Life-Cycle Budgeting? No, Thanks

Life-cycle budgeting is suddenly the latest transportation-planning fad. Even some free-market groups are promoting it as a way to save tax dollars.

The theory is that a life-cycle analysis will look ahead at all future costs, not just the initial cost, of transportation projects. At first glance, this sounds great. Most transportation fixed infrastructure needs to be replaced every 30 years or so, and rolling stock needs even more frequent replacements. Transit agencies tend to look ahead only 30 years, thereby pretending that such replacements are not needed.

The first clue we need to be suspicious of life-cycle budgeting is that its advocates overpromise the benefits. For example, they claim life-cycle budgeting could have prevented the kind of cost overruns experienced in the Big Dig or the now-cancelled Hudson River tunnel.

Spin me another one. Cost overruns take place because some of the costs were unforeseen (possibly because planners didn’t want to see them). Simply looking further ahead at the costs does not give planners any better crystal balls (especially if they didn’t want to see those costs in the first place).

The second problem with life-cycle budgeting is that its advocates want some sort of federal law imposing it on transportation agencies. “Congress should investigate life-cycle budgeting for infrastructure,” says one article. Just what those agencies need: more red tape consuming scarce transportation dollars and delaying needed transportation projects.

The real problem with federal transportation spending is that Congress has split federal funds into 40 or 50 pots of money and greatly restricted how state and local agencies can spend from each pot. If Congress puts out a pot of $10 billion or so that can only be spent on rail transit, requiring a city to analyze whether rail transit is the best of several alternatives will only insure that the analyses are warped.

Better incentives are the solution, not more bureaucracy. Congress should give state and local governments more flexibility in how to spend the money, combined with incentives (see pp. 7-8) to insure the money is effectively spent. Then it can let the agencies themselves decide what kind of analyses they want to use in making their decisions. If they decide to use life-cycle budgeting, fine, but if not, the truth is that any costs after 40 years or so, when discounted back to the present, are simply not as important (and are far less predictable with any degree of accuracy) as near-term costs and benefits.

Share

15 thoughts on “Life-Cycle Budgeting? No, Thanks

  1. C. P. Zilliacus

    The Antiplanner wrote:

    The theory is that a life-cycle analysis will look ahead at all future costs, not just the initial cost, of transportation projects. At first glance, this sounds great. Most transportation fixed infrastructure needs to be replaced every 30 years or so, and rolling stock needs even more frequent replacements. Transit agencies tend to look ahead only 30 years, thereby pretending that such replacements are not needed.

    I must respectfully disagree, at least in part, with what you have written above. I believe it is perfectly reasonable to attempt to assess what costs will be in the outyears, and to budget accordingly.

    In the case of rail transit, that includes employee compensation (fully burdened with fringe benefits and overhead, and including a reasonable estimate of pay and benefit increases that the unions representing transit system workers will demand, please) and anticipating repairs to rails, switches and signal systems, and so-called “mid-life” overhauls of all railcars.

    In the case of highways (and especially toll highways), shouldn’t the highway (including pavement, bridge decks and other structures) be designed to last at least as long as the construction bonds are outstanding? Apparently that’s what the New Jersey Turnpike Authority did back in the late 1940’s and early 1950’s when it wrote the specifications for the Turnpike’s asphalt pavement. The southern section of the Pike’s mainline (south of Exit 6) in South Jersey is little changed from the earliest days of the turnpike, and that includes the pavement (see article by Peter Samuel’s TOLLROADSNews back in 2000 here).

    I have driven that part of the New Jersey Turnpike as recently as earlier this year, and the blacktop pavement is still in excellent condition. Many of the bridge decks that carry the Turnpike over other roads and larger bodies of water have been replaced, but the performance of the pavement is quite remarkable.

  2. Aarne H. Frobom

    I have some of the same misgivings as Patrick does above. While Randal’s last sentence is generally correct, my work with highway life-cycle costing has led me into cases where the treatment of end of the life cycle is very hard to handle. Even after discounting future benefits and costs to present value, the timing and costs of the action to be taken after the life of the first construction expires can sometimes be decisive for the decisions to be taken in the first year (depending on discount rates and a horde of other assumptions). I have never found a universal good solution. While discounting and uncertainty absolve current officials from much worry, agencies still must consider whether the facility will be repaired, replaced in kind, or upgraded at the end of its life, IF the costs affect current decisions. On behalf of future users, it must be remembered that whatever these future costs are, they won’t be paid in discounted dollars.

    This goes double for transit agencies. I have never seen a life-cycle-cost study for a rail-transit plan that addressed the end of life for the trolley lines and cars. I hereby predict that some of the cities now building trolley lines will, when the life of the track expires and no money is in hand for renewal, make the same decision that cities made in the 1940’s when confronted with the same situation: pave `em over.

  3. Andrew

    Antiplanner:

    Most transportation fixed infrastructure needs to be replaced every 30 years or so, and rolling stock needs even more frequent replacements.

    That is really offbase. Most of the cost of fixed infrastructure investment is in heavy civil work like bridges, earthwork, tunnels, retaining walls and buildings. If properly maintained, the lifespan of earthwork and tunnels is well over 100 years. We are still running on railroad grades and through tunnels constructed before the Civil War 150 years ago.

    Bridges are a bit shorter lived, but its reasonable to expect a lifespan of 75 to 100 years for bridges and elevated railway structures. Elevated railways built at the turn of the century in Boston, Chicago, New York, and Philadelphia all lasted around that range of time. Bridges actually tend to be even older. Amtrak is currently attempting the replacement planning of several bridges in the 110-120 year range of age.

    Rolling stock is also a long lived item. Other than New Jersey Transit with its EMU’s from the 1960’s/early 1970’s, which were allowed to be run into the ground, most agencies have gotten their rail cars to last around 40 years, sometimes more. Amtrak’s fleet is heading in that direction, with the cars on the Northeast Corridor headed for their 40th anniversary in just three years. Most trolley cars lasted 40+ years as did older elevated and subway equipment and electrified commuter equipment in New York, Philadelphia, and Chicago. SEPTA was running cars built in the 1910’s into the 1980’s on the Norristown High Speed Line (Route 100/Philadelphia and Western), and both SEPTA and New Jersey Transit ran older commuter rail EMU’s from around 1930 into the 1980’s. VIA is running equipment on the Canadian that is closing in on 60 years of service.

    What is NOT a long lived item for transit is buses. Its rare for a bus to make it past its 12th birthday.

  4. Andrew

    CPZ:

    The NJ Turnpike has of course been resurfaced since construction. Asphalt pavement surfaces won’t last much more than 20 years. The sub-surface structure of the road should be able to last much longer -50 years is not an unreasonable expectation. If the road structure were failing before that time (as happenedon on I80 in Pennsylvania), it would really be outrageous.

  5. MJ

    What is NOT a long lived item for transit is buses. Its rare for a bus to make it past its 12th birthday.

    Rare? Not in the US. According to 2009 data, more than 16 percent of the buses in the active vehicle inventory were 12+ years old.

  6. FrancisKing

    Antiplanner wrote:

    “Transit agencies tend to look ahead only 30 years, thereby pretending that such replacements are not needed.”

    Life-cycle budgeting, considering discounted benefits and costs, in standard in the UK. What it does not appear to do is to include the costs of the replacement item. That seems to me to be a big mistake.

  7. FrancisKing

    MJ wrote:

    “Rare? Not in the US. According to 2009 data, more than 16 percent of the buses in the active vehicle inventory were 12+ years old.”

    By that age, it is likely to be in a transport museum, or well on the way there. Bus passengers don’t take kindly, in the UK, to being asked to use an ancient relic, despite being charged the full fare.

  8. Dan

    I heard somewhere that pavement surfaces in Europe are constructed better, are more durable and last longer. Is this true?

    Yes. IIRC the ones in Germany were not built by lowest bidder, but by quality record.

    DS

  9. Andrew

    bennett and Francis King:

    I heard somewhere that pavement surfaces in Europe are constructed better, are more durable and last longer.

    Anyone who has driven the Autobahn or Autostrada can tell you that European roads are significantly superior in engineering, surface, and quality of construction. The cracked, potholed, and rutted surfaces so common in America are very rare in Europe.

    I don’t doubt construction techniques (and quality of construction labor) play a part in this. So do gas taxes of dollars per gallon instead of pennies.

    I don’t know that European roads are shorter. There’s about 35,000 miles of motorways in western Europe compared to about 42,000 miles of US interstate. I think it is purely a phenomena of funding.

  10. FrancisKing

    metrosucks asked:

    “How much of the gas taxes are stolen for transit, though?”

    In the UK, most of the taxation on cars is to replace income tax – the UK voters want continental standards of government spending on US levels of income tax. Most subsidy is for buses, where the operator cannot run the service profitably.

  11. C. P. Zilliacus

    Andrew asserted:

    What is NOT a long lived item for transit is buses. Its rare for a bus to make it past its 12th birthday.

    Umm, that may be true today, but wasn’t always.

    When WMATA took-over the four private bus companies to form what became known as Metrobus in the early 1970’s, it also took over a pretty large fleet of buses, mostly GMC “New Looks,” also some GMC “Old Looks,” Whites and some Flxible “New Looks.”

    Many of those GMC New Looks were to remain in revenue service with WMATA until well after 2000, when they were well over 40 years old, like number 1317 (which was originally purchased by the AB&W Transit Company and finally retired from service sometime between 2000 and 2005).

    WMATA’s first purchase of transit buses on its own in 1973 and 1974 was an order of 620 AM General buses like this one. Without exception, the AM Generals were removed from service, and most of them sold for scrap, long before the GM New Looks were retired.

  12. the highwayman

    CPZ, in Boston the MBTA has trams from the 1940’s still running on a daily basis.

    As for those AM General buses, lemons exist in all modes.

Leave a Reply