> Do people really believe that infrastructure is not the purview of the government?

Well, the people behind these Koch-funded groups do. They believe that nothing is the purview of government, or at least nothing that can be monopolised and exploited for profit.

I think the purview of government is primarily to protect the people against this kind of monopolisation and exploitation. Privatised infrastructure is fine with me as long as it doesn't hurt people's access to it. But in this case, the government is clearly responding to a lack of access.

Do they really believe that, or do they simply need us to believe that so they can keep pillaging the country?

There was a great Freakonomics podcast[0] where they interviewed a Koch brother. It seems like they truly believe that is what would be best for the country. They are Libertarians and believe in it's tenets to their core.

[0] http://freakonomics.com/podcast/why-hate-koch-brothers-part-...

That's pretty much my take on the Kochs as well. They seem to contribute to things they believe in, which incidentally includes public radio stations and hospitals [0]. You may not like their politics--and many do not--but so far everything I have read or heard about them supports the notion they are sincere in their beliefs.

The problem is not so much the Kochs as the fact that we've permitted a political system to arise in the US that allows a disproportionate amount of wealth to flow to a relatively small number of people and to allow those lucky few an outsized influence on the evolution of said system. Those two things are not unconnected.

[0] https://www.usnews.com/news/the-report/articles/2015/06/26/t...

Upton Sinclair's "It is difficult to get a man to understand something, when his salary depends on his not understanding it." still holds true here.

Read Jane Meyer's Dark Money.

https://www.nytimes.com/2016/01/24/books/review/dark-money-b...

I think they would agree that infrastructure falls to the government... the disagreement arises in what "government" means. Their views seem more aligned with feudalism/monarchism, subjugating the public by convincing citizens to surrender key infrastructure to privatized rule.

Rich people can't create monopolies and rent seek as easily if government does the basic infrastructure well.

Peter Thiel gives the example of not wanting to open a restaurant in San Francisco because there is too much competition for it to be successful and you constantly have to work at it for little reward.

Of course the reality is that a successful San Francisco restaurant makes millions of dollars per year. Too little for Peter Thiel to be interested in but for most of us that would be considered extremely successful. The same can be true of websites and the Internet, ala Indie hackers.

>Of course the reality is that a successful San Francisco restaurant makes millions of dollars per year.

Selection bias anyone? Of course a 'successful' $X makes money, it wouldn't be successful otherwise. That tells us nothing on how many restaurants had to fail in relation to the ones that did well. From a hypothetical investor's point of view if 1 out of 10 restaurants does well enough to earn a payback and you would have to put $1 million in each one to get it started, then the one that succeeds has to profit at least $10 million for the investor alone. Yea, seems like a pretty high risk environment.

Both your analysis and Peter Thiel's are coming from that weird place where becoming filthy rich is the only marker of success, rather than opening a business, building its loyal clientele by offering a valuable service they appreciate, employing people which helps their own lives grow and being able to support your own self from all that.

What the f would Peter Thiel know about the joy derived from feeding someone good food?

> Both your analysis and Peter Thiel's are coming from that weird place where becoming filthy rich is the only marker of success

Actually, it looked like his analysis was for what was required to break even, on the average. Invest 10 million (across 10 companies), get back 10 million in profit from the 1 (of 10) that succeeds.

I am trying to point out the ludicrous attempt to apply tech venture capital strategy to starting restaurants.

There is no way anyone is going to open ten restaurants simultaneously, fail at 9 of them and earn 9 million in first-year net profit from the 10th. That's not how opening or operating restaurants works.

Instead one person opens one restaurant and they have a 90% chance of failing. That's so much worse!

I think you have that wrong, 90% of people fail at opening and operating restaurants.

Some folks have a very high success rate.

I think the failure rate I’d restaurants is more tied to the types of folks who open restaurants than the business model.

There are many other things one can derive joy from that don't require putting in 90-hour workweeks, won't cause crazy levels of stress and aren't likely to result in boatloads of debt.

Sure, if sufficient care, attention, and elbow grease made it a near certainty that even a halfway competent owner/manager could take out, say, $100k per year to live on, then you might have a reason to be a little indignant toward Thiel.

The restaurant market is notoriously fickle with considerable risk of ruin. Between opening a restaurant and building its clientele is a lot of uncertainty. It’s easy to look at the surviving restaurants and think that anyone could make a comfortable living at it. The economist must consider both the seen and the unseen — of which the latter in this case is all the restaurants that started and failed.

> filthy rich is the only marker of success

it's the only marker that others judge you by.

Said the king, before the guillotine fell and decapitated him

I think his line of thinking is more like, why spend so much time trying to get X to work, when I can just do Y and make the same amount of money? Unless you really really want to run a successful restaurant, he's not wrong.

If it takes 1/10th the effort to run a cleaning business and make $1m a year, vs a tech business, would you do it? Some people would, others won't. Operating in a space with (not just normal but) extreme cut throat competition can get tiring after a while as well.

And even if you really want to run a successful restaurant, capital doesn't give a fuck. Banks won't lend you more money because you have a passion. Investors won't either.

Drugs cartels seem to emerge without government assistance.

"Do people really believe that infrastructure is not the purview of the government?"

I don't think anyone is arguing that the government shouldn't be responsible for certain infrastructure, it's the degree to which the government should be involved, and what its role should be. If I put on my more conservative hat for a minute to make an argument, I think the idea is that the government should be somewhat of a last resort for things, and that less is more when it comes to government involvement.

Part of the reason for that is that once the government is involved, it changes the incentive structure quite a bit. Using other infrastructure projects (e.g. roads, bridges, etc) for examples that others have pointed out, they're a great way to raise taxes earmarked for those projects, but then that's when politicians start to play games and move money around to their own projects, and now the roads are falling apart, so we need to raise taxes to pay for the roads!

It's also somewhat tricky because once something is a government project, it's basically that forever. It's fairly rare for something to go back to the private sector once it has become a government program. What is more likely is that the solution is more government involvement, and that's something that conservatives tend to view as a bad thing.

Obviously I'm not speaking for all conservatives, and this isn't even necessarily my own viewpoint, but I wanted to push back against the idea that "conservatives just want what's worst".

> once something is a government project, it's basically that forever

Hardly. In my lifetime the UK government has privatised electricity, water, gas, telephones, British Aerospace, British Steel, British Petroleum, Rolls Royce, Rover, British Airways, railways, coal mining, and Royal Mail. Experiments have been made with roads, bridges, schools and hospitals.

Success of these has varied. The utilities make a big profit while people struggle to pay utility bills: https://www.ft.com/content/ef488dc0-9168-11e7-a9e6-11d2f0ebb...

BAe and BA are still in business. British Steel were owned by Tata for years and the few remaining sites are now spun off and trading under the original name. BP are making a profit except when they spill in the Gulf of Mexico. The railways are a mess. Rover were looted by their directors who narrowly escaped jail.

> Success of these has varied. The utilities make a big profit while people struggle to pay utility bills

That kind of thinking absolutely kills infrastructure investment. To use British Telecom as an example. After privatization, BT kept its monopoly on the last mile, but rates were set to ensure that it was quite profitable. (About as profitable as say Time Warner in the US.) That ensures that BT had the incentive to invest and keep pushing fiber deeper into its network. Today, broadband speeds in the UK are significantly faster than in say France. But it also results in huge political backlash. “BT makes 20% profits while people can’t pay their utility bills!”

Except that Profits = income - Expenses (in simplified form).

So, that 20% Profit you mention is money NOT spent on infrastructure investments.

Sure. But it’s money that motivates BT to spend money on infrastructure investments. No matter how you structure your infrastructure markets, ensuring adequate investment is the single most important thing. You can do this with markets or quasi markets, or you can do it through government agency decision making. With the former, you need to ensure attractive returns on investment. Not merely bare profit, but enough profit so that the billions of dollars don’t get invested elsewhere. Having the government make the decisions doesn’t magically eliminate the issue of ensuring adequate investment. It makes it even harder! With the government is on the hook for making investments, you’re always subject to pressure from other priorities. There is a reason roads and bridges and water systems in the US are crumbling. We’ve dramatically underinvested.

Your examples illustrate perfectly how government regulation establishes monopolies and cronyism

And Standard Oil, Alcoa and Carnegie Steel/U.S. Steel demonstrate how lack of government regulations establish monopolies and cronyism...

Not "regulation" in this case, but actual state owned enterprises?

Isn't the obvious objection in this case that the incentive structure at the moment is terrible? Hence the lack of competition in most markets? Without needing to endorse one philosophy or the other, I'd say the telecom system in the US is a failure (for ordinary customers, it's been a ringing success for the telcos), and municipal broadband is worth a try (being less terrible than Comcast & Co is a fairly low bar...).

Yes, what you are describing is basically the UK railway. Network rail (government) subcontract out maintainence and build to the private companies (Alstom etc..) and then lease track access to private companies (Virgin trains etc...). The UK has possibly the most overpriced railway in the world, at least in this instance it fair to say 'it didnt work' unless you want to pay 300usd to get from london to glasgow in economy.

Note that the system was fully private for a while (Railtrack PLC), until they went bankrupt. I think it's going to be expensive whether it's public or private, for complex reasons probably related to "legacy" hardware. If it was fully deregulated then the operators would immediately shut down all the tracks within 100 miles of London to build houses on them.

I believe all European countries are required to have separate track and train operators for railways with some structure for competition in rail. It's just that most of them manage it better than the UK.

The UK telecoms industry was privatised and has ended up with BT Openreach operating most of the wires, leased out with local loop unbundling (LLU) to a variety of broadband providers. It works OK although not brilliantly.

In Japan the railway companies (which are pseudo private in the way a lot of things are in Japan) run the railways effectively at or below cost and the profits are made on property development and leasing on the surrounding land.

In the UK the railway companies try to run profits on railway operation while the taxpayer subsidizes them and the surrounding land is owned by property developers who are making profits that are staggering.

The Conservative government is also doing stuff like donating 38 million pounds to govia thameslink in profit compensation for a strike they triggered.

It's taxpayers and passengers subsidizing property developers' and TOC profits, and the government/TOCs seems to be trying to get the TOC wage bills down so that TOCs and property developers can make even more.

No, the most sucessful train company in Europe, the swiss doesn’t. And those who have go e for separate companies have just bad experience with it. This separation is mostly ideologially driven with no proven benefits. Right wing parties will push through anything their economic text books say is smart regardless of pracrical results and experiences.

Switzerland is part of the Single Market but not part of the EU or EEA, just to complicate matters. I agree with the rest of what you're saying.

"I believe all European countries are required to have separate track and train operators for railways with some structure for competition in rail."

Not really, in some countries the railway and train operators is the same company, strongly controlled by the government and operating without big problems.

It's my understanding that the problems in the UK started when they stopped using that model.

100 times better than the states where local monopolies where allowed to remain and you get insane things like receiver plays for mobile.

And it went bankrupt at the cost of fixing some fundamental problems that contributed to at least two nasty rail accidents. The government decided it didn't want to continue to be responsible for the maintenance. Would rather feed subsidies to a company that would take the heat for issues.

> The UK has possibly the most overpriced railway in the world

The UK railway is still subsidised by the tax payer -- while some routes like Manchester-London make a profit, this profit is reinvested into loss making routes like the Welsh Valleys which help keep communities alive (the operating companies take about 3%, so a £100 ticket without a private company taking money would at cheapest be £97)

There is plenty of competition on the Glasgow-London route, including two Coach companies, Plane - multiple companies and routes, driving, two direct train companies (albeit with similar branding) and a third overnight.

The $240 peak time one way fare comes out at 58 cents per mile. It's valid on any train, just walk up and go.

For almost all trains it's actually $183 return, or 22 cents per mile - valid on any train other than the first arrival into London in the morning and the first couple out of London in the morning.

A train from Washington to New York will set you back $1.14 per mile, with less flexibility.

Im not sure I understand your 3% point, are you saying Virgin pay more into Network Rail on the WCL and this wouldnt happen if government owned/controlled?

Interestingly, I can take a 3.5-4hr train from Penang to Kuala Lumpur for $12, made by Siemens I believe, traveling faster than the Pendolino/Voyagers, and with a similar level of comfort (I do think the seats are bit narrower though), so in comparison I consider the UK crazy priced unless you get some off-peak deal (still not great).

I suspect if still subsidized then it's probably very inefficient, I dont know but I certainly dont feel good paying the train ticket prices in the UK even the off-peak saver.

Nothing to do with the trains themselves, everything to do with the tracks and their maintenance. There's a pie chart in that PDF I linked elsewhere which shows that about 50% of the cost is network rail, 30% staff and 20% train leasing. I suspect the UK could maintain it for a lot less if it could pay Malay wages for the track maintenance and train staff.

UK railways have been chronically mismanaged for years. Better comparison would be against other European countries.

>I suspect the UK could maintain it for a lot less if it could pay Malay wages for the track maintenance and train staff.

They could probably also do it by whacking the extremely wealthy landowners surrounding major train stations with an extra tax.

After all, if the train station made your land valuable, you should be paying for that. In Japan this helps pay for infrastructure. It's practically criminal that in the UK it does not.

Indeed, a land value tax would be very equitable. If society needs to build a power plant near you, which lowers the value of your land, then at least you don't have to pay as much in tax. If they build a new transport hub nearby, which triples the value, then you have to pay more.

Fair points. What would be the benchmark country in your opinion?

I'm hardly an expert, but Germany and Japan are the obvious examples to look for best practices. Really the UK comes off badly compared to most European countries. Worth noting that Abelio (Dutch state-owned) and DB Freight (German state-owned) operate in the UK.

(anyone in the age range eligible for InterRail, I would highly recommend spending a month travelling the railways of Europe)

Germany subsidies its rail network by €17b a year. The UK subsidises by €4b.

In the UK we have decided to go for lower taxes and lower subsidies. We do that while still subsidising lines that are good for the country but not profitable (old mining towns in south wales which would collapse if people couldn't commute to Cardiff etc, cornish rurual lines which attract tourism and offer transport for local residents who may not have 1 car per adult, etc). This means that on the profitable lines which would make commercial sense (say Manchester to London, or some London commuter lines) we have to charge more.

Remember commuter tickets are twice the price of business tickets - as they are paid after income tax (which varies from 33% to 60%). When I was offered a job based in the office, I wanted an extra £10k a year for the commute -- £2200 for the ticket, £3300 for the extra tax to pay for that ticket, and £5k for the time out of my day. They decided that I didn't actually need to be based in an office after all, and I go (to that one) about once every 3 months, which costs them £20 return (but would cost me £45).

This is offset by the massive discounts. A season ticket from Nuneaton to London, 206 miles a day, about 1 hour in each direction, is £9000 a year, for 230+ days of travel, or 47,000 miles. 19p per mile. If you're willing to take a little longer and travel on London Midland it's £6500 per year, 14p per mile.

On the other hand, a day trip from Nuneaton is £154 in peak time, £104 on the slower train. That's 74p per mile - 3.5 times the price of the commuter.

I don't think many people are happy with the way that Abelio has been operating Scotrail.

There's a constant claim from the left that train fares in the UK are overpriced because the train operating companies take a profit of about 3% of the revenue.

There is a profit margin to be made on any savings, I'm sure that if Stagecoach or First or whoever could cut costs enough to make more money they would. We have very old infrastructure, laid down in the mid 19th centuries, and very crowded lines. Building a new line is a mammoth task, and expensive, as we've seen from HS2. Rail staff are hardly cheap too - a virgin trains driver earns £51k, top 10% of earnings in the country.

I pay £11 per day for my 35 min commute into central Edinburgh (I work from home a lot so season tickets don't make much sense) and given that parking in Edinburgh would be £25 a day £11 a day looks like a bargain given that the station I commute from in Fife is 10 mins drive away from where I live and has free parking always available.

That doesn't sound too bad, but I don't think that's representational of most routes and hence most travelers, in that if you stand outside Euston, Glasgow or Edinburgh before 8.30am and ask people what they paid to get to work and if they considered it value for money you would likely get a lot of expletives

However they don't take commuter bus services, or drive, or ride a bike, or move closer to the office, or change job location, or change job times.

The media has driven this culture of people believing in 'rip off Britain'. If people were commuting 100 miles a day for the cost of a cup of coffee they'd still say it was a rip off (although oddly it would be the commute, not the £4 latte)

I suspect I would be much less happy with the service if it wasn't for the fact that I am far enough out from Edinburgh so that the train is relatively empty so I can pretty much always get a seat (and a window seat on the side with the best views).

Two stops in from where I get on the train fills up dramatically and there are often people standing.

free parking what socialist paradise is this - a while back I looked at buying a flat near my local rail station - I reckoned I could have covered all my property taxes or a large chunk of my mortgage from renting out my unused parking space

The Kingdom of Fife!

> the operating companies take about 3%

I've heard this before but I'm not sure I believe it. They have a pretty huge incentive to make it seem like they aren't making a lot of profit, and it is easy for them to do so (just like Hollywood accounting).

Well if you believe that Labour, Coalition and Tory governments have all conspired with all the train companies to somehow hide profits from the public and from their shareholders, that's not really something I can argue with.

I could fly Glasgow-London for far cheaper than the train ticket. Yet an airplane is far more expensive to operate.

I wonder why that is.

No, the railway "permanent way" is far more expensive to operate. That's the problem. The UK railway network has rather a lot of 100+ year old bridges in it, along with some less ancient but still pretty old signalling equipment.

(extremely high detail PDF on costs: http://orr.gov.uk/__data/assets/pdf_file/0019/4933/toc-bench...)

The airplane gets tax-free fuel to burn and doesn’t have to pay for environmental damage it creates.

That (plane operating costs > rail operating costs) may not be the case when you take into account the cost of maintaining the physical infrastructure (hundreds of miles of railroad) compared to that of airports (a few hundred metres of tarmac).

Not to mention that there are only two stopping points (London / Glasgow airports) where as a train needs to stations all over the place.

True. But if Heathrow you'll pay 20gbp (or is it 25gbp now?) on Heathrow Express to get from central London to the airport unless you take the slow tube.

From Heathrow the options are

Heathrow Express - Heathrow Connect - Bus to Hayes+Harlington then train - Picadilly line - Coach - Uber - Taxi

Only one of those options is for mugs, and it's not the Heathrow Express.

However from Glasgow you can also fly to Gatwick, or City if you want to avoid thiefrow.

However comparing the cost of a non-flexible ticket from Glasgow to London with a flexible train ticket is ingenuous at best.

> unless you take the slow tube.

That's competition, no?

Is the slow tube the one that takes an hour?

> A train from Washington to New York will set you back $1.14 per mile, with less flexibility.

The trip is approximately 226 miles. With sufficient advance purchase, the Northeast Regional coach fare is $49 each way and the Acela is $168 each way. I have never seen an NER fare greater than $200 each way. How did you arrive at $1.14 per mile at all, let alone thinking it's typical?

I looked at the next train they were selling a ticket, which was the closest I could come to the flexibility you get on a £182 ($240) London to Glasgow ticket that came close to the "$300" mentioned by the OP.

I've checked again for tomorrow morning, $295. Perhaps I'm using the site wrong: https://imgur.com/TnuOGt3

295/226 is $1.30 per mile.

I'm ignoring the slower trains, non flexibile, booking in advance etc, as that wasn't what was raised. There are slower options for London-Glasgow that are nowhere near the full price (for example £57 tomorrow morning if you go via Edinburgh - saving £130).

I've just checked my emails - last time I took this train (albeit from NY to Washington), in March 2013, it was $498 for 2 tickets - $249 each.

This is for a far less flexible ticket than I get in the UK. For my $240 I can travel London to Birmingham, have the day there, and then continue to Manchester, have a meal and stay in a hotel overnight, then get a train from Manchester to Glasgow the next morning.

Washington to New York is half the distance, and nearly twice the price, as London to Glasgow.

> I've just checked my emails - last time I took this train (albeit from NY to Washington), in March 2013, it was $498 for 2 tickets - $249 each.

That price is far from typical for this service. The prices I quoted are what I just pulled off amtrak.com for a Friday NY-DC train on 12JAN18. A couple months ago I purchased a ticket from Fredericksburg (south of DC) to NY with only a few days in advance for only $130. Thee ten-trip option my sibling poster mentioned gives you the same flexibility you mention there. Please do not base your idea of the cost of rail travel in the NE US based on one very overpriced trip.

Edit:

Just saw your screenshot. That $295 is for the Acela (which is premium service). I just ran the same search, and midday trains are running about $152 for the NER and $168 for the Acela. The one you chose is literally the most expensive non-first class ticket you could book tomorrow.

If I was going at midday I'd get a $80 London to Glasgow ticket. The comparison was as close as possible to the fully flexible any train any time Glasgow/Euston high speed route that sitepodmatt brought up. It's still not as flexible as a UK train - you can't stop off in Phillidelphia for a few hours on your way you New York for example, can you even buy a ticket 5 minutes before the train leaves on that train?

I still don't see his $300 ticket, the only thing I can think of is that he's looking at a return ticket, but then that doubles the price on the Accela.

I was about to chime in similarly; a ten-ride ticket for NYP->POU is $0.38 per mile and you can take any train without reservations; an off-peak MTA ticket for the same leg costs $0.21 per mile. $1.14 per mile sounds high.

Those prices are a bit high for typical UK season tickets. 85 miles Poughkeepsie to NY 90 minutes, similar distance + time and number of trains per day to Salisbury to London (88 miles). A weekly season there (so 10 trips) will set you back £131, or $175, 20c/mile, half the price of Poughkeepsie to New York.

$0.38 / mile is the long-distance rail (Amtrak); the commuter rail line is $0.20 / mile (off-peak 10-ride or 10 rides on a weekly pass anytime).

"a ten-ride ticket for NYP->POU is $0.38 per mile"

I'm only going on what you said.

MTA.info says this ride is 247.50 for a 10 trip ticket, with a trip of 85 miles that would be 247.50 for 850 miles, or 30c/mile.

From what I can see when doing a like-for-like comparison between the UK and the northeast US, the US is surprisingly pricier than the UK.

Now to be fare the Salisbury weekly season is valid for 10 journeys (or even 40), but only over 7 days, where as the Poughkeepsie one is a carnet, you can do 3 days one week, 2 days the next, and that's better, but for the Monday-to-Friday person, I'd rather be paying the Salisbury-London fare than the Poughkeepsie-NYC fare. (I'd rather be going into NYC, just without the fare)

NYP is New York Penn Station, the Amtrak station. Metro North leaves New York out of Grand Central Terminal. Sorry; the phrasing, while unambiguous to someone commuting from the Hudson Valley, was definitely confusing to most everyone else.

Metro North also sells a weekly pass for $166.75. Unlimited rides during a calendar week. The 10-ride ticket has two prices, for peak and off-peak. Peak-hour 10-ride is $247.50, as you said; off-peak is $157.25.

Very few people actually buy a weekly pass on Metro North; either you are commuting infrequently (eg, weekend commutes) in which case you want a 10-ride ticket, or you are commuting daily, in which case you want a monthly ticket. A monthly pass between Poughkeepsie and GCT is $521. If you commute 20 days per calendar month, your per-mile cost is $0.16 / mile.

(Metro North fare information here: http://web.mta.info/mnr/html/planning/faresMar2017/hhfares_m...)

Actually, since there is confusion, going back to the very beginning --

For the train route I myself am familiar with, the line between New York and Poughkeepsie, two different train systems serve the route, Amtrak (the company that does the nation-wide train service in the United States) and Metro North, a commuter-rail line associated with New York's MTA, via their Hudson Line.

I started with the Amtrak costs because the New York <-> DC train is an Amtrak train; most daily commuters will be using the Metro North, as it is cheaper.

Part of the problem is that you can't have "real" competition, e.g. two train companies can't operate competing franchises on the same line, at the same time. I get terrible service and pay high fares on my Southern Railways route, but I can't opt to use a different carrier without moving house. But if my broadband provider decides to raise their prices by 5% next year, I can move to a different provider with a single phone call and a few minutes of downtime.

Most people in the country can not realistically switch internet providers. Even when you have a choice you're often limited to two.

lhopki01's point is that we suspect you are talking about USA in your post, where the GP was talking about the UK, which has a much healthier market for high-speed internet.

To be fair this might not be obvious, if you missed the context of the thread and unless you are familiar with "Southern Railways".

Which country? If you have a phone line to your house you can pick between more than 2.

Under what circumstances do most people only have a choice of two.

Switching to DSL is not a realistic option when most of the country has broadband.

> Switching to DSL is not a realistic option when most of the country has broadband.

DSL is broadband. I've been using DSL for over 17 years (and, while away from home, using combinations of cable & satellite), and I watch plenty of TV & listen to plenty of music on it. It really does work great for me.

Depends on the wire length and quality. A whole lot of DSL connections can't even handle a single 1080p stream, or good quality 720p.

Again which country are you talking about? The majority of UK broadband is ADSL2+. There is cable in some areas but not all. In all areas with phone lines there are multiple providers giving around 20Mb/s. They all use the same phone lines because of local loop unbundling. In areas with fibre to the cabinet there are again multiple providers all using the same fibre. The only service in the UK that doesn't have local loop unbundling is cable.

"In all areas with phone lines there are multiple providers giving around 20Mb/s."

I don't know about the rest of the UK but I don't think that is the case in Scotland - once you are any distance from a city or town the speeds can drop dramatically and that's not even in the Highlands or Islands. Certainly when we were looking at houses about a year ago there were quite upmarket new houses in Perthshire that had no broadband at all.

Are there non-subsidized railways that break even, what's their quality?

The problem with rails is the same with roads.

Yes, AnCaps/Libertarians/Voluntarists/goldbugs think that "yeah, if the State weren't meddling in the market, people would form a corporation to build a [toll] road/rail", but of course, the same network effects would lead to very fine roads between major population centers and abso-fucking-lutely nothing else, because why would anyone connect a mining town in Wyoming anywhere? The mining company will use big all-terrain vehicles on dirt roads and that's it. Maybe they'll lay a track for the coal itself.

And I'd even argue, that yes, even if rail operates at a net loss according to GAAP, that doesn't mean socially it is a negative. The same applies to broadband.

However, maybe ... maaaybe not everyone should live so far from each other, so infrastructure TCO and commute times and energy use (and GHG emissions) would be lower.

Have a look at Thailand for the effects of way-less effective regulation.

Street furniture and cables all over the pedestrian areas and cars and bikes everywhere else. Huge concrete overpasses, litter and pollution.

In the railway context I believe SRT is government operated including track and trains so they follow a mandate rather then regulation. I took a new train from BKK to Hua Hin a few months ago, quite a pleasant ride to be fair. The cables are a problem, but I suspect it led to faster broadband and FFTH rollout, to give an example I was one of the first True customers and they ran a coaxial cable about 0.25km to the house, it took less than two hours, you couldnt do that if it was underground. Sometimes these things work.

Government maintained track system that lease track access to private companies is problematic system because of several points where that relationship breaks.

1: Rail maintenance dictate quality, but is shared between all competitors.

2: Rail is located on ground owned by the government. There is no practical or legal option to operate a private rail network.

3: Lack of distinguishable features to compete, high barrier to entry, and a limited number of companies create an environment thats prime for oligopolies. High price, low service.

In contrast to Internet service providers which:

1: Quality of service is generally not dictated by the back bone network.

2: Government tend to grant permission and funding to the first ISP that want to dig into government land to plant cables, but then deny completion afterwards and allow natural monopolies of those that were first.

3: Low barrier to entry could allow for many ISP.

> 1: Quality of service is generally not dictated by the back bone network.

If by that you mean that it is not the only thing that matters, then yes. But if the backbone is bottlenecked then of course quality will suffer.

> 2: Government tend to grant permission and funding to the first ISP that want to dig into government land to plant cables, but then deny completion afterwards and allow natural monopolies of those that were first.

Is that actually the case? Getting a dedicated line between any two buildings tends to be no problem, so why would doing the same on a larger scale be a problem all of a sudden?

> 3: Low barrier to entry could allow for many ISP.

That's only really possible if the last mile can be leased from either the municipal or some other company (sort of like the government maintained track system). Otherwise the barrier is exactly this last mile, which is the most expensive part of the whole thing (which is also why a lot of small rural ISPs use wireless as last mile access instead of wired connections).

Virgin Trains at least are nice and if you book well in advance very reasonable deals are to be had. We paid ~42 quid from London to Manchester in first class, one way. Alas we bought the tickets month' in advance and, of course, non-refundable.

For shits and giggles I tried to figure out what the same ride costs like when you buy the ticket 30 minutes in advance. And there we come to 280£ each, which is completely ridiculous. It's not that good (2nd ist still ~90quid)

Now, don't get me started on Northern. Filthy, Expensive, Clueless and slathered with a heavy dose of disorganisation.

I think you can honestly say that privatizing trains in the UK was an experiment going horribly wrong.

> Do people really believe that infrastructure is not the purview of the government?

Sure. Lots of people do. Japan’s railroads, subways, and many highways, are private. Water utilities in France are mostly private, as are most of its highways. Almost all electrical systems in the US and Europe were privatized in the 1980s and 1990s.

Would public internet infrastructure have been better in the US? Our internet speeds in the US are faster than our big EU competitors: UK, Germany, France, Spain, and Italy. That’s pretty much the only piece of infrastructure you can say that for. Our subways aren’t better. Our roads aren’t better. Our regional trains aren’t better. But our internet service is better.

It’s not a philosophical question, but an economic one. How do you ensure adequate spending on very expensive infrastructure. Japan does it for rail by making rail companies private and allowing them to use land grants to enter retail leasing markets. (Japanese railways own lots of the land around stations and are the landlord for the associated retail establishments.) Many European countries have the political discipline to ensure adequate funding for public infrastructure. Both models can work.

But we already know that in the US we don’t have the political discipline to properly fund infrastructure. Infrastructure gets attacked from the right, who don’t want to pay taxes. But it also from the left, who want to use infrastructure as leverage for social issues. For example, in Stockholm, the publicly owned (but completely privately funded) dark fiber utility took 16-17 years to cover 90% of the city. There was no mandate to build in historically disadvantaged neighborhoods or anything like that—build out was based on demand. That approach is completely illegal in most US cities.

Do you have a source on the US vs. the rest Internet speeds? That's not my impression from end user experience in those countries.

See: https://www.akamai.com/fr/fr/multimedia/documents/state-of-t....

To use a concrete example, 500/200 fiber through Orange in France is about $63/month after the one-year promo period: https://boutique.orange.fr/internet/offres-fibre/jet. The closest FiOS plan is ~$80/month after the promo period for symmetric gigabit. So we're somewhat more expensive, for somewhat faster speeds. Affordability-wise, it's about the same, considering the much higher household income in the U.S.

But if you visited say Baltimore and Paris, you wouldn't realize that. "France has fiber and Maryland doesn't," you'd conclude. Population distribution in the two places is very different: most people in the U.S., especially the folks who can afford higher tiers of internet service, live outside the core cities. While Baltimore has crappy internet choices, most of Maryland lives outside Baltimore, and has fiber. Indeed, Maryland is in the top-5, just behind Sweden and way ahead of France in Akamai's tests.

Our max may be higher but I too would like to see that our average is higher because I rarely get the speeds I pay for.

First I'm not sure of the yardstick you are using to measure. There is still a not insignificant minority with no internet access at all in the US and further lots of people are stuck with truly mediocre dsl at speeds we no longer consider broadband.

Lets start with Germany.

Internet (60 Mbps or More, Unlimited Data, Cable/ADSL) 25.92 € or 30.59 usd

https://www.numbeo.com/cost-of-living/country_result.jsp?cou...

United States

Internet (60 Mbps or More, Unlimited Data, Cable/ADSL) 56.60 $

Further in the US only 80% of households had a broadband connection at home or access to a smartphone with data.

In Germany its 98.

In summery the internet is almost twice as expensive in the US and lots of people don't even have access to it.

We could go through all the countries you listed and see if we could find even one where your assertion is correct but I feel like that would be a waste of my and your time because its not clear that being able to find even one nation in all of Europe worse than the US would prove that private competition has served the US well. It hasn't.

See Akamai’s Q1 2017 numbers, where the US breaks into the top 10 for average connection speeds: https://www.fastmetrics.com/internet-connection-speed-by-cou....

The whole report is here: https://www.akamai.com/fr/fr/multimedia/documents/state-of-t.... Other interesting statistics: 48% of American connections are above 15 mbps; in Germany its 33%. Maryland, which has almost exactly the same population density as Germany, is at 60%.

I don't think measuring the average speed is even remotely a meaningful measure.

Imagine

scenario 1

3 users with gigabit 4 users have dialup 3 users have nothing

Scenario 2:

4 users with gigabit 6 users with nothing

Scenario 3: 10 users have 50 Mbps

In scenario 1 the average is 300 even though only 30% of people have real internet access and 70% either can barely send email or have nothing at all. Scenario 2 is even better at 400 even though most people have nothing.

Another way to look at it is that one person with gigabit is as good as numerically as 20 people with real internet or 100 people with basic internet.

The whole networking infrastructure is privatized as well though in Germany (since 1995) - so it doesn't support the argument that private competition is somehow better, since both have it.

While in general I agree that competition is great - I don't think that applies to the last mile (from the next PoP to the customer).

More socialist leaning countries privatize infrastructure. For example Ontario power generation.

Is there conclusive evidence that privatization is worse?

It seems like an age old question. That there are positives and negatives. It's also politically expedient as it raises money.

Big difference between privatising the parts of the system that are a natural monopoly and those that aren't.

Privatising the power cables going to people's homes is a dumb move - you won't get the benefits of the free market, as there's no competition between suppliers.

But once the cables are in the ground, competition between multiple power suppliers might be possible - so long as customers can compare prices and switch suppliers easily; and the power plants aren't incentivised to cut long-term investments and imperil the country's energy security.

This guy gets it. Privatisation on its own does not magically improve services and cut costs. Competition can do so. But privatisation without competition is just a monopoly license to print money.

There's rarely any political push to privatize things that aren't natural monopolies though, precisely because nobody is going to make bank on that.

In practice almost every privatization is simply an exchange where the new owner pays the government $x to help solve its "debt problem" (because the government may one day be rendered insolvent by running out of the cash it it prints).

Then, the public gets soaked for $nx where n is usually at least > 2 and the government officials behind the privatization get some really lucrative jobs and speaking engagements after they leave office.

> It's also politically expedient as it raises money.

You don't get that money for nothing. You're selling part of the nation's infrastructure, that was paid for by taxpayers, to now be exploited for profit. The taxpayers end up paying twice.

It's better to keep the essential infrastructure nationalised, and allow private parties to use it.

> You're selling part of the nation's infrastructure, that was paid for by taxpayers, to now be exploited for profit

Maybe. If it's taxpayer funded infrastructure, it doesn't perpetuate itself by self determination. It must be maintained with more tax dollars.

Assuming the cost is equal whether it's privatized vs. state-owned, then it all boils down to a matter of ideological preference. Some object to others making a profit off of a public good. There may or may not be merit to those arguments, but they're completely orthogonal to whether or not there are costs associated with their maintenance, because there always are.

The taxpayers always end up paying twice.

If costs are equal, the private for-profit entity is more expensive, because it wants to make profit. If it bought the infrastructure, it wants to recoup that investment.

Apologies. I intended "costs" to mean "costs to consumer".

Sure, if the costs to the consumer is equal, it doesn't matter much. But I think that's unlikely. In a privatised versus nationalised comparison, I think the costs to the consumer are either going to be higher if it's an area where competition doesn't work efficiently (monopolies, duopolies, cartels), or lower if it's an area where competition does work efficiently (a transparent market with many suppliers).

> It's also politically expedient as it raises money.

Raises is a very positive sounding word.

Would it be better to couch that sentiment in terms of who's getting shafted (it extracts money from) or who's getting rich (it redirects wealth to)?

At the local level, I could deal with private ownership. But if and only if there were strict regulations in place and an oversight body. But, at that point, would it really save local residents money anyway?

The closer that the government is to me, I more that I can control it (albeit with much frustration sometimes). But, private ownership, I can't control.

> Is there a good argument against this?

I think one of the main issues is having a company that has spent millions, sometimes billions of dollars building out their infrastructure. Then you have the government come in, build out a competing infrastructure with tax payers money.

As a business, you're now having to compete with an entity that has unlimited budget resources, can build out wherever they want, grant contracts and favorable leasing options to whichever businesses they want to and quite easily put you out of business.

When you have to compete with a government entity as a business, you will always be at a disadvantage. In essence, governments don't foster competition, they basically do away with it, pick the winners they want to and then monopolize it and push other companies out.

Would your solution be better for most people on average? Probably.

Would anyone have unlimited gigabit internet access for $70 a month (like I do)? Probably not.

So yeah, I realize the hypocrisy of being willing to pay taxes to help the less fortunate but not being willing to give up my precious gigabit service to help the less connected.

>Who doesn't think the US Highway System has been great for our economy?

When one considers the hidden costs of increased urban sprawl, reliance on automobiles, and the associated increase in pollution, it isn't so clear. If roads were privatized then the true costs would be more clear.

Are you arguing that privatization of an industry is likely to result in better management of its externalities?

> Is there a good argument against this?

Yes. The infrastructure in and around NYC. All the bridges need to be replaced (you can literally see the water below through them), the subway system is in tatters, Penn station is a mess of train derailments and constant delays, and the whole thing is unbelievably expensive.

It's a textbook case of short-sightedness, mismanagement, and rent seeking from private parties.

A govt internet would work ok for a few years until it was driven into the ground by lack of innovation and special interests. With this model we'd all still be connecting to the internet via 56k modems.

And because it's a government enforced monopoly, we'd have no choice.

Municipal broadband exists because the lack of competition in the private sector has lead to "lack of innovation" and "special interests". The muni broadband projects fill in these gaps since multi-hundred billion dollar companies can't seem to roll out service at greater than 10MBps. Muni broadband is competition to a monopoly, not a monopoly itself.

I downvoted you because you relied on an analogy to train and subway infrastructure instead of looking at existing municipal broadband plans that have existed for more than "a few years."

For example, the municipal broadband in Wilson, NC, was grandfathered in because it existed before the NC state legislature wrote a law to prevent municipal broadband. It's existed for around 10 years. Rather than being stagnant, that plan delivers some of the fastest internet in the country with reportedly easy setup and decent pricing.

If you had made your argument citing municipal broadband plans in the U.S. which give worse-than-average speeds and pricing, I wouldn't have downvoted you.

Here's a more complete, if a bit dated, analysis. https://www.law.upenn.edu/live/files/6611-report-municipal-f...

I would say that Greenlight in Wilson is still fast because of the relatively recent infrastructure build out. The question of whether they are stagnant are not won't be answered until the next build out is required.

The project was financed by bonds with the infrastructure as collateral. Since Wilson's bond has a balloon payment due in 2033, any new infrastructure would probably require rolling over that debt in addition to the new infrastructure costs. Given the low positive cash flow, it remains to be seen whether they will be able to keep up with future infrastructure demands, whether driven by streaming VR or other services. They are also deriving significant revenue from cable television and landline phone services, both of which are potentially negative growth service areas over the next 15 years.

The peculiar advantage of municipal broadband is that municipalities are usually not rural, they can take advantage of the higher density installation locations which most providers want. Many of the monopoly contracts issued by various localities to providers came with the caveat that everyone in the county needed to get access to the service. To be fair, the higher costs for more rural installations should translate into higher prices for rural customers, but that would reduce demand.

So, it is certainly worth trying more of these muni-broadband projects, as this obviously is not yet a solved problem. My only plea would be that we don't give anyone a monopoly on putting wires in the ground, whether a private company or a government entity, whoever wants to safely lay cable should be given a permit.

Au contraire, we'd be connecting to the internet with _extremely_ expensive equipment. See https://arstechnica.com/tech-policy/2013/02/why-a-one-room-w...

The issue there was the top-down approach, which is almost never efficient.

I'm suggesting a bottom-up approach. Municipalities themselves paying for what they need. Then states paying for what they need. Then federal.

Isn't MTA semi-private???

I lived in chicago, they did some maintenance work on the street in front of my house but improperly installed the manhole cover. Now evertime a car passed over the manhole cover it made a loud thud. This is moderately busy street and there would be lound noise all day long, I was jolted out of sleep several times because of this.

I called 311 who said they will look it into it. I called them again couple of weeks after that, some person rudely asked me to call local alderman instead. I called alderman who said they noted down the complaint, called them many times in the following months. After a while a they said its comed( electric company) fault and I should call them. I called comed, they asked me to call local city office. I took time off from work and went upto local city office and they said I should call 311 for complaints of this nature.

Sorry, I don't want govt anything. If something goes wrong you are at mercy of some govt employees who have no motivation or sympathy towards your problems.

People who know govt officials like alderman ect get things done but everyone else is out of luck, just roll the dice and hope it fixed.

Edit: why the downvotes? I think this is a on topic response to parents comment about how well govt maintains roads. You should come to chicago and look at the roads in working class neighborhoods. People are ok with poorer people getting horrible internet cos they can't raise enough taxes to fix their internet infrastructure. Its horrible, if you take CTA in chicago underground in winter, you'll see drops of water ( is it water?) drip on your head from leaky ceilings.

People run into the same wall dealing with support for any company.

I get good customer service from comcast, amazon, costco ect.

Are you seriously saying thats typical of any company and that you have to personally know some exec in that company to get things done?

> I get good customer service from comcast

There's an entire subreddit out there who would like to hear from you on this subject...

sure comcast might suck but to say you have to deal with this at "any company" is not a valid response.

If amazon ever did something like this to me, I would never shop there any longer. Problem solved. Now who I turn to if govt refuses to listen to me. Interacting with Govt is a horrible experience almost every single time. You just have to "deal with it" if you don't get help.

Your internet experience will depend on how much your neighbors think internet is the priority given limited funds.

Most people in the US are stuck with only one choice of broadband Internet providers, and those who are lucky might have two options[1].

The difference between Comcast and Amazon/Costco is that when Amazon provides sub par customer service you have the option to shop elsewhere. The reason Comcast can be so well known for providing terrible support is because there is no alternative, your only other option is to choose a dial-up provider and sacrifice a consistent Internet connection.

1. https://arstechnica.com/information-technology/2016/08/us-br...

I am not arguing that comcast is good.

I am disagreeing with "Govt monochoice is better than private monochoice" . Govt can be equally brutal and inefficient and often is. Whats a govt employee's motivating to fix your internet in a timely manner?

Also I have a tangential question, whats stopping comcast from charging $500/month for internet?