Safety is the Watchword of Cost Disease

Published on July 22, 2017

Safety is the Watchword of Cost Disease

I tweeted this story out yesterday about stairs in Toronto:

Basically, a citizen built stairs in a garden really cheaply (at 0.3% of the cost!), and Toronto took them down.

The story is absurd, but all of it is unfortunately “rational.” You can imagine a bureaucrat at city hall realizing that the stairs don’t meet a number of requirements for safety, reliability, inspectability, and more, so the obvious answer is to tear down the whole thing. Bad stairs are only likely to invite lawsuits, and so the city would prefer stairs not exist (increasing injuries on the hill) than to allow potentially unsafe stairs to exist.

I say potentially, because the more I study the problem of cost disease, the more that safety in a very broad sense is the root of the cause. No one wants to actually fight over the safety of a small staircase. No, a couple of steps shouldn’t be $150,000, but neither does anyone want to do the work to prove that a lower-quality model would be more than sufficient for the needs of its users.

When it comes to safety, no one ever wants to stand up for the issue of cost over the value of human lives. Even though economists have built an entire literature around the concept of the “value of a statistical life,” it actually takes someone to argue in a council meeting that no one is going to be hurt by a small and cheaply-built staircase in a garden.

So we get these absurd cases where we spend 1000x on a staircase, to reduce the chance that someone (maybe, maybe!) might not fall down them at some point in the future.

There isn’t any good way to solve this phenomenon. But what you can do is focus on what I might call “progressive upgrading.” Start with the cheap stairs, and if foot traffic seems to increase beyond its capability, upgrade to a better staircase as needed. The same thing happens with stop signs versus signals — start with the bare minimum, but if things get worse, one can always replace the cheap model with something more expensive (and safe).

Instead of dithering for years, just get something in place. Adapt to problems as they arise, as opposed to trying to predict them in advance. For small public works like this one, the costs of changing the construction later are minimal compared to the wide savings that happen up front.

Photo by John Payne used under Creative Commons