[pandemicsoul]

The absurd part of this is that the solution is so clear and easy in the United States: Use the scale of the federal government.

We got to where we are, in the U.S., because we've got massive inefficiency in state and local government. There's only a handful of cities that are massive enough to do projects that would require hiring highly paid specialists to manage these huge projects. Similarly, most states have very few opportunities for long-term career-building public project management. (There's like 3 west of the Mississippi.)

But consolidating all of this expertise in a federal agency tasked with planning and executing massive public projects would allow us to hire career-oriented planners and keep them busy for ever. It's how we run the military, of course. And, ironically, the military is a leech on the U.S. taxpayer, massively bloated and throwing money at foreign problems in the most ham-fisted way just to justify itself – but if you shifted even 20% of the military's spending to a domestic public works department, all that money would not only be to the benefit of the U.S. taxpayer, but you could hire tens of thousands more American workers to do the jobs. It's a win-win for everyone. And it's probably something a lot of legislators could get behind.

Not only that, but a federal public works department would quickly be able to aggregate learnings on improving efficiency – knowing what works and what doesn't – while also being able to wield the power of the federal government to negotiate lower prices.

[nine_k]

How many are the examples when the federal government was more efficient (as in effect per unit money) than a commercial company? Certainly there are some. How many examples are there when it was the other way around?

Asking unironically, because in a monopsony situation when there's only one buyer, a state or federal agency, commercial companies compete in materially different ways than in a free(r) market.

[PuppyTailWags]

Veteran's Affairs, the postal service, social security benefit enrollment and payment processing, the irs, and public education is generally more efficient than commercial offerings at multiple points in their historical existence. Many of them are currently hampered by underfunding though, so I don't know how efficient they are now (being unable to invest enough to maintain themselves would force them to make long-term inefficient tradeoffs).

[nine_k]

This is a good list. But doesn't "underfunded" mean "the society refuses to pay the whole price of the service"? That is, it's sort of more expensive, but is kept both starving and alive by fiat?

Regarding public transportation, for instance: New York's MTA receives about half of its financing as a subsidy. London Tube is a statutory corporation, and can cover only 90% of its expenses from fares. Tokyo Metro is a private entity that shows a healthy profit, and Moscow metro also used to be profitable for a long time. Both Moscow and Tokyo underground infrastructure and service at least are not inferior to NYC's and London's (from my personal experience with 3 of the 4 of them).

[mrguyorama]

No, underfunded means republicans (and occasionally democrats) have hampered these groups for decades, under the guise that "government can't do anything", and then point to the kneecapped government actions as reason to believe the government can't do anything.

If you disallow the government to pay its bills, or effectively haggle, or build the kind of teams it needs, of course it will do a bad job. Meanwhile lots of countries have shown you can have in house teams in the government who make great things.

People will say "the government can't do anything" while cashing their social security check, after driving on the federal highway, drinking their usually clean water, and yelling at their kid's teachers for stupid reasons. We have had 60 years of starve the beast, this was all intentional.

[Firadeoclus]

There is no reason public transport should aim to be financed entirely by fares, or to make a profit from fares. It benefits people who rarely or never use it.

[verall]

It is not always comparable between cultures. If the Tokyo Metro was found to be blatantly profiteering, the president of the company would probably resign due to the cultural shame. In the US, it would be, "welp, that's the governments fault..."

[PuppyTailWags]

I didn't include public transportation in the list so idk.

[jimmaswell]

Aren't VA hospitals hopelessly corrupt with such a low standard of care that any veteran avoids them like the plague if they're able?

[tnel77]

Corrupt? I haven’t heard about widespread corruption at VA hospitals. Do you have a source for this claim?

[PuppyTailWags]

Please read my entire post.

[Firmwarrior]

VA hospitals were like that 50 years ago too though

[PuppyTailWags]

Please read my entire post. I never cited a specific time of efficiency. I only said that there exists a point in time that a service was more efficient than equivalent, private sector services. If you want to disagree you will have to prove that in the entirety of the VA's existence it has never been more efficient in any of its services (they cover housing too) than private sector.

[pessimizer]

> How many are the examples when the federal government was more efficient (as in effect per unit money) than a commercial company?

The real question I think is whether there are examples where larger companies gain efficiencies, and if the most efficient way to do something is at a national or international scale, do you want to depend on a massive private companies (likely heavily subsidized) to do it?

That's not handing a public function to the private sector for the sake of finding efficiencies or to fund emergent innovation through competition, that's handing risk-free concessions to a cronies.

[m000]

> How many are the examples when the federal government was more efficient (as in effect per unit money) than a commercial company?

Here is the thing: commercial companies have incentive to stay efficient as long as there is viable competition from a public agency, which provides a baseline in terms of cost and quality.

As soon as public agencies are out of the picture, then there is little incentive for private companies in staying lean and efficient: Why deliver what you promised when you can go over-budget? The threat that an expensive project will be left unfinished gives you leverage to ask and approve excess costs. Why compete with other companies when you can collude with them and laugh together on your way to the bank?

[lumost]

If the government is a monopsony for a good like “subways” or “highways”. It’s advisable to have an in-house alternative to the commercial sector. Commercial firms are obligated by incentive to increase profits over time - whereas in-house teams are obligated to increase organizational size/pay over time.

Competition between the two should lead to an equilibrium between organizational bloat and profit taking.

[yummypaint]

Read up on the design and construction of the pentagon and los alamos national laboratory. The relationship between the US federal government and private sector changed tremendously over the 20th century.

[pirate787]

Amtrak is a disaster, both inefficient and undercapitalized and dangerous. The U.S. Congress nationalized passenger rail in 1971, gave Amtrak the extraordinarily Northeast Corridor backbone, and yet America has easily the worst passenger rail in the developed world. Amtrak's cost per passenger mile is 3x or 4x higher than flying with a worse safety record.

[stetrain]

We nationalized the passenger services that were losing money, that’s why the railroads wanted to get rid of them.

Most of Amtrak’s service runs on track they don’t own, at the mercy of the freight railroads for scheduling and track maintenance.

They are pressured to keep their long distance trains that lose money as a public good, but they aren’t given consistent public funding in return. We treat them as a company that should turn it own profit. We don’t expect the DOT to turn a profit on highways, because we recognize the greater economic impact of having useful highways.

The Northeast Corridor is the exception and notably seems to be the best performing part of the Amtrak network.

Seems to me like the standard American public-private practice of privatizing the gains and socializing the losses, and then complaining about how the government loses money.

[jcranmer]

> The Northeast Corridor is the exception and notably seems to be the best performing part of the Amtrak network.

And, interestingly, the worst part of the Northeast Corridor is the part that's not owned by Amtrak (the Metro North territory between NYC and New Haven).

[kwhitefoot]

> the standard American public-private practice of privatizing the gains and socializing the losses

It isn't only the US that suffers from this. The UK has a bad case of it and even Norway is in danger of doing it.

[olivierduval]

Same for France sadly... but we always take US worst ideas ;-)

[cratermoon]

That's because there's a poison pill in the Amtrak world. With few exceptions, Amtrak doesn't own any of the rail network they use. They run on rails owned by BNSF, Norfolk-Southern, or Union Pacific.

It was on HN just two months ago. The One Tiny Law That Keeps Amtrak Terrible: https://news.ycombinator.com/item?id=34336653

[ghaff]

Amtrak trains could run on precise schedules--while probably slowing down cargo--and long routes would still be underutilized because most people don't want a more expensive days long train trip rather than a few hour flight.

[cratermoon]

Yes but properly speaking Amtrak should have cross-country high speed rail routes and lower costs than flying. Amtrak can't even imagine high speed rail as long as they are stuck renting routes where they are limited to 140km/h, less than half the speed of high speed rail in other countries. Even Amtrak's fastest lines, the Acela in the NE corridor, only manage 240 km/h along one or two sections.

[vikingerik]

Framing the comparison as cost-per-passenger-mile is unfair to Amtrak, though. Amtrak on the NE corridor enables many trips between nearby cities in ranges of tens of miles that are too short for an airline to serve. Using passenger-miles as the denominator misframes Amtrak's strength as a downside. Amtrak is serving a very useful service there in displacing those trips from cars or buses.

It's true that Amtrak isn't all that useful beyond the NE corridor, although that's as much a function of the scale of the US as of Amtrak itself.

[louky]

Something like the https://en.wikipedia.org/wiki/Works_Progress_Administration

[analog31]

I'm thinking health care and education.

[gersh]

We have the army corps of engineers, which maintains a lot of the dams. We could expand from there.

[larrik]

I was thinking of them, too. We could utilize the military directly rather than siphoning off their funding.

[scythe]

>There's only a handful of cities that are massive enough to do projects that would require hiring highly paid specialists to manage these huge projects.

If you use the Census Bureau's broadest definition of economically intradependent regions, the Combined Statistical Areas — defined roughly by commuting patterns, which is the relevant scale for a transit system — there are in fact eleven major cities at populations of seven million or higher: New York, Los Angeles, Chicago, Baltimore-Washington, Dallas, San Francisco, Boston-Providence, Houston, Atlanta, Miami and Philadelphia-Wilmington. Over a third of Americans live in one of these "agglomerations". But establishing a system to handle regional planning would require crossing state boundaries in six of those cases.

There are some advantages to local expertise: the geographic and social factors affecting Miami are very different from those in Dallas, which are both very different from San Francisco, which are all totally unlike New York. Establishing the necessary agencies and securing their purview would require federal action, though, I agree.

[WorldMaker]

> But establishing a system to handle regional planning would require crossing state boundaries in six of those cases.

6 of 11 is more than half of your examples. That alone seems to suggest there would be efficiencies in federally-supervised planning simply to cross-cut inter-state concerns. (Such as with roads, the use of "inter-state" there is not an accident.)

Scrolling further through the list of CSAs the ones between a million people and seven million that are multi-state seem too often to be the least well served by public transportation access today compared to their peers, and that doesn't seem to be a coincidence.

[whiddershins]

What was the size of the organizations that actually made the New York Subways, on time, and on budget?

What size organization made the Verrazano and the Golden Gate bridges?

What size organization made the Hoover Dam and the Pacific coast Route 1? Using tiny (compared to the equipment we have now) steam shovels?

[pandemicsoul]

All of those projects were built nearly 100 or more years ago. Public works can't be built in the same way anymore because laws are different and there's vastly more complexity in property ownership.

[abduhl]

Tunneling and underground work is a uniquely localized specialty. A national organization that runs underground work is going to still need specialized knowledge for each region both because of local ground conditions and local contracting practices. It is very difficult to export know-how from one area to another. Site sampling/testing practices are not even consistent across regions.

Beyond that, as someone with extensive experience with underground projects and the federal government, the federal government doesn’t pay nearly enough and its benefits lag private industry significantly at this point.

It’s also worth noting that this agency already exists somewhat but spread across departments (FHWA, EPA, etc.). They still outsource to consultants.

[bluGill]

Out sourcing to consultants is not the problem. The problem is that we don't have the ability to manage those consultants. We are also outsourcing management of consultants, including gathering requirements. At no point does anyone get cost control as part of their goals, and in many places they get instructions that are contrary to good cost control.

[abduhl]

You honestly think that the federal government would be more effective at managing consultants? The majority of additional costs in any underground job fall into one of two buckets: third party disputes or differing site conditions. A national agency will do no better than a local agency at managing these and will probably be worse because they won’t have any relationship with the locals. At the end of the day an infrastructure project has to be built locally, not in Washington.

I won’t even touch the fact that most infrastructure jobs are only partially funded by the federal government with the remainder funded by local tax or rate payers. The federal government has no interest in cost controls beyond their own money and the residents of State X have no interest in funding more of the infrastructure of State Y than they already do.

[bluGill]

I didn't say anything about federal control. Others have, but i'm not seeing anyone interested in solving the problem at any level

[ketzo]

Yeah, but that's technical knowledge, not project-management knowledge.

The article argues that the real gap is not in people who know how to build things, but in dedicated civil servants who know how to manage those people.

And that is a much more transferrable skill than "how to build this kind of rail tunnel through that kind of soil composition" or whatever.

[abduhl]

Just saying “that is a much more transferable skill” does not make it so. Effective management on any infrastructure job is, again, highly dependent on relationships with local agencies and the local population.

[xxpor]

>Tunneling and underground work is a uniquely localized specialty. A national organization that runs underground work is going to still need specialized knowledge for each region both because of local ground conditions and local contracting practices. It is very difficult to export know-how from one area to another. Site sampling/testing practices are not even consistent across regions.

Read the report the article is about. Milan fills this role for the rest of Italy and somehow figures it out just fine.

[abduhl]

Italy is 3/4 the size of California.

Does Milan run all of the EU’s jobs too?

[xxpor]

Irrelevant to the point. If the south of Italy is different geologically, the size of Italy vs anything else doesn't matter.

[abduhl]

Local contracting practices are only tangentially related to geology. They are very closely related to geography - hence the word "local."

[xxpor]

The whole point is this stuff isn't contracted out.

[starkd]

Interesting idea. Something needs to be done, because right now, the US can't even construct a nuclear power plant in Georgia on time and within reasonable budget range.

[vorpalhex]

FEMA can't even show up to emergencies in a timely manner, or produce functional online training courses. Why do you think more federal agencies are an answer?

Draining the US military of 1/5 of it's budget while Russia is invading it's neighbors and China is more aggressive with Taiwan is probably unwise.

Then bringing in federal assets to run an extremely local project (forcing those people to also travel for extended periods of time), with no jurisdiction (fed can't overrule state laws on planning typically) sounds like a disaster.

The government also has a very spotty track record with construction projects.

[tehjoker]

An interesting question is why you think the military is good when domestic services are bad. Could it be that domestic services are intentionally strangled, the military is used to pry open new markets from weaker countries, and private business benefits and fails to share the bounty with neither (poor) Americans nor with the subject foreigners?

[vorpalhex]

We spend more, separately, on health, social security and income security than the military. Military spending is tied with medicare.

https://fiscaldata.treasury.gov/americas-finance-guide/feder...

[alistairSH]

US military spending exceeds the next 10 nations COMBINED. Of those 10, half are allies (UK, France, Germany, SK, Japan). Surely, we could cut 20%, still outspend our rivals (China, Russia, India) by >30%, and be ok?

[clcaev]

The goal since WW2 has been dominance, not just in Americas, but also in East Asia, Middle East, etc. China, on the other hand, has had the luxury to concentrate its efforts more strategically. To reduce military expenditure, the US would have to dial back the breadth and depth of reach. Giving up on Taiwan, and then letting Korea and Japan fend for themselves against China would probably also be needed if it were to cut back on its ambitions. Nuclear proliferation would be an immediate outcome.

[rtkwe]

The US could cut back and still compete in the West Pacific/South China see but it would have to focus there very heavily to the detriment of the existing implied obligations to NATO. Part of the US's politically dominant position post WW2 is that we've signed ourselves up to be anywhere in the world fighting a large scale conflict inside of a month with huge stockpiles of everything you need to fight all over the world, including one floating one that can backup the standard land based prepositioned stocks. [0] Granted a lot of that was also enabled by being the only major economy not significantly bombed through the war but that advantage was heavily invested in since then.

[0] Wendover Productions has a great video about the insane amount of logistics dedicated to the goal of never having to wait for materiel practically anywhere in the world. https://www.youtube.com/watch?v=iIpPuJ_r8Xg

[mrguyorama]

Coincidentally, this whole Russia Ukraine war has pushed most NATO nations to build up their military and re-arm. If Europe can actually defend itself, the US's obligation there reduces significantly.

[rtkwe]

Yeah and it's much cheaper to operate locally and defensively than it is to have a very expeditionary and overseas force the US has so they /could/ close the capability gap for less money. There is a bit of difficulty having so many different militaries needing to work together but it's something that works with practice.

[credit_guy]

The issue is not that the US needs to be stronger than China.

The issue is that the US population has a very low tolerance for combat losses (especially losses defending a foreign country), and the Chinese population (read Communist Party) has a much higher one (especially losses towards what they perceive as country reunification). If in a fight the US inflicts one million Chinese casualties, while incurring one hundred thousand, then the US might call it quits.

The US need not only be able to defeat China, but do it with minimal losses. That's why you need to maintain the huge capabilities gap that currently exists.

[petsfed]

>Then bringing in federal assets to run an extremely local project (forcing those people to also travel for extended periods of time), with no jurisdiction (fed can't overrule state laws on planning typically) sounds like a disaster.

I won't speak to whether or not its wise to reduce military funding, but your statement about why you can't have non-locals working on local projects runs counter to my experience.

From about 2004 to 2014, the state of Oregon spent somewhere somewhere around $1.3 billion to repair hundreds of bridges. The consultant managing the effort assembled a team of dozens of highway and bridge engineers, who worked closely with Oregon Department of Transportation, the NTSB and the Federal Highway Administration to make sure it all complied. For the record, getting a state to sign off on bending state regulations is typically a lot easier than getting the Fed to sign off on bending Fed regulations, to the extant that if a state has less stringent road requirements, those state requirements only apply on roads not covered by the FHWA.

At the end of the project, most of those consulting engineers then got flown to other states (Florida and Pennsylvania come to mind) to work on projects of similar sizes. Does a transplant from Oregon have more knowledge of local concerns than a native Floridian and Pennsylvanian? Of course not. But it turns out that you can hire local for negotiating with contractors, but you cannot just hire local when it comes to managing billion-dollar construction projects, if those don't happen very often locally.

The only difference between a federal organization to manage projects of this scale and a private consulting firm is that the bill to the taxpayers on the latter is an order of magnitude larger.

[starkd]

Not to mention the existing debt problem. Just the interest alone on the debt is beginning to exceed the amount spent on the military alone. Something needs to be done, and sooner or later, the usual throwing more money at the problem is not going to work.