4 ms·

> When we added up the replacement cost of all of the city's infrastructure -- an expense we would anticipate them cumulatively experiencing roughly once a gen

> When we added up the replacement cost of all of the city's infrastructure -- an expense we would anticipate them cumulatively experiencing roughly once a generation -- it came to $32 billion. A generation is what, 25 years? If you expect to replace all roads/electricity poles/sewers etc every 25 years you seriously need to look a quality. For this I feel this whole article is off. > Humans are predisposed to highly value pleasure today and to deeply discount future pain This is not true. One of the best test for predicting a child success is seeing if they will defer a treat now for more treats later. Many people have the sacrifice now for a better future instinct. This article seems to make some vastly incorrect assumptions. Personally I think where cities are overspending 3 things could really help. 1) Making asset sales/tenders/expenditure very public and show comparable cost ratios. E.g. cost per km of sewer 2) Bringing more work back in-house. I feel this 'privatise everything is better' mentality is simply not true and does things that leads to infrastructure that needs replacement every 25 years and cost blowouts. 3) If you are going to privatise services price them on points of quality that you can enforce. E.g. Rather than jails getting a straight fee, they should get a small per convict allowance but a generous bonus for each released convict that spends 5 years crime free after release.

flukus 10y ago
> Making asset sales/tenders/expenditure very public and show comparable cost ratios. E.g. cost per km of sewer I think this is one of the main drivers in favoring cheap and punishing quality. I agree with 2 and 3, though I'd argue that private jails are one of the things that should be bought back in house.
- pyre 10y ago
  > I think this is one of the main drivers in favoring cheap and punishing quality. The problem comes with evaluating quality. There are issues with over-valuing cheapness, but there are equally issues with using high-cost as an indicator of quality too.
  - socialist_coder 10y ago
    Could cities not do something where they pay 50% of the construction bill over the expected lifetime of the project, with a big chunk of it at the end? The cost of unexpected repairs would simply get deducted from this money. It would mean that only big companies could take on these projects (because they'd only be making 50% of their money right away), but they could still subcontract out to smaller companies and pay them 100%.
    - flukus 10y ago
      If >50% of the cost is maintenance then those companies could still build it cheaply then laugh all the way to the bank.
- Gustomaximus 10y ago
  Fair point about this can push cheap. I guess this is said with the expectation people that know what they are doing have relevant quality requirements in said tender.
droopyEyelids 10y ago
Love the idea behind point #3 Like many of us, my mind was instantly drawn to how the incentives would be gamed, but you know what? The current incentives are already gamed. We all read about the judge who got kickbacks for sending kids to a juvenile detention facility, etc. Let's build our incentives in the opposite direction. A certain percentage of dishonest people will game the system. But many, and probably most, will actually work towards the idea. And nothing is accomplished perfectly in one pass, there will be reform and accountability.
otterley 10y ago
> A generation is what, 25 years? If you expect to replace all roads/electricity poles/sewers etc every 25 years you seriously need to look a quality. For this I feel this whole article is off. Isn't that the typical depreciation schedule for infrastructural assets? I've seen some useful life tables used by accountants and 25 years seems to be a typical mark. Maybe 30 years, but rarely 50.
- rhino369 10y ago
  Depreciating it over 30 years is probably reasonable, but automatically counting replacement as an unfunded liability is way too conservative. Heavily traffic roads will need replacing at least every 30 years. But in my college down, they had side streets made of brick that has lasted over 80 years. Things like water mains and sewerage pipes have a lifetime of 100 years or more. A lot of the systems are now due up for replacement for the first time and it will cost a lot. But it won't have to be done again again in our lifetimes.
  - AstralStorm 10y ago
    The problem is, since nobody budgeted for this rare but predictable replacement, there is no money for it. This is why even long replacements have to be budgeted annually. They may be cheaper, but they silk exists. It is always somebody else's problem, but then it sucks to be that person.
xenadu02 10y ago
A concrete road is good for 50 years at best. Water and sewer lines somewhere between 50-100 years.
- AstralStorm 10y ago
  Which means one will see them replaced during their lifetime and grandkids paying for it.
- nickjarboe 10y ago
  If you "over engineer" the road it can last much longer. The 580 interstate in Oakland was built in the 1960's to standard US interstate specs but bans large trucks 99.9% of the time. It is in almost original condition. The parallel interstate (the 880) in Oakland has all the trucks and is in very bad condition with constant repair.
runeks 10y ago
> This is not true. One of the best test for predicting a child success is seeing if they will defer a treat now for more treats later. Many people have the sacrifice now for a better future instinct. Surely, any intelligent child would factor in the rate of interest (additional treats earned per unit of time) and bond price (how many treats are sacrificed right now) before making their decision. Automatically saying yes to more treats later is analogous to not caring whether you earn 0.1% interest or 10%, and that's certainly not intelligent. Promising more treats later in exchange for treats now is exactly what a bond is, and any wise investor would factor in rate of interest/coupon rate.
Smaug123 10y ago
> One of the best test for predicting a child success is seeing if they will defer a treat now for more treats later. Although a different take on it is provided by [1], which suggests that the marshmallow test is not always testing for the ability to defer reward, but for the ability/will to obey instructions/please authority figures. I consider the post well worth reading. [1]: http://hotelconcierge.tumblr.com/post/113360634364/the-stanford-marshmallow-prison-experiment http://hotelconcierge.tumblr.com/post/113360634364/the-stanf...
- WildUtah 10y ago
  Also, a substantial number of children know the truth about marshmallows: they're disgusting. The study might get very different results with bittersweet chocolates with buttercream filling.

flukus 10y ago

> Making asset sales/tenders/expenditure very public and show comparable cost ratios. E.g. cost per km of sewer I think this is one of the main drivers in favoring cheap and punishing quality. I agree with 2 and 3, though I'd argue that private jails are one of the things that should be bought back in house.

pyre 10y ago
> I think this is one of the main drivers in favoring cheap and punishing quality. The problem comes with evaluating quality. There are issues with over-valuing cheapness, but there are equally issues with using high-cost as an indicator of quality too.
- socialist_coder 10y ago
  Could cities not do something where they pay 50% of the construction bill over the expected lifetime of the project, with a big chunk of it at the end? The cost of unexpected repairs would simply get deducted from this money. It would mean that only big companies could take on these projects (because they'd only be making 50% of their money right away), but they could still subcontract out to smaller companies and pay them 100%.
  - flukus 10y ago
    If >50% of the cost is maintenance then those companies could still build it cheaply then laugh all the way to the bank.
Gustomaximus 10y ago
Fair point about this can push cheap. I guess this is said with the expectation people that know what they are doing have relevant quality requirements in said tender.

droopyEyelids 10y ago

Love the idea behind point #3 Like many of us, my mind was instantly drawn to how the incentives would be gamed, but you know what? The current incentives are already gamed. We all read about the judge who got kickbacks for sending kids to a juvenile detention facility, etc. Let's build our incentives in the opposite direction. A certain percentage of dishonest people will game the system. But many, and probably most, will actually work towards the idea. And nothing is accomplished perfectly in one pass, there will be reform and accountability.

otterley 10y ago

> A generation is what, 25 years? If you expect to replace all roads/electricity poles/sewers etc every 25 years you seriously need to look a quality. For this I feel this whole article is off. Isn't that the typical depreciation schedule for infrastructural assets? I've seen some useful life tables used by accountants and 25 years seems to be a typical mark. Maybe 30 years, but rarely 50.

rhino369 10y ago
Depreciating it over 30 years is probably reasonable, but automatically counting replacement as an unfunded liability is way too conservative. Heavily traffic roads will need replacing at least every 30 years. But in my college down, they had side streets made of brick that has lasted over 80 years. Things like water mains and sewerage pipes have a lifetime of 100 years or more. A lot of the systems are now due up for replacement for the first time and it will cost a lot. But it won't have to be done again again in our lifetimes.
- AstralStorm 10y ago
  The problem is, since nobody budgeted for this rare but predictable replacement, there is no money for it. This is why even long replacements have to be budgeted annually. They may be cheaper, but they silk exists. It is always somebody else's problem, but then it sucks to be that person.

xenadu02 10y ago

A concrete road is good for 50 years at best. Water and sewer lines somewhere between 50-100 years.

AstralStorm 10y ago
Which means one will see them replaced during their lifetime and grandkids paying for it.
nickjarboe 10y ago
If you "over engineer" the road it can last much longer. The 580 interstate in Oakland was built in the 1960's to standard US interstate specs but bans large trucks 99.9% of the time. It is in almost original condition. The parallel interstate (the 880) in Oakland has all the trucks and is in very bad condition with constant repair.

runeks 10y ago

> This is not true. One of the best test for predicting a child success is seeing if they will defer a treat now for more treats later. Many people have the sacrifice now for a better future instinct. Surely, any intelligent child would factor in the rate of interest (additional treats earned per unit of time) and bond price (how many treats are sacrificed right now) before making their decision. Automatically saying yes to more treats later is analogous to not caring whether you earn 0.1% interest or 10%, and that's certainly not intelligent. Promising more treats later in exchange for treats now is exactly what a bond is, and any wise investor would factor in rate of interest/coupon rate.

Smaug123 10y ago

> One of the best test for predicting a child success is seeing if they will defer a treat now for more treats later. Although a different take on it is provided by [1], which suggests that the marshmallow test is not always testing for the ability to defer reward, but for the ability/will to obey instructions/please authority figures. I consider the post well worth reading. [1]: http://hotelconcierge.tumblr.com/post/113360634364/the-stanford-marshmallow-prison-experiment http://hotelconcierge.tumblr.com/post/113360634364/the-stanf...

WildUtah 10y ago
Also, a substantial number of children know the truth about marshmallows: they're disgusting. The study might get very different results with bittersweet chocolates with buttercream filling.