Don’t Go Chasing Waterfalls: A Review of ‘Recoding America’

I’ve long held that there are two schools of thought on strategic plans. One school says they should be as comprehensive as possible so that everybody can see themselves in it and every eventuality is covered. The other says that a strategic plan should be brief and stay at a high level to allow for adaptation along the way.

I’m firmly in the second camp, but I’ve had a hard time converting others. Jennifer Pahlka’s new book, Recoding America, finally helped me put words to my intuition.

Pahlka spent years consulting with various government agencies, trying to improve their performance. The core of the book is an argument that the separation between policy and implementation is both false and self-defeating and that it has its roots in an elitist view that when it comes to implementation, we have people for that sort of thing. That view leads to what she calls a “waterfall” theory of strategic planning, in which direction is set at the top and each lower level is supposed to do what it’s told. The flaw in that system, as she quotes Clay Shirky noting, is that it “amounts to a pledge by all parties not to learn anything while doing the actual work” (60).

As one could imagine, some people adapt to the deskilling by adopting a sort of learned helplessness. As one midlevel manager in an agency memorably put it, if we’re told to build a concrete boat, then we build a concrete boat: that way, if it doesn’t work, we don’t get in trouble. We followed the process. Whether the process makes any sense is someone else’s department.

Pahlka notes repeatedly that the incentives for career agency personnel aren’t around outcomes; they’re around process. Getting bad outcomes may lead to the occasional awkward moment at an oversight hearing, but deviating from procedure can get you fired. So they spend time and energy on compliance, rather than optimization. As she put it, “When systems or organizations don’t work the way you think they should, it is generally not because the people in them are stupid or evil. It is because they are operating according to structures and incentives that aren’t obvious from the outside” (57). Anyone who has worked in a large public organization knows the truth of this.

The opposite of a waterfall system is an agile one. In that version, the implementers and the planners work together, building iteratively as they learn lessons. In this version, someone would point out that a concrete boat would sink, and the plans would change. In agile design, strategic plans are necessarily shorter; epistemic humility is a core principle. Nobody knows everything, especially in advance, so it’s better to build plans that make room for learning along the way.

My copy of the book is laden with margin notes along the line of “YES!” and “Ugh!” The examples—and there are many—are well told, familiar and maddening. A story about signing up for VA benefits is especially vivid. Veterans who needed benefits had to sign up on an online form, but the form only worked with one version of Internet Explorer and one version of Adobe Acrobat, both of which had quickly been superseded. The computers the policy makers used, though, were set to the old versions, so the form worked fine in the building. When one charismatic vet managed to get the policy people to watch him try to sign up from a computer in a public library, they saw the issue and quickly got it fixed, resulting in a surge of applications. That surge triggered pushback from the next level of the bureaucracy; they were concerned that the new backlog of cases would be held against them. They wanted to go back to the old form so their data would look good.

I could only nod in recognition.

Pahlka’s focus isn’t higher ed, but those of us who have spent years in public higher ed will see plenty to recognize. New policies rarely remake old ones; instead, they’re layered on top, like sedimentary rock. Over years, layers of rules beget layers of regulations, which generate ever more exacting demands for compliance. Working at the Council over the past year, I noticed how nearly every bill in the Legislature includes the word “notwithstanding.” They use that in recognition that they don’t know what all of the existing layers are; it’s a sort of dismissive hand-wave.

The upshot of the book for me is about the need to recognize implementation as being just as important as policy. Among other things, that means including the implementers in the discussions at formative stages. The folks who actually have to operate the software know things about it that the pure policy people don’t. Build workflows that allow for learning and improvement. Use strategic plans to help people understand the why behind decisions, so when they have to adapt on the fly, they know in which direction to go. And for goodness’ sake, stop confusing punctilious compliance with good performance.

Thank you, Jennifer Pahlka, for putting an entire framework and copious research in the service of explaining a deeply held intuition. Highly recommended.