Mixing and Matching

In public transportation as in many other aspects, an important fact of improvement is being able to mix-and-match things that work from different sources. It’s rare to have a situation in which exact importation of one way of doing things is the best in every circumstance (and the Covid-19 crisis appears to be one of these rare situations, Korea being the best). More commonly, different comparison cases, whether they’re companies in private-sector consulting or countries in public-sector policy research, will do different things better. Knowing how to mix-and-match is an important skill in competently learning from the best.

Non-transport examples

I put this up first, but want to emphasize that this is outside my skill set so I am less certain about the examples here than in transport; I bring them up because some of the sanity checks are cleaner here.

Secondary education: high-income Asia consistently outperforms the West in international math and science tests. However, two important caveats complicate “just be like Asia” reform ideas, like the popularity of Singapore math textbooks in some segments of the American middle class. The first is that Japan, South Korea, and Taiwan are a lot more monolingual than European countries like Germany and France, let alone smaller European countries like the Netherlands. And the second is that many things that are common to East Asia (and Singapore and Vietnam), like high social distance between hierarchs and subordinates or teachers and students, are completely absent from Finland, which is nearly the only Western country with math scores matching those of Asia. So the actual thing to learn from Asia is likely to be more technical and less about big cultural cleaves like making students wear uniforms and be more obsequious toward teachers.

Public health: whereas the Covid-19 crisis specifically still looks like a clean Asia vs. West cleave, overall public health outcomes do not. Japan has the world’s highest life expectancy, but then Mediterranean Europe follows it closely. The United States, which overall has poor health outcomes, near-ties Singapore and Sweden for lowest first-world smoking rate – and even though Singapore and Sweden both have good outcomes, they both have rather unhealthy diets by (for example) Levantine standards. Public health is a more complex issue than transportation, one that unfortunately low-life expectancy developed countries like Germany and Britain, let alone the US, aren’t meaningfully trying to learn in – and it’s not even clear how easy it is to import foreign ideas into such a complex mostly-working system, in contrast with the near-tabula rasa that is American public transportation.

Transportation in cities of different sizes

Alexander Rapp’s excellent list of metro areas ranked by what he calls frequent rapid transit ridership – that is, trains and buses that run every 20 minutes or better and are either grade separated or have absolute crossing priority with gates – showcases patterns that vary by population.

On the one hand, Tokyo is far and away the highest-ridership city in the world, even per capita. It has around 400 annual rail trips per capita. My recollection, for which I don’t really have a reliable source, is that 60% of work trips in the Tokyo region are done by rail (this data may be here but copy-paste for translation doesn’t work), a higher share than in major European capitals, which mostly top in the 40s.

On the other hand, this situation flips for smaller cities, in the 2-5 million metro population range. Sapporo appears to have maybe 120 annual trips per capita, and Fukuoka probably even less. In Korea, likewise, Seoul has high ridership per capita, though not as high as Paris, let alone Tokyo, but Busan has 100 trips per capita and Daegu 65. In contrast, Stockholm approaches 200 trips per capita (more including light rail), Vienna maybe 180 (growing to 220 with a much wider definition including trams), Hamburg 170, Prague 200 (more like 300 with trams), Munich maybe 230.

This doesn’t seem to be quite a West vs. Asia cleave. There is probably a shadow-of-giants effect in Japan leading smaller cities to use methods optimized for Tokyo; it’s visible in Britain and France, where Stockholm- and Munich-size cities like Birmingham, Manchester, and Lyon have far weaker transit systems. The US has this effect too – New York underperforms peer megacities somewhat, but smaller cities, imitating New York in many ways, are absolutely horrendous by the standards of similar-size European or East Asian cities. Nonetheless, the shadow of giants is not an immutable fact making it impossible for a Sapporo or Birmingham or Lyon to have the rail usage of a Stockholm – what is necessary is to recognize this effect and learn more from similar-size success stories than from the far larger national capital.

Construction costs and benefits

Construction costs are not a clean cleave across cultural regions. The distinction between the West and Asia is invisible: the worst country in the world is the United States, but the second worst appears to be Singapore. Excluding the English-speaking countries, there is a good mix on both sides: Korea, Spain, Italy, and the Nordic countries all have low costs, while Taiwan and the Netherlands have particularly high ones.

Moreover, countries that are good at construction are not always good at operations. As far as I can tell from deanonymizing CoMET data, Madrid has slightly higher metro operating costs than London, Paris, and Berlin, PPP$7/car-km vs. PPP$6, with generally high-construction cost Tokyo appearing to hit $5.

This is not even just costs, but also the ability to build lines that people ride. Tokyo is pretty good at that. Spain is not: the construction costs of the high-speed rail network are consistently lower than anywhere else in the world, but ridership is disappointing. There is no real integration between the AVE network and legacy trains, and there is a dazzling array of different trains each with separate fares, going up to seven incompatible categories, a far cry from the national integration one sees in Switzerland.

There is likely to be a clear answer to “who is best at optimizing construction costs, operating costs, and ridership?”: the Nordic countries. However, even there, we see one worrying issue: for one, Citybanan is expensive by the standards of the Eje Transversal (though not by those of the RER E or especially the second Munich S-Bahn tunnel), which may indicate difficulty in building the kind of multistory tunneling that bigger cities than Stockholm must contend with. Thus, while “be like Sweden” is a good guideline to costs, it is not a perfect one.

Optimizing frequency

The world leader in high-frequency public transportation is Paris. Its driverless Métro lines, M1 and M14 and soon to be M4, run a train every 85 seconds in actual service at rush hour. This is an artifact of its large size: M1 has such high ridership, especially in comparison with its length, that it needs to squeeze every last train out of the signaling system, unlike Berlin or Milan or Madrid or Stockholm. London and Moscow run at very high frequency as well for the same reason, reaching a train every 100 seconds in London and one every 92 in Moscow.

Tokyo, sadly, is not running so frequently. Its trains are packed, but limited to at best one every 120 seconds, many lines even 150, like New York. One possible explanation is that trains in Tokyo are so crowded that peak dwell times must be long, limiting throughput; long dwell times have led to reductions in RER A frequency recently. However, trains and platforms in Tokyo have good interior design for rapid boarding and alighting. Moreover, one can compare peak crowding levels in Tokyo by line with what we know is compatible with a train every 100 seconds in London, and a bunch of Tokyo subway lines aren’t more crowded than London’s worst. More likely, the issue is that Japanese signaling underperforms European systems and is the process of catching up; another aspect of signaling, automation, is also more advanced in France than in Japan (although Seoul, Taipei, and Singapore all have driverless metros).

This way, cities that are either extremely expensive to build in, like London and Moscow, or about average, like Paris, show the way forward in ways that cities that do other things better do not. It’s important to thus simultaneously learn the insights of small cities in reducing operating and construction costs and maintaining high-ridership systems, like the Nordic capitals, and those of megacities in automation and increasing throughput.

Can mixing and matching work?

Why not? In small cities with successful systems, it can’t be due to some deeply-ingrained culture – what do Stockholm, Zurich, Prague, Munich, and Budapest even have in common, other than being European? They’re not all national capitals or even all national primate cities, a common excuse New Yorkers give for why New York cannot have what London and Paris have.

Likewise, what exactly about French culture works to equip Métro lines with signals allowing 42 trains per hour per direction that cannot be adopted without also adopting real problems France has with small-city regional rail, fare integration, or national rail scheduling?

These are, ultimately, technical details. Some are directly about engineering, like Parisian train frequency. Some involve state institutions that lead to low construction costs in Spain, Korea, and the Nordic countries – but on other metrics, it’s unclear these three places have state capacity that is lacking in high-cost Taiwan, Germany, and the Netherlands. So even things that aren’t exactly about engineering are likely to boil down to fairly technical issues with how contracts are written up, how much transit agencies invest in in-house engineering, and so on.

There’s a huge world out there. And an underperforming transit agency – say, any in the United States – had better acquire all the knowledge it can possibly lay its hands on, because so many problems have already been solved elsewhere. The role of the locals is not to innovate; it’s to figure out how to imitate different things at once and make them work together. It’s not a trivial task, but every pattern suggests to me it’s doable given reasonable effort.


  1. Gok

    > The role of the locals is not to innovate

    Uhg, just no, no. There is autonomous high frequency rapid transit because agencies took bets on innovation. Aversion to innovation is a great way to push transit further down the spiral of incompetence and mediocrity, particularly in the US where effective people mostly want to innovate, and jobs doing new stuff are so widely available and high paying compared.

    • Alon Levy

      Because agencies that were already near the technological forefront decided to innovate, not because some American backwater with dirty stations and delayed trains thought some app would be a better use of planning resources than running reliable service.

    • RossB

      It is surprising that South Korea has relatively low transit ridership given the density of its cities. Busan has got a lot of people in very high density areas. Other cities like Daejon and Gwanju have plenty as well, given the fairly small size of the cities. Maybe there are lots of people walking and biking?

        • RossB

          Are you thinking it is a measurement thing? Maybe. Let me just compare the Metros of a couple cities:

          Daegu versus Sapporo. The Sapporo Metro is shorter, and has fewer stops. Yet it has significantly more ridership. Now compare the cities themselves. Daegu is not only a bigger city, but has way more high density areas (like most of the cities in South Korea). Either there is a difference in the measurement, or lots more people are riding the subway in Japan, especially when you account for overall population and population density. I’m sure if you looked at similar European cities, they would blow away Daegu. I can’t imagine that many people are driving — you just can’t drive that many cars if the density is that high. Maybe they are all taking the bus, riding their bikes or walking. I don’t know, but I find it interesting.

          • Andrew in Ezo

            In Daegu, it could be commuters are utilizing bus transport extensively, including the green-colored “mauru” minibuses that run on smaller roads and thoroughfares where the regular sized city buses can’t go.

          • Luke

            My impression is that the older dense areas of Korean cities are so dense that a lot of people do just walk, but newer areas are fairly autocentric, with tower-in-a-park developments and streets wider than is really useful. While those developments are often built on or near new or old rail transit stops, the coverage of rail systems is nowhere near as comprehensive as in Japan or many European cities. With those developments often being at least a handful of 20-40 floor apartment towers, people end up using buses a lot, instead, but automobile dependency is higher than in Japan or most of Europe because a lot of Korean planners were forged on the anvil of mid-century American Modernism, though the tide is changing. I find it especially perplexing that there’s not more backlash against this mindset considering how good Korean rail services (where they exist) are, and that gas prices in Korea comparable to Japan and some parts of Europe.

          • adirondacker12800

            Saying “I took the bus” doesn’t brag about how much money you have.

          • adirondacker12800

            They do all the time. I’ve been in luxury cars. The rare wood veneer on the dashboard doesn’t work any better than the cleverly printed plastic wood grain on my dashboard. Doesn’t look that much better either. And I can wash mine with any old household cleaner. I don’t need to find someone named Heinz who has the special potion that has to be applied in just the right way. I don’t even have to find anybody because swiping it down with whatever is handy is less work than finding someone.

  2. Nilo

    A tangential, but related idea I’ve been thinking about. You’ve talked quite a bit about how every transit city you know of (except Singapore and NYC) has frequent electrified regional rail. What then are developing world countries looking to build good transit systems to do when so many of them don’t have large legacy rail networks. Obviously some places (India, Buenos Aires, southeastern Brazil) do, but most of Africa and a lot of Southeast Asia doesn’t. It seems like mixing and matching is much more difficult in these places, since they don’t simply have old rail networks to reappropriated for local passenger use.

    I’m obviously not sure what the solution is, perhaps something like the Japanese model, where developed world capital could build private lines to provide middle class residents an escape from traffic gridlock?

    • Eric

      I think you can see the solution by looking at places like India (which has some legacy rail but nowhere near enough), Bangkok, and one might even mention Addis Ababa. There are a number of arterial roads in these cities and elevated metro lines are being built over them. This is enough to provide for a large and busy metro network which broadly covers the whole metropolitan area. It isn’t enough to give everyone convenient metro access – but 20 years from now when these cities are richer, they can start building underground metro lines too. (Transfers will be annoying, but that’s a minor point)

      • Alon Levy

        India has a ton of legacy rail… and is still building completely separate metro systems; Mumbai’s first metro line already crosses the Suburban Railway without a transfer. The class valence is the opposite of the US – commuter rail for the poor, metro for the rich – but the incompatibility is similar, often even involving metros running standard-gauge rather than broad gauge, i.e. BART in reverse. So much Japanese investment, inc. a turnkey Shinkansen system, and so little learning from Japan’s metro-regional rail compatibility.

        • adirondacker12800

          Are you worried that they won’t be able to find multiple vendors with a broad catalog of things that are standard gauge… ever?

          • Nilo

            I’m pretty sure Alon’s concern is they want everyone to use metro tunnels as ways to extend suburban rail into the city center. If you’re metro is standard gauge and your regional rail is indian broad you can’t do that.

          • adirondacker12800

            They run the “suburban” trains without seats and without doors because they need the capacity. I doubt the are ever going to get the urge to run through to those.

          • F-Line to Dudley

            Project Unigauge is still thrashing away on Indian Railways after 28 years, now at 92% completion to 1676 mm broad “Indian” gauge as of March 2020. Network commonality at any one gauge has massively larger impacts on cost control than the fact that they chose something other than standard gauge. Indian Gauge (1676 mm) is third-most common in the world @ 10.2% and ~134,000 km after Standard Gauge (1435 mm; 54.9%, ~720,000 km) and “Russian” Five-Foot Gauge (1520 mm; 16.8%, ~220,000 km). With enough spread around the subcontinent to neighboring Pakistan, Bangladesh, Nepal, and Sri Lanka to have similar region-wide gravitational pull as Russian Gauge in the ex-USSR and Eastern Bloc. As well as adoption on other continents in Argentina and Chile, and interoperability with Spanish/Portugese Iberian Gauge stock (only 8 mm difference…slight operating restrictions if running 100.00% unmodified, but net-zero difference on design or procurement). You can get pretty much any rolling stock in the world catalog-sourced to the other two more widely used gauges ordered in Indian Gauge at little to no cost difference because of the network scale, as well as anything literally catalog-identical from Iberian gauge (1.2%, ~15,000 km). The narrow gauge and meter gauge hodgepodge they’re mass-replacing are the ones design-divergent enough (at 762-1000 mm) to inflict procurement premiums, limit overall buying options, and hamper savings from scalability. Which is why so much of the old stock that Unigauge is sweeping aside is ancient and low-quality from too many patch rebuilds in lieu of modern-era replacements. The pain and suffering of embarking on a standardization megaproject like Unigauge in the first place has major payoff in cost control-over-lifetime for all rolling stock procurements, in addition to the more obvious savings and revenue enhancements from finally having a universally operable network.

            The physical difference between Indian and Russian gauge is small enough to simply not require any consequential-cost design deviations for porting anything over, and standard gauge suppliers are stampeding all over the wholesale fleet replacements IR is having to make as a result of each new Unigauge milestone. American-as-apple-pie GE Transportation (now owned by American-as-apple-pie Wabtec), the world #1 in freight locos by a huge margin, is 5 years into an 11-year, $2.6B U.S. 1000-loco procurement for GE Evolution freight power…oh-so-very American “Buy India” final assembly clause and all. It’s GE’s single largest ongoing power procurement, and basically strengthens their market stranglehold tenfold by all the Service & Support goodies they’re offering IR on a procurement timetable that’ll put them in pole position for the first wave of rebuilds and replacements in 25 more years. Unigauge is basically the #2 worldwide God’s gift to the freight power leaders’ profit margins…with #1 being the new U.S. EPA emissions tiers forcing 15-year long mass turnover in Class I & II power rosters across the three NAFTA countries. It’s that huge, and that big at shaping the makeup of the whole industry. GE sold off its trains unit at a market high and Wabtec went all-in on the buy for the deep long-term tentacles that IR freight loco contract gives its market share. That all happened because Unigauge–so very massive to begin with–is now far along into realizing its payoffs.

            A lot of IR passenger stock is procured under homegrown “class” makes so it’s a little harder to parse whose global supply chain is sitting under the hood because the foreign partners don’t necessarily get their name on the finished product’s builder’s plate. But basically the whole who’s-who of global heavies (Alstom, Bombardier, Hitachi, CAF, Hyundai-Rotem, etc.) are all-in on the adaptations filling out IR’s passenger rosters–with the Chinese upstarts like CRRC itching for every means to get in–that it’s little different there from the rolling stock supply chain on all hemispheres of Planet Earth. Probably will see *somewhat* less Japanese vendor influence (except for the Hitachis and Kawasakis who play everywhere) with the purge of Metre Gauge (same as used in Japan), but that was an extremely tiny crossover market to begin with because so much of Japan is electrified and so little of Indian Metre territory ever was. Other than that, their supply chains within each self-branded passenger stock “class” conforms to the same globalization sourcing as what’s inside any other country’s stock.

          • adirondacker12800

            So for lines that rational people realize will be relatively isolated it doesn’t really matter if it’s standard gauge or Indian gauge? Or if a large city has both, in large fleets, it doesn’t matter much? Very vaguely like New York City’s A division and B division?

          • Herbert

            India double stacks containers under catenary and also double stacks on flatcars…

          • F-Line to Dudley

            Herbert: Yes…CSX partnered with IR for the pilot project of double-stacks under 25 kV electrification on a Unigauge-convert line. Pubbed a bunch of much-hyped industry white papers about how good it worked out for all involved, too. Which just goes to show you what a giant mountain of bullshit is behind any Class I’s “objections” to enhanced pax service on their trackage. Just go fishing for all that self-pub CSX did on this Indian joint venture the next time one of them bellyaches about freights coexisting under electrification. It’s just the cover story they run with for not getting their $$$ fancy properly tickled. The freights literally wrote the book on double stacks-under-25 kV passenger wire; that’s how comfortable they *actually* are dealing with those considerations at the right price.

            Now, U.S. transit agencies own their share of that unwillingness to work together. MassDOT, after years of butting heads with Conrail and successor CSX on the most basic-ass gains like individual Worcester Line passenger slots, eventually had its “Eureka!” moment in Class I relations about 15 years ago when it started negotiating the Beacon Park relocation project with CSX. Turns out if you pitch your pax upgrades as sympatico to CSX’s main intermodal profit center, they become your BFF’s overnight. And that’s how the $100M+ megadeal for Worcester Line ownership, hundreds of miles of future-considerations branchline ownerships, Beacon Park redev, and the double-stack project was born. A little “Pimp My Yard” quid pro quo goes a long way when it tickles the for-profit carrier’s actual largest growth sector. CSX went from stanfoffish gradual disinvestment in New England to eagerly cutting checks for the City of Worcester schools system under “good neighbor” memoranda, and basically rebooting the whole regional trucking industry to HQ itself in Worcester County. The state has already made back most of the cost of that transaction from the shipping boom, and that’s with Beacon Park still being a windswept moonscape under the ownership of Harvard Univ. And they just made yet another follow-through public branchline sale last month which trades some light freight fun bux for potential MBTA Franklin Line improvements, so it’s turned into basically a nonstop exchange with a few more “Pimp My Yard” moves left to tap when the state needs to mount the Boston-Springfield Inland Route.

            You can contrast that to how New York gets itself nowhere fast with the same players. They publish Empire Service study after study about all the things they want from CSX west of Albany…and not a word gets said about about how they plan to grease CSX’s profit center to get it done. So they get cockblocked on every tiny thing on the Water Level Route. “No, you can’t install cab signals and Amtrak’s ACSES PTC for simplicity; you have to work with our system.” “No, you have to space out your tracks X feet further than average from ours to run at 90 MPH.” “No, we need enormous buffering between our slots and yours that we demand nowhere else because fuck you, that’s why.” All of this when the east-west Water Level Route perennially runs UNDER capacity because of deficiencies vs. Norfolk Southern’s competing lane, compared to CSX’s north-south River Line which runs perennially over-capacity because it dominates its lane. And somehow NY has not put two-and-two together that this was EXACTLY the state of affairs in neighboring MA until the first time somebody bothered to ask them “tell us what YOUR biggest needs are.” They continue to let PANYNJ light money on fire pushing hella flawed Cross-Harbor Freight Tunnel studies no Class I has ever once asked for (but which will keep another generation of sandhogs fat and busy, so it’s all good!), but never once inquire about what kind of traffic distribution leads to maximal profit margins. I bet if there was even the faintest nod from Albany to talking about what infrastructure pieces from the umpteen Empire pax studies would improve CSX’s competitive standing vs. NS they’d be singing a different tune instantly. Maybe they need this, that, and a little of those way more than the other parts that do little for them…”but really, you can do whatever the fuck you want with your people choo-choos if you just give us a grant to improve DeWitt Yard throughput”. Sold!…that’s basically the exact same way MassDOT scored its breakthrough that keeps on giving. Or maybe NY should be baiting some grants for River Line capacity improvements to solve actual profit-threatening traffic bottlenecks as bait for pax permissions on the emptier Water Level Route…solve a problem in freight-only territory as means of securing greater payback in pax-shared territory. Same way MassDOT did the math on trucking industry coattails for its intermodal project and concluded “This isn’t an up-front overpay; this is a 25-year profitmaker in taxes!” Ironically, NY ignores the one place where they DID mimic the MassDOT template: Amtrak’s lease-to-buy acquisition of the Poughkeepsie-Hoffmans Jct. stretch of Empire for self-control and self-dispatch. Which CSX was willing to do because a non-antagonistic partner (Amtrak) was willing to take a hundred-plus miles of low-growth trackage with shot signaling plant off their backs, pay for all the state-of-repair renewal they didn’t want to, and give them some actual profit-padding IOU’s in the process like stamping out the last couple East-of-Hudson weight restrictions so the route off NYC’s Oak Point Link attains some longer-term growth prospects to fish for. Just that little bit of profit-rationalized negotiating is all it took to gain complete/total control of Empire East…and yet New York State remains deaf/dumb/blind that this is a negotiating tactic that can and does get repeated results.

            Any which way you slice it there’s no progress to be had without first talking mutual interests. Which are definitely there. Hell, if Indian Railways can find a way to make a U.S.-based Class I rich on a consulting gig half a world away from where they operate just because it happens to open up buff supply-chain coattails to everyone they deal with in domestic biz, then MassDOT certainly didn’t pull off any Jedi mind trick act with its slate of moves. The common negotiating ground is a lot fatter than “Passengers are from Venus, freights are from Mars” Not Invented Here attitudes. At least with one ogre of a Class I (if not necessarily all the others) you can have a productive conversation by talking to their stomach first. They have lots of leverage they’re willing to immediately barter for your pax expansion and coexistence dreams if you’ll just spare them one small bit of finger-lifting from having to pay for upgrades to some legacy yard or physical plant. Not enough states have even bothered to ask the right set of questions to test the bounds of what’s possible.

          • F-Line to Dudley

            ad12800…there’s no “relative isolation”.

            The whole freaking point of Unigauge was to de-isolate the country’s rail network from its own damn self, which is why it’s now paying real dividends for private biz while it’s at >90% completion. The freight bucks being spread around the world into GE’s pockets is one obvious leading indicator we can count here on this continent, as Wabtec kept that business within the U.S. on the strength of this Indian Railways megadeal.

            Local effects go incalculably deeper than that. Both pax and freight service is now able to interline across former breaks-in-gauge that used to economically isolate the other-gauge regions from the Indian Gauge regions. And it’s an enormous quality-of-life improvement solely in the amount of Unigauge-spurred rolling stock procurement that’s going on. Metre Gauge and Narrow Gauge territories were the “forgotten lands” in India. Most of the rolling stock there dated to 1940’s-50’s, a lot of the Metre stuff UK-imported from the Irish market and almost as old as Indian independence from Britain itself. The ingrained stereotype of Indian trains being smelly smoke-belchers so cosmically overcrowded that passengers are hanging off every exterior crevice is basically Metre gauge territory in one visual. Rosters were never expanded or renewed, only consolidated through gauge conversions and band-aid rebuilt to keep pace with attrition. While their trains may always be overcrowded, the barbaric imagery of people hanging onto the exterior for dear life is finally becoming a thing of the past now that Unigauged territories can finally purchase in-bulk to lengthen trains, run brawnier power to pull more coaches, run more frequent trains period from having enough equipment, and run trains that have post-WWII livery and emissions standards up to basic human decency. That all stems from from network commonality finally opening up world market-rate procurement purchase power at proper scale, and the great changeover now reaching all of the forgotten lands. Generally speaking the incumbent Indian Gauge territories have been well-stocked the last few decades with modern train classes that, while maybe not as across-board spiffy as their Euro standard-bearers, definitely slay that barbaric stereotype of “forgotten lands” Indian trains.

            Right now there’s only 2910 km (4.62%) of Metre Gauge and 1571 km (2.43%) of Narrow Gauge left. 1021 km (~22%) of that remaining total has been set aside to stay as-is for “heritage railways” (tourist lines, “foamer specials”, etc…same thing it means in the U.S.). 277 km (~6%) is being abandoned…in most cases because of route redundancy with close-duplicating Indian Gauge routes. And the remaining 72% is budgeted for conversion. In 2020 to-date about 200 km of new Unigauge lines have opened, so the pace is still torrid even in the “remainder of remainders” last 8%.

            The other thing to keep in mind here is that all neighboring countries on the subcontinent–Pakistan, Bangladesh, Sri Lanka, Nepal–likewise use Indian Gauge, and Bhutan (which has no current operating railways) has a treaty with India to construct a new border-crosser line. Bangladesh is undergoing its own Unigauge-like megaproject to unify its half- Indian, half- Metre network; the other countries are all unified. While political relations with Pakistan and Bangladesh are a tad frosty for the prospect of international pax trains (though freights do cross daily), as a unified network it’s got similar economic heft to Russian gauge territory. And relatively few breaks in gauge with other countries to deal with when fully built-out, as connections to China (Standard Gauge) and the ex-Soviet Bloc are sharply limited in number by the mountains, the freight break-in-gauge in Afghanistan (Russian Gauge) hasn’t yet been dealt with because the Pakistan-Afghanistan rail link is still in planning, and there is no potential for Southeast Asia rail link whatsoever because Myanmar’s completely isolated Narrow Gauge network stands in the way and doesn’t go near any border. So long-distance passenger connection–at least to the friendlier neighbors–will be less complicated than they are with Central and Southern Europe’s gauge breaks, and Unigauge is most definitely India’s bulwark for regional economic supremacy. Their own network’s fragmentation was arguably a weakness they had to shore up to keep the Chinese and Russians from becoming an outsized economic influence on their neighbors. They (and Bangladesh to some degree) are spending themselves on Unigauge like their lives depend upon it…because their lives pretty much do depend on it for how it controls regional balance of power. Almost like a Monroe Doctrine laid out in 1676 mm parallel lengths of steel to the attention of China and Russia.

          • adirondacker12800

            How much does the Flushing Line interact with the national system? There are baraque moves occassionally to get stuff from the Linden Yards to the rest of the system but they don’t interact to a great extent. The subway or the el in Mumbai is going to be like that.

        • Eric

          “India has a ton of legacy rail”
          It might have “a ton” relative to some other countries, but that “ton” is only a small fraction of what’s needed to provide transit to Indian cities.

          “Mumbai’s first metro line already crosses the Suburban Railway without a transfer.”
          Incorrect. There are transfers to both lines, at Andheri and Ghatkopar. At Andheri, the elevated walkway for the transfer even appears on OpenStreetMap.

          “so little learning from Japan’s metro-regional rail compatibility.”
          Perhaps they are learning from the Chūō Shinkansen that if one line is expected to be entirely full, there is little value in making the next line with the same technology, as there won’t be significant through-running anyway?

    • Alon Levy

      This is a really hard question. It really depends on the city, since many do have passable mainline rail (e.g. Karachi, Lahore, Bangkok, and especially Jakarta) that can be modernized (and to a large extent is, in Jakarta).

      But in Africa south of the Sahara and north of South Africa, the best you can do is widen the mainline rail right-of-way to allow for separate local and intercity tracks, and treat the local track as one more urban rail line. This isn’t just about the lack of legacy rail – you can always build something like the Tsukuba Express, imitating legacy rail but not actually legacy. It’s also about the lack of interesting destinations to connect RegionalBahn to. There aren’t significant town centers at commute range that you can string RegionalBahn to, and there’s just a vast size gap between big cities and small ones that the everywhere-to-everywhere model doesn’t really work. You can build to a small town for development’s sake, but population growth is so high that you can build up neighborhoods from scratch with a metro too.

      One possible exception: in Nigeria, there are enough cities that the state should be thinking about an intercity rail network. Moreover, there’s strong potential for lines going east and west of Lagos, whereas the current legacy mainline only goes north. So it should bundle metro planning with intercity rail planning and try to go for 6-track mainline, 2 tracks each of local, express, and intercity traffic. (It may sound excessive, but Lagos has a metro area of 21 million with just about no good rights-of-way for later.)

      • adirondacker12800

        NIgerians are never going to be rich enough to build subway? From a quick glance at satellite images and a peek at street-view they managed to blast modern limited access highways through.

        • Eric

          I suppose that in Nigeria, the buildings that need to be destroyed to create a ROW are not very valuable.

          • Alon Levy

            They are kind of valuable in Lagos Island and Lekki. But it’s fine, because there are very wide roads for els; where Lagos does wrong is a) planning on two-track els rather than four- and six-track ones, and b) skirting the CBD with a waterfront el instead of building short tunnels in Lagos Island to serve it better.

          • Eric

            BTW, are there examples anywhere in the world of 4+ track elevated railways? If yes, are there examples built in already populated areas?

          • Tonami Playman

            In addition to the Ringbahn example in Berlin, London and Tokyo has multiple examples.

            Most of the Corridor between Tamachi and Tokyo Station is just one big 10 track elevated railway carrying four JR East commuter lines and the Tokaido Shinkansen. Another example is on the Tohoku Shinkansen corridor. Here’s the 4 track section that carries the Tohoku Shinkansen and the Shonan Shinjuku line. They remain parallel before the 2 tracks of the Tohoku Shinkansen rises over the Shonan Shinjuku for a short double deck section to squeeze in between the area between the Saitama Super Arena before approaching Omiya station. view it in 3D mode for more context.

            Tohoku Shinkansen corridor before Omiya Station

            Here’s a section of the same corridor, It’s actually a 6 track elevated as it carries the 4 tracks of the Tohoku Shinkansen and Joetsu Shinkansen in the middle and Saitama New Shuttle rubber tired people mover on the outer edges of the 4 track section.

            Tohoku, Joetsu Shinkansen and New Shuttle people mover

            Here’s a google maps location of the undulating Berlin Stadtbahn heading towards Alexanderplatz

            And finally the 10 track elevated section heading towards London Bridge station.

          • Tonami Playman

            Correction, I meant Saikyo line instead of Shonan Shinjiku line.

          • Eric

            Also re Nigeria, I’m not sure if they reliably and fairly compensate the residents and/or landowners (not the same, I imagine there are huge numbers of squatters in the Lagos area).

          • Alon Levy

            Re Nigeria: that’s fine, the required demolitions for the trunk lines are minimal. The squatters don’t live in the way of main roads. (P.S. this is true of Nairobi as well – there’s a railroad going through Kibera making a few stops today.)

            Re 4-track els in general: Berlin Stadtbahn! You can even see the traces of old development around Alexanderplatz, where the railroad undulates to avoid what was tightly packed then and hosts wide streets now.

          • Tonami Playman

            Squating is not that big of an issue at this point of Nigeria’s development. It might cause some initial delays, but the government can always blast through sentiment as can be seen with large number of 36m – 45m ROWs being blasted through houses for highway and stack interchange construction across several cities Uyo, Abeokuta, Lagos, Port-Harcourt, Aba just to name a few.

            The main issue is not planning for multitrack rail ROWs. In Lagos for example the current 30m rail corridor was initially planned to hold 5 tracks, 2 for metro, 2 for Intercity and 1 for freight, but they seem to have given up on that plan and instead gone with building just 2 tracks in the middle of the ROW to run intercity trains, commuter rail and freight. I can only imagine the conflict once the trains start running.

            At least the greenfield corridor in Abuja has 3 dedicated tracks, 2 for commuter rail, and 1 for intercity with enough room to expand the tracks. Abuja also has reserved ROW in it’s masterplan for it’s Urban rail network since it was built from scratch. Mostly 75m wide and 90m in some sections as can be seen here. though the Autocentric urban layout has far more reserved ROW for highways as can be seen here

          • adirondacker12800

            The Chicago L has four track sections. But then Chicago, instead of making it 6 tracks built 2 tracks some place else to shift demand to someplace else. I can see where 3 two track lines in different places might be more effective than one 6 track line but one 6 track line keeps it simple for foamers half a world away to think about. And it saves on crayons!

          • Nilo

            There is no place where the Chicago El built a two tracks instead of adding two. The Northside Mainline has four tracks only from Armitage to Howard, and south of that reverse branches with two to the Loop and two to the State Street subway.

          • adirondacker12800

            He didn’t ask if there is someplace that has four tracks that was not a desperate attempt by people to increase capacity quickly. He asked if there are four track sections and there are. Metra Electric a few blocks away from an L line might be more interesting to look at. Or the history of Sixth Avenue in Manhattan, that was rather complicated. It has six underground tracks. But he was asking about elevateds. Sixth Avenue started out with an El, they added two underground tracks, then they added two more and some four track sections, tore down the El and eventually built the missing link between the four track sections. But that is underground these days. But he was asking about elevateds.

          • adirondacker12800

            In addition to the Ringbahn example in Berlin, London and Tokyo has multiple examples.

            He does have a problem determining scale. Wikipedia says Lagos metro area is 21 million people and growing fast. In those terms one six track line isn’t going to be enough. And two, two tracks lines to start is probably a better idea than one six track line. People in Nigeria can talk about where to put what. Since they have actually seen the modern elevated highway I saw less than minute after typing “Lagos” into the search box on Google Maps, they probably have a better idea than a crayon scrawl. Which I looked at because I hope Nigerians are someday rich enough to build subways and probably have existing routes that could be used.

        • Nio

          Adirondacker don’t confuse the issue you said

          >> But then Chicago, instead of making it 6 tracks built 2 tracks some place else to shift demand to someplace else.

          This statement is false and doesn’t represent the history of the Northside main line. You’re of course right about one thing the Illinois Central did build six tracks on the ROW of what is now the Metra Electric. Four electrified ones for local service, and two non-electrified ones for long distance and freight. Though I wouldn’t call it an El, since it’s really built on a berm not elevated over an alley or a street.

          • adirondacker12800

            Building two subways under the Loop and calling it the L is two tracks someplace else twice.

      • Herbert

        If Managua ever decides to build rail, they should tear down those idiotic highways I stilts to build it…

        • Eric

          What highways on stilts? I looked on Google+OSM and see only a handful of overpasses and everything else at grade level.

          Anyway Managua is a small city (1.4M) and Nicaragua is below replacement fertility and poor, so it will be a long time before they feel the need for grade separated rail…

          • Herbert

            Greater Managua, especially if you include Masaya has a larger share of the population of the entire country than most capitals of countries that aren’t city States. And the migratory pressure on Managua shows no sign of abating any time soon…

            As for sub replacement fertility, you seem to have different data than I have… Yes, Nicaragua went down from a TFR above 5 to one a bit over 2 since the days of Somoza sr. But I thought it was still above 2.1

      • MucklaSSR (@mucklasuntermf1)

        The Train Express Regional in Dakar most definitely can be extended towards Thiès (65 km from Dakar with 320,000 inhabitants), onto Djourbel (120 km from Dakar and 60 km from Thies with 130,000 inhabitants) and finally Touba (165 km from Dakar. 45 km from Djourbel and 880,000 inhabitants). This are Regionalbahn/express distances in my book.

        Nairobi has Nakuru/Naivasha to the west and Ruiru/Thika to the north. Kampala has Mukono/Jinja/Igana on its legacy rail corridor to the east. Addis Abeba has a string of rabidly growing string of cities (Debre Zeyit, Modjo, Adama) to its south all along the legacy rail corridor and in commuting distances.

        • MucklaSSR (@mucklasuntermf1)

          BTW Charles Hinga Mwaura, principal secretary at the State Department for Housing and Urban Development, would love to read ur take on his S-Bahn proposal for Nairobi. He is really open minded and is actively looking for solutions Kenya/Nairobi can emulate.

        • Tonami Playman

          Abidjan with a metro population of 4.7million is also building a 38km TER with 20 stations connecting the Northern suburb of Anya a through the Plateau region where the CBD is located to the Airport in the south. That too can be extended further east 26km to Grand Bassam and further to Bonua which is roughly 60km from the Plateau district.
          Up north it can be extended to Agboville. The TER alignment follows the legacy narrow gauge cooridor. There will be 3 tracks. 2 standard gauge and fully grade separated for the TER and 1 narrow gauge with grade crossings to keep the freight connection to the port. With Cote d’Ivoire already being one of the most urbanized country in Africa with Abidjan accounting for 20% of the country’s 25million population, I expect that to continue with more people living within the metro area and high growth.

          • MucklaSSR (@mucklasuntermf1)

            Unfortunately Abidjan’s TER ROW is only decently located betwenn Anyama Centre and Abobo Banco between Gare International and Plateau Lagune there is probably no cost efficient alternative but walking distances to the stations are okish but beginning from Treichville it’s totally off going through rail yards and industrial areas with little demand. An alignment following Boulevard Valéry Giscard d’Estaing would give decent accessibility to Treichville and Marcory as well as half of Koumassi that’s 750,000 people instate of -at best- 100,000 with the current alignment.

          • Tonami Playman

            Yeah you’re right about the poor alignment. It follows the existing freight route which skirts the left edge of the CBD and then runs through an industrial zone in the south bypassing the populated areas. Making that 7.5km detour to Boulevard Valéry Giscard d’Estaing and then rejoining the current alignment at Boulevard de Marseilles would be far more useful to the serving the population.

            The current cost of $45million/km is already very expensive for developing on an existing right of way, I’d say adding an additional $337million to the $1billion project cost for a 7.5km elevated detour to increase the line’s utility is worth the extra cost. But I have a suspicion the planners are more interested in connecting to the Airport by the path of least resistance without weighing the benefits of serving the population along the way.

          • MucklaSSR (@mucklasuntermf1)

            The current plan is actually to bypass the airport to keep the possibility to extend it further east where rapid growth is happening. There will be a station called Aérocité but it’s located over 1 km south of the terminal.

    • Henry

      Seoul is building the GTX from scratch since it has some legacy rail, but not a wild amount like Europe or Japan.

  3. SB

    “What do Stockholm, Zurich, Prague, Munich, and Budapest even have in common, other than being European? They’re not all national capitals or even all national primate cities”
    Four are biggest city in the respective country and three are the capital.
    So Munich seems to be the outlier.

    “Munich-size cities like Birmingham, Manchester, and Lyon have far weaker transit systems”
    Birmingham and Manchester doesn’t have subways so expecting similar transit use as cities with subways is bit puzzling.
    Lyon does seems to underperform considering the amount of infrastructure it has.

    While keeping cost down is very important, you can’t expect good transit service without some investment.
    For example, I don’t think Birmingham received comparable level of investment in transit compared to Munich.
    So it is not surprising why there is more transit usage in Munich.

    • Nilo

      Lyon if you look at trips per person seems pretty good for a city its size. 20% mode share, and 40% more trips per person per annum than NYC with 141 trips per person

    • Alon Levy

      Munich built a subway in the 1970s as well as an S-Bahn tunnel, whereas Birmingham had long had infrastructure for an S-Bahn but didn’t do anything with it; even London wouldn’t do anything with the Widened Lines until the 1980s and Thameslink Programme.

      • Herbert

        There isn’t a Birmingham based party that runs the country half of the time and Birmingham never hosted Olympic Games…

        • Alon Levy

          The 19th-century Liberals were strongly identified with Birmingham and Labour between 1945 and 2019 with Northern towns and cities…

          • Herbert

            Yeah, but it wasn’t the “Birmingham Labour Party” which by right had to have several seats at the cabinet table whenever Labour was in power.

            Munich has that. Munich has the “CSU of Bavaria” which, whenever the center-right is in power, has several seats at the cabinet table in excess of their share of the population…

          • fjod

            From 1885 to 1929, all of Birmingham only elected one MP who wasn’t conservative. From then until the 1980s, it was split relatively evenly between Labour and Conservative representation, both nationally and locally. It has since swung Labour but it remains relatively weak for the party compared to cities like Manchester and Liverpool; it spent half this century governed by Conservative/Lib Dem councils. And it definitely isn’t northern, and thus doesn’t form part of the Labour ‘heartland’ myth (though partly this is because Labour have historically done unexpectedly poorly there, as I just mentioned).

            That said, Birmingham has had an S-Bahn in all but name since 1978, in the form of the Cross-City Line.

    • Pennengineer

      One thing being overlooked here is that Munich is an economic powerhouse and one of the wealthiest cities in Germany by many measures, whereas Birmingham and Manchester are…not. Germany is the exception to the European rule of concentrating wealth and power in the national capital; Berlin has a largely administrative and — to some extent — cultural role. It’s not by accident that the quote by the former mayor is so well known for describing Berlin: “poor, but sexy”.

      • fjod

        Barcelona, Milan, and Rotterdam all have transit that’s as good as (in Milan’s case, better than) that of their nation’s capital. And Rotterdam, in particular, is significantly poorer than Amsterdam.

      • Herbert

        Germany may not have the tendency, by Bavaria certainly does. The CSU can’t stop spending on Munich even if they keep electing “red” mayors…

    • Herbert

      Thanks to being governed by Munich, Nuremberg has a subway. Does Nuremberg have better ridership than other German cities of its size?

  4. adirondacker12800

    Covid-19 crisis appears to be one of these rare situations, Korea being the best

    Your sample size appears to be very small to me. And they did a whole bunch of stuff public health workers have been doing since we began to understand germ theory and that it wasn’t bad air in most cases.

  5. Martin

    Minor point: Taiwan can hardly be called monolingual. They announce the subway stops in four languages. It is one of the more linguistically diverse places I know.

  6. Lee Ratner

    My brother’s girlfriend is from Singapore and through her meets many other people from Singapore. Being into literature, he often asked them what novels the read as part of their high school curriculum because he wanted to see what was canonical in another country. The response was apparently none and that a lot of education in Singapore can be described as math and science followed by more math and science. Some internet research leads me to believe this is the case through out a lot of Asia. There is a lot of emphasis on math and science along with English and not a lot of focus on history or literature.

  7. Herbert

    Actually Munich is a capital and all the other cities on the list are either the primary city of their state (Zurich) or its capital…

    You may argue that Bavaria is not a sovereign state, but often enough it acts like one and the CSU tail wags with the CDU dog. Just look at Söder’s grandstanding regarding CoViD19…

    Anyway, a problem Nuremberg faces is that it is administered by Bavaria and thus built a subway which has awkward branching (interlining to 100 second headways on the shared trunk of U2 & U3)…

    To add insult to injury at this point in time when virtually all subway construction that is even on the same planet as KNF>1 has either already been built or is planned to be built, Nuremberg elected a CSU mayor, i.e. A mayor of the party that never saw a harebrained subway plan they didn’t love. And who are the main culprit that the Nuremberg tram network hasn’t grown noticeably in the last two decades…

  8. Tonami Playman

    More likely, the issue is that Japanese signaling underperforms European systems and is the process of catching up

    I remember back in 2014 JR East planning to install Thales CBTC on the Joban line to compare to their proprietary ATACS signalling installed on the Senseki and Saikyo lines, but in 2017 decided not to go further due to incompatibility with existing ATOS equipment much higher cost of re-calibrating existing signals meanwhile ATACS piggybacks on them. JR East claims ATACS is more useful to them since it can handle more complex network operations including grade crossings while CBTC has only be used successfully on fully self contained lines. I don’t know how much of this is NIH syndrome or if for their operations, their own technology is a better solution.

    Tokyo Metro Marunouchi line is getting CBTC by 2022 with it’s recent fleet replacement with the 2000 series EMUs and since it has no through running, should be easier to implement. I expect Ginza line to follow suit. Maruouchi line is already the highes frequency line with 32tph, so CBTC could possibly take it to 36 or 40tph. I don’t expect the other Subway lines that have heavy interlining with though running services to get close to 30tph any time soon.

    • Andrew in Ezo

      Some in Japan have surmised the relative lack of domestic implementation of CBTC/driverless systems (other than on lower pax volume AGT lines) to operator sensitivity to public and media criticism of railways in cases of mishaps/accidents, not to some lack of expertise. (as an example Nippon Signal’s SPARCS CBTC system is used on the Delhi Metro Line 8 as well as the Beijing Metro line 15). Just last year there was a grade crossing accident on the busy Keikyu line that resulted in a truck drivers death (who was at fault), and the criticism was swift from the media for the railway operator, when in fact the driver of the train by all accounts did everything in his power to stop given the difficult circumstances (curve before grade crossing, limited express service, only 570m to stop from a 120km/h running speed). There was in fact considerable pushback to the media reportage from railfans and more thoughtful users of social media. But in general the public is extremely demanding wrt to train service reliability (stories of conductors apologizing for 2 minute delays, or even early arrivals are true), and public distrust/unease of driverless operation in cases of accidents due to malfunctions, passenger medical emergencies, and possible carriage evacuation is high.

      • yuuka

        You can have CBTC with GoA2.

        Tokyo Metro and Tsukuba Express at least have GoA2 – and in the case of Metro, they have ATO on lines with through running to other railways. I also seem to recall JR East working on automated operations of the Yamanote Line.

        Or is it they get a free pass because of full grade separation? ZPTO is hard but, at the very least, I’d guess Alon wants to eliminate the conductor.

      • yuuka

        Speaking of Keikyu, I forgot this:

        Downright medieval, and all the transit activists in the West want to get rid of it. But, Keikyu is doing that even today. Perhaps that’s also what Alon is referring to.

    • yuuka

      Dwell times may become an issue with the Marunouchi line, especially with the 3 door 18m cars (unfavorable compared to Ginza, 3 doors 16m). They may need to invest in wide door vehicles, or they may find themselves unable to go past 36tph or so.

      As for ATACS, I guess it’s fair to blame NIH. Japan is oddly insular with them=se kind of stuff.

  9. RossB

    >> whereas the Covid-19 crisis specifically still looks like a clean Asia vs. West cleave

    I think most people would consider Australia to be part of the West, given its history. Its response to the virus has been about as good as any country. It took the same approach as the Asian countries that have been lauded: test extensively and react accordingly.

    • Alon Levy

      For at least a week, maybe two, I’ve kept saying I’m not making any prediction for Canada, Australia, and New Zealand, all of which take events in East Asia far more seriously than the US and Europe do, but at the same time have white majorities that look down on Asians.

      • RossB

        Canada seems to be struggling. New Zealand has just recently been infected, and despite the additional lag time, the disease (apparently) has spread pretty quickly. In comparison, you can make the case that Australia has handled the disease as well, or better than Asian countries. The began testing early, and found their first case just a few days after it was found in South Korea (not by a sick patient, but with routine screening). At one point, Australia was testing more than anyone per capita. I think when this is all over, the report (from various countries) will all point to the same thing. The key is to test, test, test. The U. S., in comparison, discovered the disease as part of a different study, looking at the flu (https://www.nytimes.com/2020/03/10/us/coronavirus-testing-delays.html) and still doesn’t test nearly enough (despite what you might hear from the president).

        • Sarapen

          Canada’s problem seems to be more the reluctance to close the border to Americans and require returnees from the US to self-isolate. This wasn’t done until a deal was hammered out with the US, clearly to avoid angering its mercurial president, which allowed infection from the US to enter.

          Well, that and the neoliberal austerity which cut health care resources going back to the 90s, but that’s basically the same as most Western countries.

          • Alon Levy

            In Israel the solution to the “it gives Donald Trump sads” problem was to close the borders regardless of country of arrival.

          • Sarapen

            Wasn’t that at Pence’s request, though? He knew that beloved Bibi needed to restrict US travellers but it would make the US look bad, so he suggested “go global” and that Israel should restrict everyone instead. So it was more like it was Trump’s VP’s solution.

          • adirondacker12800

            I don’t know if the Vice President is a Premillennialist, Postemillennialist or not a Millennialist. I almost always briefly view his actions using from that point of view(s). Many of them are motivated by their interpretation of the pivotal role Israel plays in their eschatology. And the Vice President is really good at inserting a dog whistle to them. Really good.

        • RossB

          By the way, for the U. S., I think South Korea offers a better model than Australia. In South Korea, the virus spread quickly because a religious sect ignored the recommendations. Americans deal with this sort of thinking every day. In Australia they didn’t have to deal with that problem. Yet if I was a politician (like Joe Biden), I would take the opposite attack. I would de-emphasize South Korea, and lead with Australia. For example “Trump diddled away opportunities to contain this virus — even claiming its severity was due to some kind of Democratic hoax — while countries like Australia, South Korea and Taiwan quickly suppressed the virus. They first tested for the virus in Australia back in January, finding and quarantining people a couple days later. As a result, Australia has only a couple dozen fatalities, while the U. S. has thousands. …”

          This is because of the general level of racism in the U. S., where lots of people can’t relate to South Korea, but can relate to Australia. To be fair, from a land use perspective, the U. S. is probably more like Australia (lots of open land, lots of huge suburbs, etc.)

          • R. W. Rynerson

            Ross, I think that you are over-generalizing. Our governor in Colorado cites the Korean experiences almost every day as exemplary. And his references are relevant, not just because our commuter rail cars come from Hyundai Rotem.

          • RossB

            Well good for him, but Colorado is no longer a swing state. It is like Washington State (which used to be a swing state). It is quite reasonable for a politician to mention South Korea in either place as it would be in California.

            I’m talking about areas of the country that voted for Trump, like Florida, North Carolina, Georgia, Ohio, Pennsylvania, Michigan and Wisconsin. Those are states that Trump won, but Biden could win.

          • michaelrjames

            While it’s true that Australia was second only to Korea in testing per capita, you seem to be suggesting that they were doing contact tracing. Unless the politicians and journalists have simply failed to understand (entirely possible) and the medical-pathology people have kept it secret, I am not convinced. In fact there have been plenty of stories and complaints by many people who suspect they have the disease but have been refused testing because the criteria were quite narrow (overt disease characteristics that tried to eliminate common colds etc; age, other susceptibility factors). Only the past few days has there been overt discussion of wider testing to contact trace. Politicians have tended to be narrowly focussed on lockdown, and the police have typically been over-reacting. Someone would have to explain to me why sitting solo or even two, on a park bench is absolutely interdite. The police used a drone to harass a solo walker with his dog in a national park as if he was a refugee boat person.

            Speaking of boat people. Meanwhile officialdom let 2,700 boat people (Ruby Princess cruise ship) disembark into Australia without so much as a warning or recommendation for them to self-isolate. Some 600 have tested positive so far, and they account, and will account, for a significant fraction of all cases (and maybe eventually deaths given the age skew).

            Honestly, it is not clear why Australia has such low death rates. Earlier I suggested the hot dry conditions of this time of year could have been part of it; and now they think severity of disease (and thus death in the aged) is directly related to the infection load, and that could be reduced by the hot-dry conditions. Of course our hospitals and ICUs haven’t been taken to any kind of stress point … yet.
            I think Alon is right that it is still too early to give Australia a pass. I’m not sure the real impact of those Ruby Princess passengers has come out in the data yet (only 2 weeks).

          • Alon Levy

            …Australia is letting in thousands of boat people if they come by luxury cruise, but deporting the ones who come by raft to Manus Island to a concentration camp with typhoid outbreaks? I thought the White Australia policy had been abolished.

          • michaelrjames

            Exactly. There have been endless jokes based around that irony. Plus, the current PM was the Immigration Minister who “stopped the boats*”, and the current guy running the show (Dept Home Affairs, = Homeland security) and revels in his hardman image, Peter Dutton, is the one who came down with covid-19 (suspected on his recent trip to the US where he met with Ivanka T. and Andrew Barr!).

            *He was infamous for refusing to answer questions with the reply “that is an ‘onwater’ matter and I can’t discuss it”. Also, it turns out he didn’t exactly stop them at all, with about 50 boats having been intercepted and turned back illegally on the Conservatives period of government. This nasty government keeps pointing the finger at Labor saying they let in 50,000 boat people, yet in the time of their own government it turns out there have been 50,000 arrivals by air who have claimed asylum.

          • Herbert

            I have heard that Germany started out with comparatively high testing capability and has significantly ramped it up since…

          • michaelrjames

            It is still a long way from an explanation for these vast difference in morbidity.
            There might still be some funny business over some of these numbers. France took a big jump overnight because for the first time they included deaths in nursing homes! In fact Spain has revealed that quite a lot of their deaths are from nursing homes.

            03 April 2020
            Nation Cases Deaths CFR
            World 1,097,909 59,131 5.4%
            Italy 119,827 14,681 12.3%
            France 63,633 6,496 10.2%
            Netherlands 15,723 1,487 9.5%
            Spain 119,199 11,198 9.4%
            UK 38,690 3,605 9.3%
            Belgium 16,770 1,143 6.8%
            Iran 53,183 3,294 6.2%
            Sweden 6,122 351 5.7%
            China 81,639 3,326 4.1%
            Brazil 9,082 359 4.0%
            Switzer 19,702 604 3.1%
            USA 277,355 7,128 2.6%
            Portugal 9,886 246 2.5%
            Turkey 20,921 425 2.0%
            S Korea 10,156 177 1.7%
            Canada 12,531 187 1.5%
            Austria 11,521 168 1.5%
            Germany 91,959 1,277 1.4%
            Norway 5,296 54 1.0%
            Israel 7,428 40 0.5%
            Australia 5,529 29 0.5%

          • adirondacker12800

            <i….hot dry air…

            There is inconclusive evidence that adequate levels of vitamin D are protective for many infectious respiratory infections. Before there were antibiotics people with tuberculosis would go away for “the cure”. Where they got plenty of rest, lots of healthy food, fresh air and …………..sunshine. There was more than some advocacy for brisk cold mountain air. they would bundle the patients up and leave them out ………….in the sun….. where they could lots of cold fresh air.
            People who live in warm dry climates go out …..in the sun…. more. They are usually closer to the equator where there is… more sun… during the winter.

          • michaelrjames

            Not exactly sure of what your point is. Today deficiency of vitamin D is pretty rare. Even Americans eating a fast-food diet get enough, and one doesn’t need much sunlight. Though it is true that there is more UV-B content in sunlight at higher altitudes.
            No magic mountains needed, but lo, maybe you are writing from such a place?

            The Adirondack Cottage Sanitarium, established in Saranac Lake, New York, in 1885, was the first such establishment in North America.

            Hmm, do you think these lockdowns might cause a resurgence of vitD deficiency? Bring back those tanning beds!

          • adirondacker12800

            I didn’t say anything about staying inside for any reason. I do know that one of the main features of the santioria, most of them anyway, was spending a lot of time outdoors in the fresh air and ….sunshine. That perhaps it wasn’t the fresh air that was curing them. Though living in a city that has killer smog probably didn’t help and being in fresh air would have some effect. For whatever ails you. That in air the mountains isn’t very warm. One of the features of the mountains today that the tourist industry still touts. That perhaps it isn’t the warm air, it’s the sunshine. Just perhaps because there are double blind studies that seem to say it does and other studies that say it doesn’t seem to have any effect. And that perhaps it doesn’t have anything to do with there air it’s so polluted the UV can’t get through. Or are you still clinging to miasma theory?
            Are you deliberately obtuse or do you work at it?

          • RossB

            I have no doubt that Australia made some mistakes with their handling of the crisis. I also think there is the very real possibility of a spike in cases, just as there is in Asia (https://www.nytimes.com/2020/03/31/world/asia/coronavirus-china-hong-kong-singapore-south-korea.html). But at this point Australia has handled the crisis much better than most of Europe and North America (see the early part of this interview: https://www.youtube.com/watch?v=SVwIjxYJRko).* Weather may have something to do with it, but the (relatively small and very warm) state of Arizona had twice as many cases as Australia yesterday, despite the first (known and likely) infection occurring much later. Spain has been bit hard, while northern (colder) countries in Europe have not. There may be a seasonal aspect to the virus, but at this point, that isn’t known.

            It would seem strange for a center-left politician in the U. S. to praise the center-right government in Australia. But in this case, it is about competence, and nothing more. American voters — with a very weak party system — tend to vote more based on the people running for office, rather than their ideology. That is why Biden should — at some point — praise Australia, as well as South Korea, as countries that knew what they were doing, while the U. S. diddled around.

            *I hate to link to a video, but I can’t find a transcript for that. The point is, a former advisor for the WHO said that at this point Australia has been “quite successful at controlling this infection, same as [South] Korea…” You can’t find anyone lauding the response in (most of) Europe or North America, let alone saying it is close to, or on par with South Korea.

        • Tonami Playman

          I’d guess because of 3 things, 1. The tiny population. The entire population of New Zealand is 4.9million which will fit into the metro areas of either Sydney(5.2million) or Melbourne(5million). 2. Distance. The distance between the two Largest metro areas Auckland (1.5million) and Wellington (420,000) is about 645km and an extension to Christchurch (385,000) is about 1000km from Auckland and would require an expensive underwater tunnel to cross the Cook Strait. Compare to Taiwan’s 345km to connect Taipei (7million) to Kaohsiung (2.7million) serving Taichung (2.8million) along the way. Also South Korea’s Gyeongbu HSR uses 346km to connect Seoul (9.7million) to Busan (3.5million) while serving Daegu (2.5million) along the way.

          And finally the Difficult hilly terrain. The cost just outweigh any benefits for such a small population. Japan has similar terrain and similar distances, but has metro areas of other worldly size to connect compared to New Zealand. Japans Shinkansen network connects Tokyo (38million) to Osaka (20million) in 515km and reaches Fukuoka(2.5million) in 1069km. A line in New Zealand would cost just as much as a line From Tokyo to Fukuoka while not getting much utility out of the Investment due extremely low utilization. Now Melbourne to Sydney HSR, that should have been under construction right now.

          • Alon Levy

            (Seoul is 26 million. There isn’t an officially-defined metro area so a lot of international orgs use the city proper, but that’s incorrect, Incheon and the entirety of Gyeonggi are suburbs of Seoul and together with the capital have 26.something million people, just as Kanagawa and Chiba and Saitama are suburbs of Tokyo and just as New Taipei and Taoyuan are suburbs of Taipei.)

          • Tonami Playman

            Thanks for the added info. That 9.7million on wikipedia just seemed too small for the sprawling mass of the built up area around Seoul.

          • adirondacker12800

            None of that changes that there are 4.9 million people in New Zealand and a lot of them live in metro Auckland.

          • Tonami Playman

            That might be a result of the International penalty be Estonia and Finland. But the tunnel does have the advantage of connecting to the rest of Europe to increase its utility. A tunnel under the Cook Strait connecting the Northern Island of New Zealand to the very limited population on the Southern island has not extra population to connect to. Investment in electric aircraft will be better use of that capital.

          • adirondacker12800

            Authorizing a formal study could provide solid information that it isn’t worth it…….. You don’t have to travel through Russia to get from South Island to North Island. The rest of Europe is beyond Estonia and there is a lot of ocean south of South Island until you get to Antarica. Which isn’t much of a destination and quite far away.

          • Herbert

            Nowhere in the Baltic is even nearly as dangerous a waterway as Cook Strait. Plus New Zealand is a lot wealthier than Estonia…

            But obviously the intra- North Island lines should be built before the tunnel is built…

            That said, the tunnel might be good for cohesion and whatnot…

          • adirondacker12800

            There are 1.1 million people on South Island, there are these things called airplanes that have adequate capacity to serve them.

          • Herbert

            New Zealand is rapidly growing in terms of population and for various reasons, it may at some point want to change that its larger island has far fewer people in it. If you can entice them onto that island with a nice all-weather tunnel that gets you to the “good island” in no time, that makes things easier…

            Of course the tunnel shouldn’t be the first line. The first line should be Auckland-Wellington, currently a ten hour slog on a good day…

          • Reedman Bassoon

            The Faroe Islands (population 52k, part of Denmark, out in the North Sea halfway between Scotland and Iceland), has two tunnels being built to connect islands.

            … the three-branched sub-sea tunnel measures 11.24 kilometres (6.8 miles) long, including an underwater roundabout. Construction costs are estimated to be around a billion DKK. Drilling commenced on 21 February 2017 and the final blast was made on 7 June 2019. …. making it the longest sub-sea road tunnel in the world at this time, surpassing Tokyo Bay Aqua-Line, but likely to be surpassed immediately by the new Ryfast and Rogfast sub-sea car tunnels in Norway.

            Sandoyartunnilin is an undersea road tunnel under construction in the Faroe Islands. It will connect the main island of Streymoy with Sandoy to the south. The length of the tunnel will be 10.9 kilometres. The estimated cost is 860 million DKK.

  10. RossB

    Thinking about this some more, I wonder if there is value in comparing countries on more of an apples to apples way. For example, Korea has very high density, even with their secondary cities. Yet they lag much of Europe and Japan. That seems like a flaw (or at least it could be — more analysis is called for).

    New York is the only city in the U. S. with significant density. Overall, the U. S. looks like a large version of Australia. But Australia has way more transit use than the U. S. (especially when you pull New York City out of the equation). This suggests something a little different. If you are in New York, then you should definitely look at big cities in Asia and Europe, while you can pretty much ignore what is happening in Australia. But if you are in, say, Austin Texas, a growing (but still pretty low density) city, then maybe you can look at Australia. Or maybe there are even better examples of low density areas with relatively high transit ridership.

    • Herbert

      The “Karlsruhe model” is often cited as an example of high ridership in less dense areas through tram trains…

      • RossB

        Wikipedia has an entry for that, and lists the San Diego Trolley as an example. It carries 120,000 riders a day, which is tiny for a city that has 1.4 million within city limits, and around 3 million in the general area. Like most light rail projects in the U. S., it fails to get many riders, while the buses do the bulk of the work (such as it is). I’m sure there are places where it makes a lot of sense (especially if it is easy to leverage existing rail) but in the U. S. it has often failed, and failed miserably. It is not really a philosophy, or a model, but simply a tool which can be leveraged. The inability to tell the difference is probably the biggest reason why the U. S. continues to build crap.

        • Herbert

          I’m pretty sure if you got the guys running the Karlsruhe trams on a two week fact finding trip to San Diego and talked to them off the record afterwards, they could give you several reasons for the apparent failure in the U.S.

          • RossB

            Yeah, that’s my point. If you took people from low density areas of Germany, the rest of Europe or even Australia, they would all have something useful to say about the U. S. transit system. In some cases they might recommend more Karlsruhe type trams, but that would be only a tiny bit of what they recommend.

  11. Tonami Playman

    What do you think about adopting Singapore’s elevated AGT based LRT systems for feeder services to metro stations? Singapore uses 3 loop lines to connect new suburban housing developments to their nearest metro station. Personally I find them as rather overkill for the purpose. I’m thinking an at grade BRT or Rail based LRT would be a more cost effective solution.

    • Eric

      I suppose it all depends on what price they can build them for, and what increase in ridership they get as a result? I agree it is unlikely the math will work out. Though, if these are built by a semiprivate housing developer for what they think is economically viable reasons, that counts in their favor

    • yuuka

      It really hasn’t worked out, and with the newer project in the west they’ve actually upgraded it to a full medium capacity system with actual trains.

      Doesn’t help that the main benefit of the AGT implementation is just grade separation, which you probably can do with a BRT guideway too. They could probably find a way to upgrade the Sengkang/Punggol system to three-car trains at least, which would be where the carrying capacity of AGT vs BRT can really flex its muscle.

      • Eric

        BRT guideway is worse, it has to be more massive because the buses won’t naturally stay on a guideway. Also a lot of driver labor is needed to drive those buses.

        • yuuka

          Well, you could get by with something like the Cambridge Busway/Adelaide O-Bahn and battery-powered self-driving electric buses.

          It occurs to me that a large portion of the operation and maintenance cost of such AGT is the provision of electrification systems.

          • Herbert

            Why such a needlessly complicated system when rail is cheaper anyway?

    • SB

      U Line in Korea and Yukarigaoka Line in Japan are similar elevated grade separated feeder lines.

    • Andrew in Ezo

      I think it very much depends on the housing density as well the footprint of the development. I’m familiar with three AGT systems in Japan- the Yurikamome in Tokyo, and the Kobe Port Liner and Rokko Liner in Hyogo Prefecture. All three share characteristics of providing discrete access to artificial islands with mixed commercial/industrial/residential developments built on a grid pattern, with housing being primarily high density high rises. The Yurikamome in particular also serves convention centers and shopping/entertainment facilities, with higher traffic on weekends than weekdays The AGT has better capacity that BRT to deal with surge crowds, and has the advantage of grade separation along with ability to negotiate sharp curves and steep grades- giving it a capability as a distributor with stations at frequent intervals and otherwise difficult locations spatially. Of course, as others have stated, the AGT is expensive due mainly to it’s non-standard, bespoke nature with proprietary electronics, etc. BRT or merely frequent regular bus service would be the better choice for spread out, low density developments (mainly single family detached housing), while light rail would be a choice for infill development or as an extension of an existing system, not functioning merely as a feeder to a “heavy rail” station(s).

      • Eric

        To me the closest non-Asian comparison is the Miami Metromover. That too connects high-rise developments to the main metro system in a tropical city. Perhaps the weather in these places is so steamy that few people are willing to walk 10 minutes to the metro, so a high frequency low capacity system is needed to bring them there.

        Now, why make this system an elevated people mover rather than a cheaper ground level bus? One reason is that the bus is seen as a low-class, unreliable mode which only poor people are willing to take. This is probably the situation in Miami, but less likely in Singapore. Another possibility is that it’s a way of signalling the system’s permanence. A bus route can be changed or cancelled at any time, to the detriment of people who have bought flats along its route. But it’s unlikely that an elevated people mover will be shut down the same way.

  12. Pingback: Mixing and Matching from Pedestrian Observations – Hogg1890 – Blog

Leave a Reply to RossB Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.