Hacker News Reader

Comment by kevindong | Hacker News Reader

> the Twin Cities-Chicago market is definitely a route that deserves better service

For reference, the distance between Minneapolis and Chicago is about 400 miles.

If the Amtrak schedule [0] is strictly adhered to, the travel time is ~7.75 hours. Current prices are ~$116 roundtrip.

Google Maps says the estimated time by car (assuming no traffic) is ~6.25 hours. At ~30 miles/gallon at $2/gallon, the cost of a roundtrip for just gas would be ~$53.

A roundtrip, non-stop flight lasting ~1.5 hours each way can be had for ~$80~$175 with the passenger's choice of at least 10 distinct flights/3 separate airlines.

I like rail. But it's clear that there's not a market fit for rail between Minneapolis and Chicago without massive subsidies. A car is faster and cheaper. A plane is much faster and about the same price. Both car and planes are strictly more convenient.

[0]: https://www.amtrak.com/content/dam/projects/dotcom/english/p...

View on HN

It's fun that nobody ever includes the cost of the Interstate Highway System into these throwaway calculations, as if the roads just magically appeared and are maintained by faeries. As if the roads, car companies, and the airlines aren't "massively subsidized" with energy (fuel and corn->ethanol) subsidies.

We spent half a trillion dollars building the IHS over five decades, and have historically thrown at least a few billion a year at US oil producers - is it really surprising to anyone that driving is cheap? We keep bailing out the airline industry repeatedly - we already gave them $25 billion in grants this year because of the coronavirus, and they're still asking for more.

But, dare we build an alternative to those systems, one that could be more ecologically and financially sound? Nope, too busy feeding our sunk cost industries. Keep blaming Amtrak for being terrible while refusing to give them their own run of rails so they don't constantly get stuck behind freighter traffic. Keep shuting down High Speed Rail projects around the country because the billionaires don't want it in their back yards.

We raise ~36 Billion dollars in fuel taxes per annum. That's why no one complains about the Interstate Highway System, it pays for itself.

https://www.fhwa.dot.gov/policyinformation/statistics/2016/f...

From the American Road and Transportation Builders Association, hardly a negative source about highways: https://www.artba.org/about/faq/

> Highways: The FAST Act provides $45.3 billion for highway and bridge improvements in FY 2019 (the fiscal year from Oct. 1, 2018 through Sept. 30, 2019) and grows this amount to $46.4 billion by FY 2020. Most federal highway investment is used to upgrade and maintain the nation’s core highways, including the Interstate Highway System, and to repair and replace deficient bridges.

> An additional $3.4 billion was provided for highway and bridge investments as part of the final FY 2019 spending bills signed into law Feb. 15, 2019.

> Since 2008, revenues to the Highway Trust Fund (HTF) have been insufficient to fully support the level of federal highway and transit investment authorized by Congress. Prior to enactment of the FAST Act, the annual investment gap was nearly $15 billion. Congress has not increased the federal motor fuels tax since 1993, but federal highway and public transportation spending has grown substantially over the last 26 years. It should surprise no one that holding the trust fund’s inflow of revenues constant while attempting to address the nation’s growing transportation needs would lead to an unsustainable situation.

It is quite clear that even just the Interstates do not "pay for themselves." There's also the issues with the fuel taxes, namely that motorists are not necessarily the only ones paying those taxes, and in addition to the depreciating value from a flat fee not tracking inflation, the value of the tax is continuing to diminish as some motorists switch to alternative fuels and others are driving ever more efficient vehicles.

Hm, using your own link, I see "Federal aid to highways" totaled at $43,421,077,419, and I see "Total excise taxes" at $42,329,411,402. I do a subtraction and I see a $1,091,666,017 deficit. Seems to suggest that, in fact, the Highway System does not pay for itself, but rather we pay for it, no? (It get worse if I just use gasoline taxes and ignore the rest.)

I do see their balances coming up positive though, thanks to a very massive balance transfer from the General Fund per the FAST act (https://www.congress.gov/bill/114th-congress/house-bill/22), straight from US taxpayers' wallets.

I also see no mention to climate impact anywhere. $36 billion dollars of gas taxes at 18.3 cents per gallon suggests sales of nearly 200 billion gallons of gasoline were burned. That's a hell of a lot of climate damage that everyone's being forced to pay for.

So, you were saying about us not subsidizing the IHS?

The budget for the highway system is managed by the Highway Trust Fund. The trust fund had been solvent until 2008, but has since needed additional funding to meet outlays. Prior to that the system had been working as outlined in the Federal Aid Highway Act of 1956. So yes the taxpayer has been subsidizing the highways for the last 12 years.

So the system is broken, but it has a simple fix, just raise the gas tax. We haven't done that since 1993 and it isn't indexed to inflation. This is something that should have been fixed years ago, but alas I don't control congress.

If you want to go even further and discuss the environmental impact, again the solution is more taxes.

> So the system is broken, but it has a simple fix, just raise the gas tax.

Given the latency on legislatively fixing things in the USA these days, any work at fixing stuff now should probably include the likelihood of a major shift to non-gas vehicles in the next couple decades.

Your simple fix would change the calculus when comparing road & rail travel, which I think is the point of the person you were disagreeing with.

Which is a fair argument, but using the current math it's about half the cost of rail. So even if you need to increase the gas tax in order to "make it pay for itself," you're absolutely not going to do that so much so that you double the cost.

It's hard to make an argument that rail is cheaper than driving without getting into a lot of mental gymnastics, and I really like taking the train when able.

I don't think "cars cost more than just the price of gas" is mental gymnastics, but an exact $ figure we can all agree on will be tough.

Once they become a thing, the price of an autonomous taxi between the cities should approach the real amortized cost of road travel.

Inflation-indexing might help for a bit but there are longer term issues on the horizon.

Namely, as fuel efficiency increases and some drivers switch to alternative fuels, people will need less gas for the same amount of driving.

conveniently ignoring the external costs of driving.

And the external costs of rail.

I am often guilty of ignoring real costs of car travel. You didn't even mention the purchase price of the car which is included in the rail ticket. Or the rest of the $0.55/mile in tire maintenance, insurance, oil changes, battery replacement, etc.

For a personal calculation it is a moot point. You're paying the taxes to maintain those roads even if you take the train or a plane.

A true market based solution would be to make all major roads toll roads and use the funds collected to maintain said roads. With EZ-Pass/SmartTag/etc... this could be practical. The problem with toll roads is that local governments tend to use them as a piggybank so you end up with poorly maintained expensive road.

While this looks good at first sight, it reminds me of robbery, wegelagerei/highwayman when thinking longer about it. I mean you need roads anyways, always, if only for emergency- and utility services. Are they exempt, priced in? What about bicyclists? Will there even be roads accessible to bicyclists between cities in such a system? Or can they simply fuck off, because no market for it? What about freedom of movement?

Gas cost isn’t a great stand in for car trip prices, as there are a plenty of other costs to drivers.

The IRS per mileage reimbursement rate is 57.5 cents per mile, putting the “real” cost to an 800 mile roundtrip around $400.

A different stand in might be the per mile rate I can get from a rental company, which is often around 20 cents a mile. That would be $160 in other costs + the $53 in gas $213.

Now, people chronically underestimate the cost of driving (and many of them are fixed costs) so the “competitiveness” of rail doesn’t really change, but IMO we shouldn’t just look at gas costs for automobile transport.

In Germany, the state rail company (Deutsche Bahn) tries to address this problem by selling rail discount cards.

The idea is that people only take into account fuel costs when considering driving. They don't typically take into account capital costs like depreciation. So the DB sells a 250 Euro card that gives you a 50% discount on tickets for a year. The card creates an artificial sunk cost, and lowers the marginal cost of taking the train, making it more comparable with driving.

> So the DB sells a 250 Euro card that gives you a 50% discount on tickets for a year.

A nitpick that's so small as to almost be irrelevant: For people up to (and including) the age of 26, it's actually 60€/year. This makes going by train a financially pretty attractive travel method.

Or even the 100, which retails at 3878€ (2nd class)/6560€ (1st class) a year and allows pretty much unlimited travel, forever, everywhere.

The SBB equivalent is Switzerland is even cheaper! 165-185 CHF/year (a bit lower in EUR) and now it's cheaper for younger passengers, which means many people get it without really thinking about it because it almost inevitably pays back even with many local transport tickets or certainly intercity travel. A few trips from Zurich to Geneva would mostly or completely offset the cost depending on your ticket.

These too are just the financial costs of driving, they don’t account for the long term environmental costs.

> people chronically underestimate the cost of driving (and many of them are fixed costs)

Once someone has paid the fixed cost of owning a car, that cost is sunk and shouldn't be factored into calculations for individual trips. The only thing that matters at that point is marginal cost.

For public transportation to be competitive, the overall public transportation system has to provide a level of service that makes it feasible for many people to not own a car at all. For example, having great inter-city train service doesn't help if I need a car to get to the train station. If I have to own the car anyway, and the marginal cost of driving is lower than the marginal cost of taking the train, I'm going to drive.

In most European countries, the overall public transportation system does provide this level of service. In most places in the US, it doesn't. A significant part of the reason for that is simply that people in Europe and people in the US have a different mix of preferences. Many more people in the US do not want to live in population dense areas that can support a public transportation system that makes it feasible to not own a car. (I'm one of them: I have lived in population dense areas and I know from experience that I prefer not to.)

This is a chicken and egg problem.

Of course there's hardly any demand for a slow, expensive and often late service which only runs once a day. It's not useful.

A journey I do often in the UK is London to Edinburgh by train. The distance is almost the same (400 miles). London is a bit bigger than Chicago (12 million metro vs ~8 million metro) but then the Twin Cities are quite a lot larger than Edinburgh (1 million metro vs ~4 million metro).

However, the track has been improved, is at least doubled the whole way and has significant portions at 125 mph. As a result the journey takes just over 4 hours. If the trains are more than 1 hour late you get all your money back.

On an average weekday there's 41 trains travelling each way between those cities. And many of them are full. They're constantly doing improvements to try and increase the number of trains travelling.

I understand it's a different situation - the result of decades of political apathy towards railways. But I think it's important to understand that the Amtrak situation doesn't have to be the way it is. There's not an inherent lack of demand for fast, regular train services between major cities.

Berlin to Frankfurt is also 4 (ICE) to 7 (IC) hours, and has a bunch of trains every day, and in my experience (pre-Covid) was usually pretty full when I took it.

Lots of countries have good rail service that people actively choose over cars or planes between major cities. I find it maddening that my own country can't bring itself to really want that.

> A car is faster and cheaper.

I don't own a car and never learned to drive. What's my upfront cost before the $53 incremental cost?

Car ownership begets car ownership.

AAA says car ownership costs $8,558 per year. I'd have a hard time spending that much on public transport.

https://www.aaa.com/autorepair/articles/what-does-it-cost-to...

Also consider the health benefits and reduced medical costs if your lifestyle includes bicycling or walking.

until someone hits you with their car or you catch pneumonia from cycling in february

All-cause mortality is significantly lower for cyclists. Yes, motorists do kill a lot of people, but in the big picture they're still less dangerous than a sedentary lifestyle.

As a former bike messenger, I can't relate to this kind of fear mongering.

Winter biking is enjoyable when you dress properly for the weather and use spiked tires.

https://valientemott.com/blog/chances-of-dying-in-a-car-cras...

Motorists kill more motorists than they do cyclists. Did your car calculation include the odds of getting hit by some other moron in a car while you were driving to your destination?

Then you are doing it wrong. Where does one catch pneumonia from driving in the cold?

The linked article clearly states that number only applies to new car ownership. For reference, I have a 20 year old vehicle which I own outright, runs great, and drive maybe 6k miles a year. I usually bike and/or walk to work (or did, before COVID.)

Buy a used car and do your due diligence before purchasing.

Not sure what state they're using for license, tax, and registration. I pay less than $100 a year for that.

Maintenance is about right and can be lower, especially if you don't buy junk.

My insurance is lower than that, but that's a perk of my state.

I fill up maybe once every three weeks, so my fuel usage is way lower than that.

It'll probably go to the scrap yard once I'm done, and this thing was worth around maybe $3k when I bought it some years ago.

All in all, I'm sitting somewhere around $2000 a year for my car. Totally worth having.

Just be smart about it. Cars are very useful to have.

You listed a lot of ways to pay in time and energy instead of cash.

(Disclaimer: I own a car)

The only thing I see in the list that takes time is doing due diligence when buying a used car, which isn't really a reason not to buy a car. You should be doing a little research for any $10k+ purchase, I would think.

I agree with you but there's no way the number is that high. My used car cost $3,000 and for three years the total cost of service has been less than $600. Add in $300 per year of insurance and you're looking at a yearly cost of $1,500 tops

> What's my upfront cost before the $53 incremental cost?

$400ish (get lucky on a $250 "ran when parked" car, pay by month for insurance and live in a low tax state) on up.

>AAA says car ownership costs $8,558 per year

They say it costs that on average. That's very different from the floor or baseline cost which seems to be what you're implying.

>$250 "ran when parked" car

How to get an absolute hooptie sitting in your driveway waiting for replacement parts.

I said "lucky" for a reason.

If you can't source parts in this day and age it's not the car's fault. Anything made in the last 30yr you can readily get service parts for. What you can't get at the parts store are interior trim and non-wearing suspension items that are specific to a narrow range of vehicles (also true for new cars) but if you expect to keep the interior of a bottom dollar car in tip top shape you're out of your mind. The internet makes finding everything that isn't a normal wear item super quick and easy but you're gonna have to wait 24-96hr to get it.

Hrm. If the average car lasts 200,000 miles, then 800 miles is 0.4% of that, and if the car cost $30000, that alone is $120.

Just some food for thought -- of course, if you have passengers it'll quickly be worth it even if you decide to factor in this kind of calculation.

That analysis is incomplete because any car that gets used at all ages by time as well as by mileage. The car is going to depreciate and degrade on a similar timeline regardless of how much you use it. Keeping miles off just changes the shape of the depreciation function.

That's true especially in regions where there's salt on the road in the winter. But depreciation is usually a lot more mileage-based than time based.

At the end of a car's expected life, absolutely. But compare a two otherwise-identical 2016 or 2017 models, one with 30k miles and one with 120k miles.

And ~8yr from now the depreciation functions for both those will intersect.

The only difference is depreciation now vs later.

There are a ton of subsidies at play behind these numbers. Cars and planes get the massive subsidies of non-taxed externalities due to carbon emissions and air pollution, for instance.

There's a lack of long-term policy here: do we want to be in a society in 20-30 years that's still using personal cars to do that journey, when a high speed rail would be much more efficient and pleasant? If a true high speed rail were enabled, that 400 mile trip might take 3 hours instead of 6-8 for other modes of transit, and it would incur less negative externalities to boot.

I'm all for individuals making informed decisions in the short term, while government and society should be taking long-term steps to promote the health of the commons as well.

The core use cases that the long-haul Amtrak solves are not day trips between large cities. You are right, there are other solutions for those.

But Amtrak is the best fit for multi-day trips across the entire country for people who cannot fly. Amtrak also solves transportation for travelling directly from small rural towns to the large cities, and sometimes even from one rural town to another. It is the safest way to get across mountain ranges in the depth of winter (although not the fastest).

It doesn't solve every problem, and it does need subsidies, and it always runs late... nevertheless, it connects rural Americans to the large cities in ways that city folk don't even realize.

With a car, you can go point to point. With a train, you have to arrange transportation to the station on both ends. Also, you can leave when you want, you don't have to time your travel with the train or plane schedule or worry about delays and cancellations. To me, up to 6 hours, a car almost always makes sense. Above 6, usually a plane. Rarely does a train beat out both alternatives.

The only time I've found the train beneficial for me is rural/suburban -> urban travel, where the train ride itself is about 3 hours. I live in the middle of nowhere and have gone to NYC via train many times. The train is about 20 minutes longer than driving, I don't have to drive in NYC, and I can park at the rural train station for $5/day. Now granted that could very easily go out the window if there are strict time constraints around when I can travel, and I end up driving because I can leave when I need to, and it's actually faster.

Agreed though that car or plane beats train 99% of the time.

It doesn’t have to be this slow or expensive. A sprinter train from Berlin to Munich, which is a similar distance, takes a bit less than 4h, non Sprinter is 1h more with Tickets starting at 17.50€

But this starting price can be had only if you book way in advance. If you hop on the train from Berlin to Munich today, it will set you back between 93€ and 135€.

Yea the pricing structure is weird.

https://bahn.guru/

Helps to find cheap tickets, so when you are somewhat flexible you can still travel for cheap.

There are a lot of great counter arguments to what you said, but as someone who is from the Twin Cities and studied in Chicago, I'd recheck each trip and always go car or plane. Often I could car pool, and splitting with one person made zero interest with going train. Train should be easy and way more competitive than it is.

The cost of tolls, gas, depreciation per mile, depreciation time, monetary risk of accident, and etc could definitely make driving less economical than taking the train.

Unfortunately car owners tend to ignore depreciation and risks of accidents in their cost estimates, but include comforts like the ability to carry more luggage.

What happens when you get to your destination? Do you suddenly just stop needing a car?

Flying is pretty popular and cars don't generally fit into the cargo hold of airplanes.

They do... ok, you could afford a pretty decent used car (or several months of car rental) for the shipping price: https://www.autoshippers.co.uk/air-freight-car-shipping.htm

From the 1950s to the 1970s there was a regular service transporting cars by air between the UK and various airports in France, Belgium and the Netherlands. It cost about twice as much as the boat from Dover to Calais.

They initially used Bristol Superfreighters, which were replaced by Aviation Traders Carvairs (converted for the purpose from second-hand Douglas DC-4s). The Carvair could carry 5 cars and 22 passengers- the cars were loaded through a door in the nose.

The "air ferry" service was killed by competition from the new RO-RO car ferries (when it started flying, cars still had to be loaded onto cross-Channel ferries by crane) and especially by the hovercraft service which began in 1968- a crossing by hovercraft with a car didn't cost much more than the conventional ferry while being as fast as the plane. The hovercraft, in turn, stopped being viable because of competition from the car-carrying trains through the Channel Tunnel- it stopped running in 2000.

Alternatively, you pay up to $60/night for the privilege of keeping your car underneath your hotel while you walk or ferry or scooter or even uber your way around your destination.

I'm a car owner but it's stupid expensive and if there were a reasonable alternative I'd stop in a heartbeat.

Rumors have it that there are cities which are actually walkable AND have acceptable public transportation!

Imagine!

The other difference with a car is that, unless you're going downtown to downtown, you may need a car on the other end anyway.

One of the reasons I always take a train to Manhattan is that I not only don't need a car but I hate driving into the city and parking is expensive.

That applies equally to air travel.

I've heard Uber and Lyft are popular.

Cities and regions built with density, walkability, bikability, and transit in mind avoid or greatly reduce this problem.

You can take limos/taxis for multi-hour drives but I expect that's relatively rare.

I'm about 200 miles from NY and actually backtrack to take the train. But I'm still guessing that an Uber (if I could get one) would be a a good 2x the train and a private car even more.

Some areas make it easier to get around other than in an auto than others. Of course, I'm often going somewhere out in the suburbs of a city anyway.

Uber or Lyft would provide local point-to-point transport once at the long-haul destination.

If you're making extended or multi-stop day or multi-day trips at a destination, you no longer have a simple long-haul transport need. A local car hire, group tour bus, bicycle rental, transporting your own bike, or similar arrangement could satisfy this.

Boeing 737s and Airbus A330s would be equally or more handicapped under such circumstances, though a Cesna or comparable bush plane might be appropriate in some regions.

A lot of people are pointing out that you are only including gas costs, but they are also ignoring the fact that cars get you precisely where you want to go. Trains only get you into the city you are going.

Cars get you to the nearest parking lot to where you want to go. Depending on the city, that could be further away than the nearest metro or rail station to your destination.

Once I got above a threshold compensation package where my time became far more valuable than occupying it with attention-grabbing activities around driving and flying, I came to appreciate trains. I can see how this calculus can change yet again back to valuing money more than freed-up time when one has a family though, if one were strictly evaluating as homo economicus: non-working spouses and children can add significantly to travel costs. However, I've frequently seen families prefer train travel for the time spent together it engenders, with one challenge being coordinating travel logistics once at the destination while wrangling children (ride sharing has alleviated that, I'm sure).

On the super-long transcontinental trips, apparently the cell reception with the right gear on trains is much better than when I rode half the country westward many years ago. That makes the train travel mode even more valuable to me if I'm not on a tight schedule and the seating arrangement enables me to look forward to getting work done.

What prevents me from using trains more than I do is the demands of many clients upon my time. Not quite lucrative enough for me to justify a satellite data link, but if my financial picture improves by many orders of magnitude, a digital nomadic life with a private railcar set towing shipping container footprint custom structures as a "home base" that I operate vehicles from would be an enjoyable way to my mind to get deeply into locales, while staying in touch with the Net and maintaining full fabrication capabilities with me.

The Obama Administration tried to build high speed rail from Chicago to Madison, WI. It was supposed to be the first segment that would eventually extend to the twin cities. They have $700M dedicated to it. Then WI had an election for governor, and the incoming Republican (Scott Walker) managed to somehow conflate high speed inter-city rail with 'light rail' and scrapped it. He claimed it was too expensive, since the state would have to spend $7Million or so each year for their part of the maintenance (even though the feds were going to fully fund construction). After Scrapping it, the State then had to pay $90Million to upgrade the tracks between Chicago and Milwaukee (that was going to be paid for) and then had to pay $40M for the high speed Talgo Trains. Plus, Talgo moved their US headquarters from Milwaukee to Chicago.

Interestingly, it wasn't the Obama administration that tried to build the high speed rail. It was the Republican governor of Wisconsin (Tommy Thompson) that pushed for it. Scott Walker used it as a wedge to split the electorate and win the election. Wisconsin Public Radio did a great podcast about it here: https://www.wpr.org/derailed/wisconsins-high-speed-rail-saga...

The route is seriously dumb -- goes straight to Milwaukee and then south, versus going directly through Madison. Badger bus already does MSN-MKE cheaply and quickly, and there's almost no chance that this routing could ever be time-efficient versus the direct route.

The route in Madison was supposed to go to Monona Terrace, which is singularly inconvenient for most people, limiting the catchment basin. Getting out of Madison requires huge numbers of grade crossings, limiting speed, and disrupting traffic.

So this route was never going to be anything other than a slow drain on the treasury, and SW wisely killed it.

We could really need the Milwaukee to Chicago train up to speed. I'm in Milwaukee and would go to a Chicago office twice a week on that train and it just got tough with 1:30 each way. The train is packed too, where on the way back in the evening, people can be standing because not enough seats. Higher speed, like cutting it down to :45 or :50, would mean a ton. The problem is the value of that isn't able to be calculated and in pretty much all infrastructure cases, it gets purely political.

FWIW when driving, tolls along that route are roughly $10 each way... and that's low for the midwest.

so that's worth about $10 more for a grand total of ~$63 for the trip

Tolls are actually around $5.50 one way, and they're only in Chicago. You can bypass tolls by taking 41 through Chicagoland and add maybe 15 minutes to the total drive time. No tolls once you're in Wisconsin and Minnesota.

Depending on whether you fly into O'Hare or Midway, you've got a significant or a somewhat significant journey into Chicago after you land. This doesn't negate your point, but needs to be taken into account for city-to-city service.

Depends on where you go. The arrival is written by someone who lives near the airport in Minneapolis and had to drive (or?) a considerable distance to the train station.

I don't know if the the schedules are better out east, but here in CA the Amtrak schedules are wildly optimistic. Like, double the listed travel time. And they run infrequently, at odd hours.

Surely this calculation is incomplete without considering depreciation. Also, it makes sense to consider a portion of the cost of the insurance, maintenance, etc. towards the vehicle cost.

Chicago - Indianapolis is also another route that would benefit both cities immensely.

And Chicago - Milwaukee. We used to have a train that ran as often as the CTA that you could hop on on the Union Loop downtown that would take you to Milwaukee. Now? You still see some people commuting on the Hiawatha but there's still only 9 of those a day.

Reply on news.ycombinator.com