264
This capital investment is irrecoverable—climate change creates a huge stranded
capital risk for oil rigs and refineries, ports, large dams, and irrigation districts. As
infrastructure is a long-term investment, small mistakes in planning may result in
large impacts in a rapidly changing, unpredictable future.
We cannot afford to make infrastructure mistakes—financially or otherwise.
Most of US infrastructure, and much global infrastructure, was built during a different era: burgeoning economic growth, low public debt, and large government surpluses. Rebuilding infrastructure during lean fiscal times is an entirely different
proposition. An ultimate cause of FEW infrastructure collapse in the twenty-first
century has been, and will continue to be, fiscal exhaustion. In the usual cycle soft
socioeconomic infrastructures fail first, followed by fiscal exhaustion of the proximate infrastructure-governing institutions, followed by the slow neglect of hard
infrastructures, followed by “sudden” collapse of hard infrastructures. Our twentyfirst century infrastructure investments need to be smarter, more resilient, more
cost-effective, more modest, and more fiscally sound.
So, how can we build infrastructure resilience?
Eliminating lock-in is hard to imagine, but lock-in could be reduced by shifting
to distributed infrastructure that facilitates services on a smaller, more local scale;
increasing the diversity of the system; shortening lead times to make changes; and
reducing the risk of large losses and stranded capital. Nonetheless, distributed
small-scale infrastructures tend to have higher marginal costs, which advantages
traditional infrastructure during long-term infrastructure planning. Local governance is a form of distributed soft infrastructure. Is distributed infrastructure still
“infrastructure”? Yes and no. Yes because it serves the same functions. No because
it lacks the systemic lock-in property and it mitigates many of the systemic problems associated with large scale infrastructure. If we can provide the services of
infrastructure, without the problems, this is a good thing—regardless of whether we
still call this “infrastructure.”
Improving our foresight by investing in science and information is also a good
strategy to improve our resilience to future change. However, foresight is emphatically not 20/20, and history is full of examples of catastrophic failures to anticipate
change… and we should not make the mistake of imagining we are exceptional in
this regard. The world is full of Black Swans, Knowable Unknowns, and Knightian
Unknowns. The Black Swan is an unforeseeable and consequential event that has
never happened before, like the emergence of the atomic bomb during World War 2,
the founding of Islam, or the creation of social media and the internet. Knowable
Unknowns are important decision factors that we include in our calculations and
that we could in principle estimate or observe, but which have not yet been estimated or observed to an adequate precision (perhaps due to cost or lacking knowledge), like the exact volume of water remaining in California’s freshwater aquifers
or the likelihood that the Earth will be struck by a large meteor tomorrow. Knightian
Unknowns are important decision factors that we know we cannot know, or that we
do not even know exist, like (from the perspective of a medieval plague doctor) the
existence of microbes as a key factor in medical pathology.
B. L. Ruddell et al.
This capital investment is irrecoverable—climate change creates a huge stranded
capital risk for oil rigs and refineries, ports, large dams, and irrigation districts. As
infrastructure is a long-term investment, small mistakes in planning may result in
large impacts in a rapidly changing, unpredictable future.
We cannot afford to make infrastructure mistakes—financially or otherwise.
Most of US infrastructure, and much global infrastructure, was built during a different era: burgeoning economic growth, low public debt, and large government surpluses. Rebuilding infrastructure during lean fiscal times is an entirely different
proposition. An ultimate cause of FEW infrastructure collapse in the twenty-first
century has been, and will continue to be, fiscal exhaustion. In the usual cycle soft
socioeconomic infrastructures fail first, followed by fiscal exhaustion of the proximate infrastructure-governing institutions, followed by the slow neglect of hard
infrastructures, followed by “sudden” collapse of hard infrastructures. Our twentyfirst century infrastructure investments need to be smarter, more resilient, more
cost-effective, more modest, and more fiscally sound.
So, how can we build infrastructure resilience?
Eliminating lock-in is hard to imagine, but lock-in could be reduced by shifting
to distributed infrastructure that facilitates services on a smaller, more local scale;
increasing the diversity of the system; shortening lead times to make changes; and
reducing the risk of large losses and stranded capital. Nonetheless, distributed
small-scale infrastructures tend to have higher marginal costs, which advantages
traditional infrastructure during long-term infrastructure planning. Local governance is a form of distributed soft infrastructure. Is distributed infrastructure still
“infrastructure”? Yes and no. Yes because it serves the same functions. No because
it lacks the systemic lock-in property and it mitigates many of the systemic problems associated with large scale infrastructure. If we can provide the services of
infrastructure, without the problems, this is a good thing—regardless of whether we
still call this “infrastructure.”
Improving our foresight by investing in science and information is also a good
strategy to improve our resilience to future change. However, foresight is emphatically not 20/20, and history is full of examples of catastrophic failures to anticipate
change… and we should not make the mistake of imagining we are exceptional in
this regard. The world is full of Black Swans, Knowable Unknowns, and Knightian
Unknowns. The Black Swan is an unforeseeable and consequential event that has
never happened before, like the emergence of the atomic bomb during World War 2,
the founding of Islam, or the creation of social media and the internet. Knowable
Unknowns are important decision factors that we include in our calculations and
that we could in principle estimate or observe, but which have not yet been estimated or observed to an adequate precision (perhaps due to cost or lacking knowledge), like the exact volume of water remaining in California’s freshwater aquifers
or the likelihood that the Earth will be struck by a large meteor tomorrow. Knightian
Unknowns are important decision factors that we know we cannot know, or that we
do not even know exist, like (from the perspective of a medieval plague doctor) the
existence of microbes as a key factor in medical pathology.
B. L. Ruddell et al.