to the Fukushima nuclear disaster it caused) and an estimated 411 billion inflation-adjusted dollars worth of damage. That same year, the Christchurch earthquake in New Zealand cost an estimated $44 billion, itself one of the most expensive modern disasters. Second costliest was another Japanese earthquake, in 1995 in Kobe (6,400 deaths; $330 billion). Third place was the 2008 Sichuan earthquake (88,000 deaths; $176 billion). The next five were hurricanes in America, all since 2005 (Katrina); three in 2017 alone (Harvey, Maria, Irma). Yet even the 2017 hurricane season as a whole cost less than either of Japan’s big earthquakes.

Of course, these do not come near the figures of the deadliest epidemics, such as the 1957-1958 Asian flu (~2 million), the 1968-1969 Hong Kong flu (1 million), or the AIDS epidemic (~32 million in its 60 years, for an average of 530,000 per year, with a peak of 1.7 million deaths in 2004). Nor (as we have often been told lately) do they approach the number of deaths from other horrible problems, such as car accidents (~1.3 million per year). They also don’t come near the death tolls from the very worst natural disasters, like the floods that occurred in in northern China in 1887 (~900,000-2 million, perhaps half of whom died because of a resulting pandemic and famine) or in 1931 (~400,000-4 million).

These Wikipedia statistics obviously need to be taken with a huge grain of salt. They often range widely: the death toll estimates even for the recent 2010 Haiti earthquake, for example, go from 46,000-85,000 (according to a report made by the US Agency for International Development) to 160,000 (according to a University of Michigan study) to 316,000 (based on numbers from the Haitian government). The death toll from the 1976 North China earthquake, perhaps the deadliest post-WW2 natural disaster, ranges from 240,000-650,000.

All of these estimates may also overlook indirect causes of death and destruction, and certainly they do not include the significant non-fatal consequences disasters usually cause. The 2015 Nepal earthquakes, for example,  led to around 8,000 deaths, but 3.5 million people were made at least temporarily homeless by them.

Nevertheless, these numbers do show that the deadliest disasters in recent years have tended to be earthquakes. Searching online right now, I see that I am hardly the only catastrophist to wonder what would happen if  “the big one”  were to occur in an earthquake-prone area like Tokyo or the US Pacific Northwest in the immediate future, while the current pandemic is still going on. The probability of this happening is small, but not zero. I recommend reading this New Yorker article, which was awarded a Pulitzer Prize, about this subject.

One of the regions impacted most by the virus thus far, the Mediterranean, is also among the most seismically active, ranging from countries with a medium risk of earthquakes, such as Italy (its deadliest modern disaster was the Messina earthquake in 1908, with an estimated 75,000-123,000 fatalities), to those with a high risk of earthquakes, like Turkey. Iran too, which has suffered the most deaths from Covid-19 of any country outside of the US or Europe, is a high-risk country where earthquakes are concerned. It experienced a deadly earthquake in 1990 (50,000). 

China’s Hubei province, of which Wuhan is the capital, is itself used to earthquakes. The province experienced an earthquake this past Boxing Day, just five days before Chinese authorities first told the World Health Organization that there was an unusual pneumonia in Wuhan, less than a month before much of the province went into quarantine.

Historically speaking, northern and central China have suffered some of the deadliest earthquakes, in large part because of how populous they are. Before the terrible ones in Sichuan in 2008 and Hebei in 1976, there was the Gansu-Ningxia earthquake in 1920 (273,000). [Three years after that, the 1923 Great Kanto earthquake in Japan (100,000-143,000) destroyed large parts of Tokyo and was, at the time, probably the most destructive disaster experienced by a modern industrial city]. According to Wikipedia, possibly the deadliest ever earthquake occurred in Shaanxi, in 1556, killing more than 800,000 people.

Before the horrific Indian Ocean tsunami of December 2004, other recent big, deadly disasters include the 2003 European heat wave (70,000), the 1991 Bangladesh cyclone (140,000), the 1976 North China earthquake (240,000-650,000), the 1975 typhoon and resulting Philippine dam failure (230,000) and the 1970 East Pakistan (Bangladesh) cyclone (500,000+).

There have also been a number of disasters with death tolls in the 10,000-50,000 range: earthquakes in Gujarat in 2001 (20,000), Turkey in 1999 (17,000),  Iran in 1990 (50,000), and Armenia in 1988 (28,000). The only non-earthquake disasters in this range during the past few decades were a volcanic eruption in Colombia in 1985 (23,000), and cyclones in Central America and Mexico in 1998 (11,000) and Bangladesh and India in 2007 (15,000).

Certain places have been struck repeatedly by large earthquakes. The most notable of these may be Valdivia, in Chile. It experienced the most powerful earthquake on record, in 1960, an earthquake so powerful that by itself it accounted for roughly 25 percent of the world’s seismic energy released in the 20th century. (The next two biggest in the century, in Alaska and Sumatra, together accounted for roughly another 25 percent). The first really big earthquake recorded was also in Valdivia, in 1575, according to Wikipedia.

The next three big ones after that, all in the 1600s, were in Chile as well, including one in the capital, Santiago. Valparaiso (in central Chile, near Santiago) was then hit with big ones in 1730 and 1822, and Conception (on the coast between Valdivia and Valparaiso) in 1751 and 1835.

The other area to flag in this regard is the island of Sumatra, in Indonesia. It has been hit with one of the only two recent earthquakes with a magnitude of at least 9; namely, the deadly Indian Ocean earthquake and tsunami in 2004. (The other magnitude 9+ magnitude quake was the costly Japan earthquake in 2011; until then most experts had not believed that an earthquake above 8.4 was even possible in Japan). Before that, no 9+ magnitude earthquakes occurred since Alaska in 1964 or Chile in 1960. A magnitude 9 is about 33 times more seismically powerful than a magnitude 8, and over 1000 times more powerful than a magnitude 7. Sumatra was also hit by two of the three only recent earthquakes in the magnitude 8 range (in 2012 and 2005). The other was just off the coast of Conception in Chile, in 2010. Before 2004, there was no magnitude 8+ since Alaska in 1965.

Solar-Based Space Power

Hello! I hope you are all doing as well as possible right now. Since many of us are stuck inside for the time being, I’m going to try to revive this blog a bit.

For this first post, let’s talk about space-based solar power. The idea is a popular one; it is exciting (outer space!) and has some logic to it (there are no clouds, nights, seasons, or land-use-constraints in space!). Sadly, though, if you look into the topic – for example, if you read this very well-thought-out piece from Do The Math –  you see that even with extremely optimistic assumptions, it seems unlikely that sending energy from solar panels located in space to earth will become economically worthwhile any time soon, if ever.

And yet, perhaps these analyses are missing something. From what I can see — with the huge, twin caveats that I haven’t looked too deeply into the subject, and that I understand almost nothing about the physics involved in it — recent discussions about space-based wireless power transmission have been limited almost entirely to the idea of generating power in space and sending it to earth in order to provide civilians with clean, reliable energy. There might, however, be alternative uses and methods for wirelessly-transmitting energy via space. For example:

  •  sending space-based solar power to military outposts, in order to provide soldiers with power that is not reliant on vulnerable supply lines, is not bulky to haul around from place to place, and is not intermittent. This is how I first heard about the topic of space-based solar power: George Friedman discusses it in his book The Next 100 Years.
  • the same purpose as above, except that instead of the military outpost receiving power that is generated from space-based solar panels, it would instead receive power that is generated conventionally on earth, then ‘triangulated’: sent up to space, then back down to a different location on earth. Such a system could perhaps also work in tandem with space-based solar panels. Over time, for example, more panels could be launched, so that as the years go by the system would use more power generated in space and less power generated on earth. [Or, perhaps, if the system was located as an array of satellites in Low Earth Orbit (500-2000 km),  rather than much further away in Geosynchronous orbit (40,000 km),  having the system use power generated by earth-based sources might allow it to provide power 24/7, as would otherwise not be possible for a Low Earth Orbit system because Low Earth Orbit satellites spend about a third of the time being eclipsed by the earth, preventing any solar panels in such orbits from receiving sunlight at those times]
  • Because wars are themselves intermittent, a space-energy system built for military purposes might be able to double as a civilian system during times when military demand is low. […Also, if the system was built as an array of satellites in Low Earth Orbit, rather than in Geosynchronous orbit, then each satellite in the array would only pass over a given military outpost or region on earth for a very short amount of time.  Most of each satellite’s orbit in such a system might be freed up for civilian uses as result].
  • There is also the question of how to provide satellites with energy, so that satellites themselves can be powered, both for military and non-military reasons. Militarily, for example, if satellites were to be physically attacked, it might perhaps be the case that their ability to protect themselves from any incoming projectiles would depend on whether or not they have more energy available to them than do the projectiles, which they could then outmaneuver. Thus it might be useful for satellites to receive power wirelessly, either from earth or from other satellites. And again, once such a system is built, it might also be able to find non-military uses, particularly since the military might not need to use the system much during peacetime. The system might then be available to power satellites for non-military uses. Or perhaps it could be used to power locations on earth

Obviously, I have absolutely no idea what I’m talking about here. So I’m asking, is there anything to these ideas? Are there other similar ideas that I’ve left out? How might these factors change the math when it comes to thinking about the future economics of energy in space?

Finally, if the civilians-piggybacking-on-the-military-surplus-capacity-of-triangulating-energy-from-earth-to-space-back-to-earth idea is anything other than totally ridiculous, which power sources would be best suited for it? Would solar panel companies in the Australian Outback benefit, for example, by being able to wirelessly send their otherwise-remote, Southern Hemisphere-summer energy to military and/or civilian locations in other parts of the world?










2019 Year in Review: GDP Growth

An in-depth version of this article was originally published on Rosa and Roubini Associates

GDP can often be a misleading measurement, and a year can sometimes be a misleadingly short period of time to measure. A review of a past year’s GDP growth trends may nevertheless serve as a useful starting point for understanding the world’s markets. Carrying out such an exercise in economic hindsight for 2019, we might settle upon the following list of approximate growth trends:

  1. Slowing growth occurred in all major regions and countries

    Global growth in 2019 was estimated to have been 3%, down from approximately 3.5% in recent years. This trend also held at both the regional and national levels. Regionally, North America, Europe, and Northeast Asia all faced slowing growth. Euro Area growth slowed from 1.8% during 2018 to 1.2% in 2019; US growth slowed from 2.8% to 2.2%; China’s growth slowed from an estimated 6.6% to 6.2%. (Elsewhere in Northeast Asia, Japan’s growth remained low at around 1% and South Korea’s slowed from 2.6 to 1.8%). No major country saw an increase in its growth rate, except perhaps a slight increase in Japan’s.


  1. America, China, and South Asia provided most of global growth

    With European and Japanese growth little greater than 1%, and with many commodity-exporting economies struggling too, global growth was carried mainly by the United States, China, and to a lesser extent India and other countries in southern Asia. US growth was estimated at 2.2 percent, which given its size (roughly 25% of global GDP), and the slow growth of global economy, is still a substantial portion of the world’s total growth this year. China’s 6.2% growth (assuming this figure is accurate) is even more substantial. India, meanwhile, which is only around 3% of global GDP in nominal terms (7.5% in purchasing power parity-adjusted terms), experienced 4.9% growth this year. Other smaller South Asian economies grew even more quickly, such as Bangladesh (7.7%), Vietnam (6.5%), Indonesia (5.1%) and the Philippines (5.7%). Thailand, however, which is by far the largest economy in Southeast Asia apart from Indonesia, grew only 2.4%.  
  1. Europe continued to struggle – and not just in the European Union

    The EU’s growth in 2019 is estimated to have been below 1.5%. The Euro Area’s growth was even lower than that, because unlike the European Union it does not include the faster-growing East European economies, notably Poland and Romania with 4% growth and Hungary 4.6% growth. Even outside the EU growth was slow, however. Russia’s growth this year was only an estimated 1.1%, down from 2.3% in 2018. Norway’s was 1%; Switzerland’s 0.8%. Britain’s was 1.2% (that is, assuming you consider Britain as outside the EU). And Turkey’s GDP, after growing at 2.5% in 2018, did not grow at all in 2019.
  2. Central Europe in particular experienced slow growth 

    Perhaps the most notable regional trend in Europe was the slow growth within Central Europe, most notably in the Germany-Switzerland-Italy corridor of nations. Germany and Italy had by far the slowest growth among G7 economies: Germany grew at 0.5% (down from 1.4% in 2018), Italy grew at 0.2% (down from 0.8% in 2018). Most countries around them also had slow growth: France 1.3%, Belgium 1.3%, Sweden 1.3%, Austria 1.5%, the Netherlands 1.7%, Switzerland 0.8%. Even the Czech and Slovak economies slowed, to around 2.5%, down from the 3-4% range they had grown at in previous years. The Central European slowdown was probably the dominant trend in the EU in 2019. The previous dominant trend, namely Southern Europe’s slow growth, did not disappear (Italy, after all, still struggled) but it was eclipsed. Spain’s economy grew at 2.1%, Greece 1.9%.


  1. Europe’s North-South dynamic has become more complicated

    There is no longer any clear divide between a sluggish South and nimble North, either within the EU, the Euro Area, or Europe more broadly defined. At all three levels, the fastest and slowest major economies in 2019 were both Southern states: Spain was the fastest, Italy the slowest. Northern Europe was divided too: major economies such as Germany, Britain, Russia, and Scandinavia (ex-Denmark) grew slowly, while others like Poland, Ireland, and to a lesser extent the Dutch and Danes grew quickly. In the ex-EU Mediterranean region there were divides too: Turkey did not grow, but the Levant grew quickly (Israel 3.2%, Egypt 5.6% for e.g.). In the Maghreb, Morocco grew at 2.5%, Algeria 2.6%.

  2. Latin America had a rough year…

    Venezuela remained in crisis, and Argentina experienced a recession in which its GDP shrank by an estimated 3.3% in 2019. The two largest economies, Brazil and Mexico, grew at just 0.8% and 0.1%, respectively. The Pacific economies that had previously been strong, such as Chile and Peru (both significant commodity exporters), slowed as well. Chile grew by 1.8%, Peru 2.6%. Colombia’s growth did rise however, from 2.6% to 3.1%.

  3. so did the Anglosphere

    The Anglosphere is a tricky group to define. Arguably it is does not even warrant being considered as a group to begin with. Even for those who do think the concept is useful, it is difficult to know which countries it should include. Certainly, it includes countries like Britain, Canada, Australia, and New Zealand. More broadly, it could perhaps also be used to include economies such as Singapore, Hong Kong, South Africa and/or Nigeria. Wherever you do decide to draw the Anglosphere’s lines, the group had a year of slow growth. Britain grew at 1.2%; Canada and Australia at 1.6%. Singapore grew at just 0.8%; Hong Kong actually shrank by 0.3%. South Africa grew by 0.6% and Nigeria (starting at a lower income base) grew by 2.2%. Jamaica grew at 1%. Only New Zealand and Ireland had strong growth, at 2.5% and 4.2%. Ireland’s growth slowed too though, from 6.7% in 2018.

  4. East Africa grew quickly, but Africa in general did not 

    Rwanda may have led all countries in 2019, with 7.8% growth. Ethiopia may have led among all large developing countries, with 7.4% growth. Uganda, Kenya, and Egypt all grew between 5-6%. There were high growth numbers in some other parts of Africa too, but in the largest regional economies, such as South Africa, Nigeria, Angola, and Algeria, growth was slow. Nearby in the Middle East, the Gulf Arab states’ GDP stalled and Iran’s shrank.

  1. In North America, the US kept ahead of Canada and Mexico

    US growth was 2.2% in 2019, compared to an estimated 1.6% in Canada and 0.1% in Mexico. This is the second year in a row that the US grew the fastest of the three. Before then, not since the 2009 recession did the US do so. And before then, not since 1999 was US growth the fastest. (The US grew at 4.7% in 1999, more than double its current pace). In contrast, as recently as 2014 the US grew slower than both Canada and Mexico. 


  1. In America and China both, heartlands outgrew coastlands

    Unlike in the previous two years, US growth in 2019 seems to have occurred at a faster pace in the centre of the country – in the Rockies, the Greater Midwest, or in certain areas along the Gulf of Mexico – than it did along the eastern or western coasts. In China, somewhat similarly, the interior states in the south-west, centre-west, or central China, such as Yunnan, Jiangxi, Hubei, and Sichuan, grew faster than most of the country’s coastal states. The slowest-growing Chinese region of all was, as it has often been in recent years, the northeast: states like Heilongjiang, Jilin, Liaoning, and Inner Mongolia.