Wednesday, May 21, 2008


NOTE: We've moved! Visit us at the California High Speed Rail Blog.

I have to confess that the arguments over the CHSRA's ridership projections have seemed somewhat pointless to me. Are there really people out there who think folks won't flock to a high speed train?

Amtrak California routes like the Capitol Corridor are seeing record ridership - ridership between Berkeley and San José is up a whopping 75% on that route, and 17% from Emeryville to Sacramento. This is being repeated across the country - 20% in North Carolina, 19% between St. Louis and Chicago, 8% on the Amtrak Cascades.

High gas prices are driving this - and as more and more Americans are coming to realize, high gas prices are a way of life. We might see a retreat under $4 later this year, but it's unlikely we'll see under $3 for any sustained period of time - and we might instead be headed MUCH higher. Arjun Murti, who tracks oil markets for Goldman Sachs, predicts $200/bbl oil this year - and his past predictions have proved accurate:

From here, there are only two options, and both of them suggest a need for HSR. Either the price of oil continues to rise - 8 years ago Californians were fretting about $2 gas; where will we be in another 8 years? - or the price will collapse. But the only way the price will collapse is if there is demand destruction - meaning people will be driving less. The long-term rise in prices is being driven by peak oil, which implies the possibility of supply shortages sooner or later. Cheap gas isn't useful if you can't get enough of it to meet your needs.

Some argue that we'll just develop alternative fuels. But it's not that simple. Ethanol has turned out to be a disastrous policy, helping contribute to inflated food prices. Electric vehicles cannot be used at the same rate as gas-powered vehicles; there just isn't enough generating capacity. Hydrogen fuel cells haven't proved to be an affordable mass solution either. Alternative fuels may help some folks continue driving sustainably, but they'll never be able to allow Californians to drive as much as they are today. For alternatives to have any positive impact, we need to reduce the usage of cars for commuting and intercity travel purposes and use the alternative fueled cars elsewhere.

All of the above quite strongly indicates HSR ridership will be quite high when the system opens in the 2010s. I'm no expert on the exact numbers, and I don't actually think they matter. The specific numbers only matter to those who want to argue HSR might not cover its costs. Of course, the notion that HSR absolutely MUST cover its costs at the farebox is another case of trains being held to standards that freeways and airlines aren’t. Freeways don’t recover their own costs, we subsidize the hell out of them. We gave the US airline industry a $15 billion bailout in 2001 when their "ridership" plummeted and nobody batted an eye - and they likely won’t when we do another bailout in the next 24 months.

Ultimately, this isn't really an argument about ridership at all. It's an argument about whether we will admit that things have changed. That the 20th century is over, and its automobile-centric attitudes are as well. It's an argument about whether we'll agree with Paul Krugman that Americans need to start driving less, or whether we'll foolishly and expensively delude ourselves that the 1950s can continue forever. If we are honest about our energy, transportation, environmental, economic, and fiscal needs, the case for HSR will become obvious.


Anonymous said...

In your mind nothing seems to matter but that HSR be built. You don’t care how much it will cost and now you don’t care how many people will use it. You don’t care if the project ends up costing the taxpayers billions of dollars in operating subsidies rather than producing the grand surpluses CHSRA is predicting and banking upon.

You have exclaimed that the project is needed to solve air quality problems in the central valley. What non-sense. The central valley is looking for massive job and residential housing creation as the result of this project. The pollution saved by having riders use the train rather than driving to work or going on vacation will be insignificant in comparison with the increase in pollution caused by the population increase that will occur with the normal pollution that is produced from everyday living.

Gas prices are high today and going higher. On this short term basis, there are certainly more people using public transportation than driving independently. But we have seen high gas prices and shortages influence driving habits before. In the early 70’s, during the gas shortages, consumers drove less and started a trend towards buying smaller, more gas efficient automobiles. It didn’t last very long, however. In a short time, the gas guzzling SUVs became the norm rather than buying a compact, and old driving habits returned.

For you to use the subsidies to the airlines after 9/11/2001 as a meter is shameful. When our government cannot protect the country from terrorism, an industry that was on the verge on collapsing needed and deserved this help.

Anonymous said...

Robert, you are spot on again. Peak oil took a lot of flak because some people jumped the gun and predicted peak oil much earlier than today. However the evidence abounds from all over the world that production is peaking or declining, from Saudi Arabia to Russia to Mexico to Ecuador.

Just today, the Ecuadorian oil minister resigned because he couldn't halt a decline in production. Ecuador may not be a large producer, but it is telling of a larger trend towards declining production.

Even if we were being too optimistic and oil supply is not declining as sharply as thought, all you need is for demand growth to outstrip supply to send prices to the moon. And demand shows no signs of letting up on a world scale, though with prices high enough, demand will drop (The question is: how high?)

On a related note:
American now charges for the FIRST checked bag, $15 amid capacity cuts:

Tony D. said...

Drove by a VTA light-rail parking lot today and it was PACKED with cars! Sister-in-law states that she'll put away the "Beemer" to take the trolley from South San Jose to downtown. Anyhow, anon 9:38 states that "the Central Valley is looking for massive job and residential housing creation as a result of this project." Isn't "massive" job creation a good thing, especially in light of our current economic state? "Residential housing" in the form of transit oriented development is also a good thing. The jobs and smart development that HSR creates will be just plain awesome for our state and the economy. How anyone could argue otherwise is beyond me.

Robert Cruickshank said...

anon should ask the logical question, "why does Robert want HSR to be built so badly?" Obviously anon hasn't been reading this site very long, otherwise he would understand that if we don't build HSR this state faces much higher costs thanks to soaring fuel prices, which will in turn have a devastating economic impact.

The cost and the ridership must be analyzed in context. That means it must be compared to the alternatives. The cost of not building it is FAR higher than the cost of construction. The ridership will materialize and it will be substantial - haggling over whether it'll be 90 million or 120 million is pointless. Since ALL forms of transportation in America are subsidized, what's so bad about HSR being subsidized? And since HSR systems around the world generate profit, what's to assume it wouldn't do so here?

The pollution issue in the Valley is NOT nonsense - it's a very serious health problem, with high asthma rates and the like. Further, HSR spurs dense growth - the kind that produces much FEWER pollution.

If you think the current gas price spike is temporary, you've got another thing coming. As to 9/11, our government has failed to protect us from high gas prices - so why is it any different?

davisgrad, those are excellent links. I had the airlines in my mind all day as I thought about this post but ultimately felt I'd gone on long enough. The airline industry is slowly but steadily collapsing before our eyes. Surely this should convince more folks of the need to invest in passenger rail, and HSR in particular.

Rafael said...

CHSRA and MTC commissioned Cambride Systematics to produce a ridership forecast based on scientific models.

They are forecasting 896 million total annual passenger trips in the state by 2030, up from 264 in the Authority's business plan of 2000. The discrepancy is due mostly to a change in scope and improved model calibration.

Of these 896 million trips, some 57 million are forecast to be on HSR. This is up from 37 million in the business plan. This was already enough to cover operating revenue. The increase in the forecast is due to an expansion in the HSR network and the observed rise in the cost of operating a car.

The trunk line from SF to Anaheim is designed to handle up to 86 trains each way every day, with trains up to 1300 feet long providing up to 1600 seats each. That adds up to an ultimate capacity of 100 million annually. The spurs to Sacramento and San Diego, respectively, add to that ultimate capacity.

Of course, only a fraction of this will actually be used early on because it takes quite a long time to build ridership to numbers that require such monster trainsets. SNCF in France only recently started running TGV Duplex trains with 1000 seats on the core Paris-Lyon route. Eurostar must share the Channel Tunnel with slower auto trains, so they run relatively few but extremely long trainsets. Most HSR trainsets around the world currently offer 350-700 seats.

However, train frequency on California's system should be high from the outset, e.g. one train every 20-60 minutes each way, depending on time of day.

Rafael said...

@ Robert Cruickshank -

you've presented some arguments here that are only indirectly related to ridership and deserve closer scrutiny:

1) "the only way the price [of oil] will collapse is if there is demand destruction"

That strikes me as overly simplistic. The price would also collapse if significant additional production capacity were to come online, eliminating the hefty risk premium related to the historically low level of global excess production capacity. There is actually plenty of oil left in existing production reservoirs, but most of these are either in war zones or controlled by foreign governments that lack the technical expertise to maximize production. At some point, those governments will negotiate deals to let western experts back in because they need the extra revenue - and emerging economies like China and India need the oil. However, it takes 5-10 years for capacity expansion to translate to higher production.

Another big factor in the recent run-up of the price of oil is the falling US dollar, a direct result of the subprime mortgage asset bubble.

Fortunately, concerns about energy security and global warming are now beginning to translate into policies to promote fuel conservation. Europe and Japan are ahead of the US, which still covers 45% of its voracious demand for oil from domestic sources.

2) "Ethanol has turned out to be a disastrous policy, helping contribute to inflated food prices."

This is arguably true in the US but much less so in e.g. Brazil. Moreover, by the time HSR goes into service, we will almost certainly have industrial-scale quantities of synthetic fuels based on feedstocks other than food and feed. However, these designer fuels will carry a hefty price tag - especially those that rely on solid biomass.

Similarly, advances in diesel fuel, engines and exhaust gas aftertreatment will massively reduce criteria emissions from US trucks, diesel locomotives and even ships over the next decade.

Of course, all-electric trains will be cleaner still, especially if the electricity comes from renewable sources.

3) "Electric vehicles cannot be used at the same rate as gas-powered vehicles; there just isn't enough generating capacity."

This is true only if you insist on a rapid recharge in the middle of a hot summer day. As long as BEVs stick with trickle-charging during off-peak hours, even California could support millions of them without any new power plants or grid upgrades.

HSR is actually a great complement to BEVs precisely because of the range issue. You can buy a lot of train tickets for the cost of the battery capacity needed to drive an extra 20 miles per charge cycle. Indeed, HSR will let city dwellers get away with not driving a car at all in summer - an electric bicycle will do nicely around town, especially in the new transit-oriented neighborhoods.

4) "Of course, the notion that HSR absolutely MUST cover its costs at the farebox is another case of trains being held to standards that freeways and airlines aren’t."

The civilian aviation industry actually does pay for airport construction, air traffic control etc. via airport taxes. However, kerosene fuel is not explicitly taxed. In addition, the feeder infrastructure for getting passengers and cargo to and from airports - principally, freeways - is paid for out of general taxation.

The 2001 bailout was indeed an exceptional case. More valid example of hidden subsidies include inflated military R&D contracts with aircraft manufacturers and, overly generous Chapter 11 protection of airlines.

As for freeways, they need to be built and maintained, but the public does not need to fund operations staff or fuel - private citizens take care of that. Of course, this glosses over the fact that freeway construction cost per passenger-mile is massively higher than for passenger rail.

In fact, the state could easily afford to subsidize operations for many years with the difference. Fortunately, HSR is a service many will gladly use, so that won't be necessary. The savings can and should instead be invested in expanding and upgrading general rail infrastructure: freight, commuter, subways etc.

Anonymous said...


Thank you for providing some quantitative data. I had missed reading that the trains were going to carry 1600 passengers and previously used 800 passengers as being the capacity of each train. 1300 feet is one very long train.

The bottom line economic numbers still make no sense when using $55.00 fares in 2020 when today operating systems are charging at least twice that much and up to 4 times that much on comparable stage lengths

The Cambridge study being funded by CHSRA is extremely suspect, to say the least. They invoke sensitivity analysis to justify their numbers. The academic studies by Levinson and others do not agree at all with their numbers.

On HSR being a really green and fuel efficient alternative to air and auto, one should really think about why this train should try to achieve speeds of 200 MPH and higher. Aerodynamic drag is the major demand on power for the train. Drag increases at the square of the speed. So if the train runs at 100 MPH instead of 200 MPH the drag would be one-fourth. Drag on the system consumes about 60% of the power, so traveling at 100 MPH would use only about 55% of the electricity needed to power the train as will be needed at 200 MPH. Electricity is not free nor of infinite supply.

Robert Cruickshank said...


The only way American gas prices will come down is if global demand drops, leaving us with more oil. If new production comes online it will merely feed global demand - we'll be competing with China and India for that oil. Further, it's not at all clear that new production will offset declining production in places like the North Sea, Mexico's Cantarell field, or even Saudi Arabia.

No, gas prices would only come down in America if demand for that gas had dropped dramatically - and stayed low. In other words, it'd be low because nobody wanted it. Which then requires available alternatives, like HSR.

As to airline and freeways subsidies, I never claimed that their entire cost was subsidized. But much of that cost is and remains subsidized by government - when gas taxes are insufficient to maintain the freeway network we turn to local sales taxes, where nonusers are subsidizing users. Airlines and passengers pay some of the cost of airport expansion but taxpayers foot the rest of the bill. Airlines exist because of government subsidies, especially carrying mail in the 1920s as well as the ongoing subsidies you mentioned. Finally, is 9/11 really so different from the impact of peak oil and global warming? All were unanticipated shocks to the system, but with 9/11 we happily opened our wallets to bail out the airlines. Why should trains be treated any differently?

Ultimately you and I agree that the likelihood of HSR covering its costs and turning a profit is extremely high. I'm just saying that in the remote case that it doesn''s still OK, and the system should still be built.

Morris, the EIR has taken the electricity generation factor into consideration and even with that, HSR is still a significant savings on carbon emissions. Although I do hope that the CHSRA is able to implement its carbon-neutral generating plans, which are the subject of an ongoing study that should be released in June or July.

Rafael said...

@ Morris Brown -

please don't put words in my mouth. I didn't say HSR trainsets in California would be 1300 feet long and offer 1600 seats each.

I merely pointed out that this is the ultimate capacity the system is being designed for and, that this would not actually be used for several decades. For example, architects are reserving 1300 feet (~1/4 mile) for the platforms at major stations like San Francisco Transbay Terminal and LA Union Station. This is comparable to building freeways with a generous median for future expansion.

As for the ridership study using sensitivity analysis, how is that methodology "suspect"? You need to look at the supporting documents before you disparage the whole effort. Predictions are always wrong, especially the ones about the future. However, you don't even provide a link or proper reference to the Levinson study you hold in such high esteem.

Also, the Cambridge Systematics study explicitly refers to 2005 dollars for its estimate of ~$55 for a one-way ticket from SF to LA. Quentin Kopp recently asserted otherwise but I think he was mistaken.

Other HSR systems around the world do indeed charge more, but that's mostly because they can: gasoline and diesel are taxed much more heavily in Europe and Japan. The study actually suggests that increasing fares by 25% would reduce ridership by 13% but only increase revenue by a paltry 2% - moving empty seats around the state is not free. Of course, the price of California gasoline has risen sharply since 2005, so HSR fares and operating profits may yet end up higher than forecast.

As for energy consumption, even at 220mph trains are much more benign than transporting the same number of people with either cars or aircraft. Peak electricity consumption for the entire HSR system running will be on the order of 300-500MW, about the size of the Ivanpah solar thermal plant in the Mojave desert. Such plants are not cheap to build, but the fuel cost to run them will be zero forever. Same for wind and geothermal. However, the state as a whole currently uses ~50,000MW generating capacity - HSR isn't going to make a huge difference either way.

The 220mph top speed was chosen to lure passengers away from both of these alternatives. At 110mph ridership would be much lower, undermining the whole business case. Don't let the tail wag the dog.

無名 - wu ming said...

while we're talking about peak oil, it's also worth remembering that asphalt road surfaces are made from - you guessed it - oil, and subject to the same price pressures as the gas in the cars above the surface.

concrete roads are a better deal in that they don't use oil and last a hell of a lot longer, but then again they tend to be a really big source of greenhouse gas emissions.

rail's a good move for peak oil for a host of reasons. maintaining our road system's going to get a lot more expensive in the near future.

Anonymous said...


For one of the studies that Levinson was involved with go to:

Since this was from an academic study I certainly trust their numbers much more than what Cambridge produced. What was remarkable about the new Cambridge numbers was how much more glorious and wonderful they were as compared to earlier work by the CHSRA

Anonymous said...


Sorry the blog cut off the end of the URL In two pieces the URL is:

If you copy and paste this should work

DK said...

So I looked at Levinson the study, (yes I probably should be working more diligently at my job)

Several problems emerged:

1) they model airport delays based on number of runways and scheduled flights at each airport instead of using actual historical delay times. This creates an average delay of 2.24 minutes at LAX and 4.47 minutes at SFO.
-These do not account for real delay times at these airports
-They do not account for the fact that intra-CA flights are delayed more often than others
-They use 2000 flight volumes instead of more recent flight volumes
(see page 3-40)

2) More importantly, the schedule does not use door-to-door times for any mode. Instead it assumes a fixed 677km trip. It also assumes a 877km per hr speed by plane, i.e. a 45 minute trip by plane start to finish. This underestimates the time cost of flying (which they do calculate)

The whole advantage of HSR is that it goes city center to city center and doesn't have to go outside of the city to the airport wasting time in security.


Basically: do you think that it takes longer than 45 min. to get to your LA destination from SF origin by air? If so then you disagree with this study.

That's just one of the "academic" assumptions that doesn't necessarily agree with reality.

Don't forget, it assumes that your time on HSR is wasted, which with WiFi it isn't! Hooray for technology, maybe planes will have that eventually...