[elvinyung]

I also like this analysis of general high infrastructure costs in America: https://pedestrianobservations.com/2019/03/03/why-american-c...

This tidbit is particularly infuriating:

> In California, the problem is, in two words, Tutor-Perini. This contractor underbids and then does shoddy work requiring change orders, litigated to the maximum. Ron Tutor’s dishonesty is well-known and goes back decades: in 1992 Los Angeles’s then-mayor Tom Bradley called him the change order king. And yet, he keeps getting contracts, all of which have large cost overruns, going over the amount the state or city would have paid had it awarded the contract to the second lowest bidder. In San Francisco, cost overrun battles involving Tutor-Perini led to a 40% cost overrun. This process repeated for high-speed rail: Tutor submitted lowest but technically worst bid, got the contract as price was weighted too high, and then demanded expensive changes. It speaks to California’s poor oversight of contractors that Tutor remains a contractor in good standing and has not been prosecuted for fraud.

Edit: oh, wait, just realized this is from the same blog, so the same body of work.

[kazinator]

Ah, but in what way is the under-bidding contractor the problem?

Who gave the contract to the contractor?

The problem is the process of going for the lowest bidder, or one of the lowest.

Moreover, in this case, incredibly, going for the same low bidder with the knowledge of all the history of the bids from that contractor being unrealistic lowballs and requiring costly change requests.

Don't blame the contractor. They get the job and make their money. From their angle, they are successful. They know that the city is aiming for the bottom and so they adjust their bidding accordingly. If they didn't submit a low bid, the job would go to someone else.

True story: some decades ago. My father was bidding on a contract with the GVRD (Greater Vancouver Regional District). Something in the tens of thousands of dollars, probably. He was out-bid by $5. That was all they cared about. So he pulled out a $5 bill and plonked it on the table.

If you ever drive in Vancouver, Canada and wonder how the roads can be so shitty, remember that story.

[tshaddox]

> The problem is the process of going for the lowest bidder, or one of the lowest.

The phrase "lowest bidder" gets a lot of mileage in jokes and social commentary, but I'm pretty sure that the original idea is "the lowest bidder who fulfills the requirements of the contract." In that context it's a pretty obvious process, not something to be feared or mocked. The contract should be "build X to exactly these specifications," and of course the contract should be given to the lowest bidder who can reasonably be expected to fulfill the contract.

In this case, the contractor is clearly unable (or unwilling, or unincentivized) to fulfill the contract, and thus the contractor's bid should be irrelevant. The contractor should not even be considered. After all, if it's fine to not fulfill the terms of the contract, then I'm confident I could submit an even lower bid.

[wahern]

I think the issue in California is that this particular contractor's M.O. is notorious to state officials, but nobody has completed whatever formal processes are required to establish the contractor as disreputable, permitting officials to readily reject their bids in subsequent projects. There has to be a formal process because the whole point of anti-corruption regulations is to remove discretion from individual officials, but state officials aren't being diligent on the backend of that process.

The underbidding contractor is at least one half of the problem because they're serial fraudulent bidders--they have it down to a science. IMO they're flat out the problem from an ethical standpoint. You don't get to cheat somebody just because you've figured out how to exploit a victim's infirmity (the infirmity in this case being the state's bidding rules). You can't shift your blame to the victim; whatever blame the victim is due is independent.

[kasey_junk]

Lots of places have legal requirements to take low bid contracts. This was largely an anti-corruption measure as it was deemed easier to get work out of a low bid contractor than an accurate accounting out of a corrupt process.

But now there exist a whole class of contractor that exist purely to provide the lowest possible level of compliance with municipal contracts. They spend more on legal defense than construction in many cases.

[kelnos]

Certainly CA shares blame for not managing their contracts better (whether due to incompetence or malice).

But it's unethical behavior to be a contractor who intentionally under-bids with the plan to later (repeatedly) charge exorbitant amounts for modifications and fixes.

The government isn't doing its due diligence, but I take a dim view of people who exploit and waste taxpayers' money.

[kazinator]

> But it's unethical behavior to be a contractor who intentionally under-bids with the plan to later (repeatedly) charge exorbitant amounts for modifications and fixes.

If the ethical alternatives all ensure that someone else gets the contract, then that particular ethics in that situation is rather good for nothing.

[kelnos]

I'll never claim that taking the ethical choice will ensure that others won't choose unethically. But "everyone else is doing it" is not an excuse for unethical behavior.

Suggesting that ethics are good for nothing in this case because of that is... well, even more disappointingly cynical than I'm willing to go.

[jobu]

> The problem is the process of going for the lowest bidder, or one of the lowest.

As a homeowner I usually try to get three bids on a project and then take the middle. In my experience the lowest bidder usually doesn't understand what you want or has no idea what they're doing.

[consp]

This is basically the problem with IT Contracts in the Netherlands at the moment (though it is getting better due to proper oversight). Lowest bid winners with no technical knowledge on the proposal side to include the technical challenges in the process. Resulting in expensive shoddy work projects which always overrun in cost due to several factors (design change, minimal/poor work, non extensible etc). The problem is that the overruns are by design as they know there are little to no penalties involved which keeps the companies in the loop for far more than the original sum.

[mc32]

Do what they used to do in the old days. You pay 50% now and the rest in X decades, so long as it meets some minimums set in advance. This would eliminate under bidding. It might put financially weak vendors sag a disadvantage but them’s the breaks.

[Frondo]

Another way to do this involves a shift from lowest-cost purchasing to best-value purchasing. Under lowest cost guidelines, the agency cannot take any factors but price into consideration, leading to this garbage. With best-value, they can look at the contractor's past performance in making a determination. Slowly, agencies are making the change, but a lot of them are still required by guidelines to go with the lowest bid.

[my123]

Best value is hard to evaluate objectively, which is one of the reasons why this mess even happened.

And corruption can affect it evaluating the value too

[Frondo]

It's true that best value can be gamed, but any system can be gamed.

When buyers can define some evaluation criteria like reviewing the bidder's past performance (delivery on-time vs late, cost overruns) in addition to current bid, they can better assess how smoothly this project will go, and past performance is difficult to game without lying, i.e. submitting a fraudulent bid package.

The procurement field has a number of membership organizations and certifications and these people are largely good and ethical workers who take pride in doing good for their employers. Not all, but the good ones far outnumber the lazy, incompetent, and corrupt. (I'm not one, but worked adjacent to the field for a while, and I think buyers often don't get a fair shake, because the corruption can be shocking when it's visible.)

[jlmorton]

Another way to avoid this is with a Design-Build-Operate contract.

The transit agency is on the hook for the agreed construction cost, and for ridership estimates and minimums. If ridership does not develop, then the contracting agency is responsible for making the operator whole.

The contractor is on the hook for the design, build and operation of the system for N years.

The worst possible way to design a large, complex system is the way most US transit agencies do it: the agency operates as the prime contractor, and it issues an initial design subcontractor, which submits an alignment and maybe a 20% design. Then the agency issues bids for each segment, and the new contractor completes the design.

[magduf]

>The transit agency is on the hook for the agreed construction cost, and for ridership estimates and minimums. If ridership does not develop, then the contracting agency is responsible for making the operator whole.

> The contractor is on the hook for the design, build and operation of the system for N years.

This won't work: the contractor will just declare bankruptcy when things don't work out that great. But the owners will have bags full of money by then.

[jlmorton]

This is not an idea I just dreamed up. It works and is in practice around the world.

For a successful example, look to the Canada Line in Vancouver.

https://en.wikipedia.org/wiki/Canada_Line

[adrianN]

Make suitable insurance for the case of bankruptcy a prerequisite then.

[rstuart4133]

From https://www.abc.net.au/radionational/programs/backgroundbrie...

> In the early 2000s after the collapse of insurance giant HIH, NSW and Victoria abolished some classes of builders' warranty insurance. The governments said it was to try and stabilise the industry and ensure its viability. > > So if you're in a new apartment building of more than three storeys, instead of making a claim with an insurer, it's now you versus the builder. You've still got a warranty, but you may need a lawyer.

The rational was builders of that size and above would avoid being put out of business by insurance claims. It had been true up until that point. But from https://www.abc.net.au/news/8403744 :

> Then, in February this year, almost three years after the owners commenced legal action, the developer, Payce Properties, announced that it was being placed into administration. > > The parent company, Payce Consolidated, is still trading, and has plans to roll out more than 7,000 more apartments in the next five years.

Surprise, surprise, there has been a rash of shoddily built large buildings with arguments about was footing the bill every since.

[Reason077]

> "Another way to avoid this is with a Design-Build-Operate contract."

The linked PDF seems to argue for the opposite. It suggests that "design & build" contracts tend to inflate costs, and countries with lower subway construction costs tend to separate design contracts from construction contracts.

[i_am_proteus]

Could anyone comment on whether or not awarding contacts to the second-lowest bidder (whose bid meets technical merit) has ever been practiced by governments? Or other alternative auction strategies.

[jcranmer]

Most competent contracting doesn't go for lowest-cost bidder. Rather, bids are generally scored for some metric, and a good contracting agency should additionally be evaluating the contractors to figure out if they actually are capable of performing their bid without cost overruns.

One of the contentions of Alon Levy [author of the blog in question] is that the primary reason for inexcusably high costs in the US is the hollowing out of agencies to the point that they can't do this sort of evaluation anymore.

[Gibbon1]

I've heard repeatedly that California in particular does not have enough engineering staff to manage contractors for state projects. That's a result of the US's obsession with neoliberal economics which sees those people are a waste of money and an impediment to getting things done.

My dad worked for the government doing management and oversight. It takes a lot of work to evaluate and reject bad low ball bids. Even more work to create bids that you can hold contractors to and to weed out shysters.

[magduf]

If you have to put this much effort into managing contractors and avoiding shysters, at what point does it make more sense to just not use contractors at all, and do it yourself?

This reminds me of when I used to own homes. It was generally easier to just do some home-repair job myself than deal with the hassles of finding a good contractor, and then having to deal with more hassles if the contractor didn't do it right.

[Gibbon1]

I completely understand that dilemma. Outsourcing allows you to tap economies of scale and line up the tasks with contractors experience and capabilities. But you pay for that with another layer of management.

I suspect that there isn't a magic set of incentives that replaces appropriate levels of funding, good faith and sufficient oversight.

[ComradeTaco]

Most large infrastructure projects have a large number of trades for a short amount of time Contractors have people,experience and equipment that would be cost prohibitive to maintain full time but can be relatively inexpensive when hired for just their needed duration.

The kinds of work that you describe fall more under maintenance and basic repair, which most public agencies have full time employees that perform that roles. For instance the NYC subway has signaling workers for a constant stream of signal work, track workers to maintain track and sign workers for endless lifecycle replacement of signs. New construction work, like tunnels and stations is done by contractors.

[kelnos]

If sufficient regulatory oversight is politically infeasible, I would think that the government directly employing construction staff would be politically impossible.

[markus92]

I got no source for this, but I've heard about this happening before on European construction projects (lowest 2 bidders being excluded for being too low).

[Someone]

Blindly dismissing the lowest bid would get you in court and lose in no time.

Dismissing a very low bid must be dismissed, giving good arguments that the bid is not realistic is something completely else.

[ComradeTaco]

Generally bid that is not "technically complete" .i.e. does not demonstrate a real ability to undertake and complete the project can be rejected. There is some engineering discretion to this.

[markus92]

The argument given was that it was an unrealistic bid as it was too cheap to be able to be completed. It all depends on how you set up the tender.

[epanchin]

Tender price submissions should be weighted by previous estimates vs reality.

[munk-a]

This may work for obvious bad actors but would probably penalize genuine new market entrants.

[BeeOnRope]

Seems like that would also be subject to gaming.

[0xffff2]

Every system is subject to gaming. The current system obviously is. The relevant question is "is the proposed system better than the old system?", not "can the proposed system be gamed?".

[kelnos]

Thank you for this. I feel like a lot of people like to shoot down new ideas by coming up with a list of flaws, without considering whether the new idea -- even with the flaws -- is still better than the status quo.

[fzeroracer]

Generally speaking this remains true in government contracting as well. Back when I used to work in that sphere the government always went for the lowest bidder. This is usually billed as a way for the government to be fiscally responsible but usually the result is that they get back poor software with the contractor embedding themselves deep into the design and a huge pricetag on actually fixing the damn thing.

That among many other issues is where I realized that the way competition works in this scenario simply doesn't work when the incentive is to keep yourself and your company on the government's payroll rather than write quality software.

[usr1106]

The hot thing in Finland at the moment is the alliance project model. I'm not an experienced expert in the field so I cannot explain it to detail. The newspaper explanation is that the buyer (city council or similar) form a common organisation and share risks and costs. I must admit I was sceptical when it first came up and assumed that either they end up in quarrel or they will find common excuses for the 3-fold costs in the end. But so far I have been wrong. A road tunnel was completed about in time and only exceeding the budget by a relative small percentage (which they explained with added requirements during project lifetime). A tramway project is not complete yet, but so far there have heen only good news about the progress and budget, which is not typical for such a project.

This is the marketing explanation by the contractor: https://www.yitgroup.com/en/media/why-is-the-alliance-contra...