A Short Masterpiece on the History of Sewers


Frequent commenter cynic chimed in on a discussion of San Francisco's sewer system with a short masterpiece of a history lesson.

I had suggested in the original post that sewers were sort of miraculous and that if we had to build them today, we would fail on account of our inability to stick to long-term projects.

Cynic's narrative is just great reading and it's also valuable for delivering two reminders. One, the short-term, get-off-cheap mentality is no specific feature of our time, even if I would suggest that it is slightly more prevalent now. Two, the "cost" of something is not a fixed entity, particularly with sociotechnical projects that take a long time to build -- and must last even longer. As cynic concludes, "The cost equation, you see, can be inverted." 

American cities once relied on private cesspools and privy chambers, which had to be emptied on a periodic basis. Mounting density and a growing concern with disease stressed these older systems. But it was the spread of municipal water works that really precipitated the change. Raising water from a well, or carrying it from a town pump, served as a natural brake on consumption. With water piped directly into homes, usage shot upward. That's why people started putting in brick sewers in the 1860s. And by the time Grunsky was called in, the residents of San Francisco were using some 23 million gallons of piped-in water each day, and returning three-quarters of it to their sewers. No system of septic tanks and privy vaults could possibly have kept pace with that volume. Running water was a solution that created an enormous problem.

San Francisco's solution to this problem was not as admirable as you suggest. The problems with the sewer system it constructed, although exacerbated by growth, were actually designed into the sewers from their inception. The city realized, quite early on, that rainwater would stress its system beyond its limits. Grunsky estimated that the sewers would overflow twenty-six times a year. But he wasn't overly worried about it. He knew that the rainwater would serve to flush out sewers, performing a necessary task at no cost, and he designed special pipes to ensure the further dilution of the overflowing effluent with rainwater before it exited the system.

There was, of course, and alternative. Every subsequent coastal city in California (which is to say, all of them) installed a separated sewer system. One set of pipes for the storm-drains and rainwater, and one for the sewage itself. San Francisco, too, considered such a system. It alone faced the challenge of a substantial retrofitting of its sewers - portions of its system were separated, but large portions of the hodgepodge were combined. And it opted for the cheaper, easier, quicker fix. It preserved separated systems where they existed, and even planned some new ones in very specific places - flat districts where costly pumping and flushing would be necessary, making it more cost-effective going forward to pump and flush sewage alone than to deal with the combined flow. But the city of some 340,000 residents decided not to shoulder the expense of separating out the entire system, with the attendant infrastructure and higher operating costs. And what looked prohibitively expensive then has become a little less plausible with every passing decade. In retrospect, San Francisco would have saved itself a great deal of money to make the upfront investment in a fully separated system. But it had a cheaper way of solving its pressing problems, and it took it.

I should add that San Francisco isn't unusual in the national context. Almost every nineteenth-century city in America installed combined sewers, and I don't think that any combined system has, as of yet, been fully separated. But if you'll pardon my challenge to your narrative of moral decline, there's reason to hope that we're capable of more far-sighted action today than were the city fathers of San Francisco. In Cambridge, the sewer system is only about a tenth the size of the San Francisco network, but the challenges and costs of replacement are roughly commensurate. The city commenced building separate systems in the 1930s, but by then most of its infrastructure was already in place. It opportunistically replaced small portions of the system over the subsequent decades, but began to do so more systematically about twenty years ago. Over the last decade, in particular, it has used its burgeoning tax base to fund infrastructure projects in general, and the sewer separation scheme in particular. I expect that at some point in the next ten years, Cambridge will hold a press conference to announce that it has become the first American city to fully separate an old, combined system. And though it has received some aid from state agencies in the process, most of the work has been funded out of the capital budget, without further assumption of debt.

Will San Francisco or other major cities follow its lead? That's doubtful. The real problem here isn't lack of leadership; it's the time-horizon. If Cambridge pulls this off, it will have taken decades. The city has had the same city manager in place for the entirety of the project; he seems determined to see it through to its conclusion. Cities run by elected mayors tend to focus on projects with shorter life-cycles. Public authorities are sometimes one way around this problem, but that doesn't seem to have done San Francisco much good. Another answer is to offer water-quality waivers for cities that demonstrate steady and consistent progress toward this longterm solution. The cost equation, you see, can be inverted - it would have been cheaper on a short-term, annual basis for SF to have spent thirty years separating its system, than for the quick-and-inadequate fix that the Feds and the city installed. But I'm not holding my breath.