Infrastructure Service Life as a Matter of Sustainability

The word is everywhere.  People are talking about sustainability, but what does it really mean?  We want sustainable growth, sustainable development, sustainable communities.  In a recent blog posting, the Director of Sustainable Communities at the Natural Resources Defense Council admitted with remarkable candor about the latter topic, “even a lot of environmentalists don’t quite know what to make of the phrase.” 

Most people would probably agree that not using up all available resources needed for life has something to do with what “sustainability” signifies.  Physicists, chemists, and biologists can define with some precision the oxygen, water, and food, required by a group of organisms for a particular period of time, and the quantity of waste products they will produce.  For humans, their communities and their civilizations, there is certainly a lot more to it.  Our population, our history, and our ideas continue to grow, increasing our demands on an expanding range of resources. 

The American Society of Civil Engineers (ASCE) tried to cover everything by defining sustainability as “a set of environmental, economic, and social conditions in which all of society has the capacity and opportunity to maintain and improve its quality of life indefinitely without degrading the quantity, quality, or availability of natural, economic, and social resources.”  Improving quality of life “indefinitely” with no degradation of resources sets the bar quite high.

The Brundtland Commission took a more modest perspective, defining sustainable development as “…development that meets the needs of the present without compromising the ability of future generations to meet their own needs.”  If one considers that our concept of “needs” has evolved over time, along with our technology and other capabilities to meet our needs, then the Brundtland criterion is considerably less stringent than ASCE’s.   As we look to more distant generations, we necessarily have less confidence that their concept of their own needs will be similar to ours.  Although some people will judge it to be technological hubris, we also are inclined to presume that these future generations will find new ways to meet whatever those needs are. 

The idea of discounting the significance of the more distant future is fundamental to the benefit-cost and cost-effectiveness analysis practices.  The practices, developed primarily in the middle decades of the 20^th Century, are now widely used to assess the wisdom of decisions involving expending resources to obtain services or other benefits over a period of years.  These capital investment decisions—for example, to build infrastructure facilities—certainly have the capacity to foreclose future options.

A crucial assumption in any such analysis is the discount rate, usually expressed on an annual basis, representing the decline in value of resources that are made available or used at some future time, as compared to their value today.   For a simple example, many people would readily accept the idea that being given $98 today is preferable to the promise of $100 to be received one year hence.  If the proposition is $90 today versus $100 next year, more people may be willing to wait.  These examples represent discount rates of 2% and 10% respectively.  The discount rate used in any particular analysis reflects economic conditions and analyst’s and investors’ expectations about the future.  Higher discount rates imply less confidence and a requirement that the initial investment should more quickly become profitable.  (An interesting footnote: Comparisons of the value of the goods traded by the Dutch to purchase Manhattan from the Native Americans in 1626 to the current assessed value of the real estate suggests an annual increase of 5 to 6%.  It seems unlikely that Peter Minuet could have imagined what would become of the island.)

What can this tell us about understanding sustainability?  If we imagine that life will carry on more or less smoothly for some time, we might apply a low discount rate, perhaps 2% annually, in deciding whether or not to invest in infrastructure.  The discounting calculations would tell us that benefits accruing to a new investment more than 160 years from now would have a present worth less than 5% of their future amount.  If a human generation is about 20 years, we might then conclude that any long-term benefits that will not be realized until the 9^th generation after we make the investment and building the facility are not worth much.  If the discount rate is 5%, fewer than 4 generations (perhaps 50 years) are needed.  With a discount rate of 10%, it takes about 30 years to reach the point that returns promised further in the future may not be influential on our decisions today. 

In other words, it may not make much sense to try to think beyond our grandchildren when it comes to sustainability.  Consider this: the initial demonstrations of transistor technology occurred in the mid-1940s, when I was a toddler.  Only a few years earlier, ENIAC I was developed as the first modern computer, using some 17,000 vacuum tubes.  Who then would or could have imagined the microchip and digital computers—not to mention cell phones—my grandchildren today take for granted. 

Perhaps it is only coincidence, but designers of bridge typically assume their structures will last about 50 to 70 years; for highway pavements, the numbers are 30 to 35 years.  Regardless of such assumptions, most of the people responsible for managing the nation’s infrastructure strive to get as many years of service as possible before a facility must be substantially refurbished or replaced.  Only seldom do public-sector owners of infrastructures make provisions at the beginning of the service life to ensure that funds are available to do the right thing at the end.

If we want our infrastructure systems to be sustainable and to serve as a basis for sustainable communities, such a management strategy may be counterproductive.  Long service lives discourage adoption of new technology and responsiveness to changes in users’ demands and society’s values.  Long-lived infrastructures fix the patterns of land development, social and economic activity, and perceptions of spatial relationships on the regions they serve.  Anecdotal evidence suggests we might be better of planning that infrastructures should be replaced, substantially renovated, repurposed, or retired no more than 3 generations after they are initially constructed.  Where rapid changes are occurring in economic conditions (for example, population growth in Phoenix or shrinkage in Detroit) or the technology (electricity generation and control, for example, as compared to municipal waste disposal), the lifetimes should be much shorter.

Seattle’s Alaskan Way Viaduct, for example, an elevated highway constructed to carry traffic past the city’s downtown core, was completed in 1953.  Over the years since it was built, traffic levels and truck sizes and weights had grown to levels that exceeded what the original designers had in mind; the structure was effectively obsolete.  Replacement would have been costly and very disruptive for the region’s highway users and the Viaduct’s neighbors. 

It was the 2001 Nisqually earthquake that moved Seattle finally to declare an end to the structure’s useful life.  The Viaduct was seriously damaged; sections had to be closed for a time. Repairs were made, but much of the public understood that more would need to be done.  It nevertheless took strong political leadership to effect the Viaduct’s replacement—with a tunnel—initiated in 2011.

The now-infamous The Fukushima I Nuclear Power Plant in Japan, first commissioned in 1971, uses “Generation II” technology.  Newer “Generation II” plants have better safety features.  Because nuclear plants are very expensive to build, reducing their allowable service would significantly raise the hurdle for establishing feasibility of a new plant but might also reduce the risk such infrastructures can pose for their neighbors.

There are many more examples that might be given of infrastructures destroyed or removed before the time envisioned by their builders and operators.  Certainly many elements of any particular piece of infrastructure are likely to remain useful beyond the 3- to 5-decade life implied by sustainability considerations.  However, viewing these elements as evidence of overinvestment and designing to ensure that infrastructures can be salvaged, recycled, or put to other uses would be important steps to improved sustainability.