Malice, incompetence, or a secret third thing

There's a saying about bureaucracies that you should never attribute to malice what you can attribute to incompetence. This is typically the axis along which explanations for dissatisfaction with local government live: Are they evil or stupid?

As a recently ex-bureaucrat, one thing I tried to convey to people was that while, being human, bureaucrats can be influenced by malice or stupidity, most of the time my experience was that people in local government were well-meaning and conscientious. Nevertheless, they—I—frequently did things that frustrated people.

Sometimes, this was simply about a difference of values, or a different conclusion about weighing tradeoffs. ("You're not being ignored, you're being disagreed with" is a line to memorize, in bureaucracy and life in general, via my former boss.)

But a lot of other times, we agreed on the best course of action and that, theoretically, it was achievable. And then I would be in the position of explaining why we couldn't, or wouldn't, do it.

People use the word "structural" a lot, often kind of vaguely to refer to things beyond their control, but I have rarely experienced "structural" problems as tangibly as when I worked for local government. By that I mean: I had to be acutely aware of the Rube Goldberg machine of rules, norms, and institutional incentives that would be set into motion by any given action I might take, and ultimately produce the result of my action.

For example: One project involved workshopping a set of policies and procedures to implement a program, and specifically how to handle applicants to that program whose applications were processed by a third party. We had received some thoughtful feedback from a group of advocates, who laid out a vision of implementation that was generally reasonable and would have represented a significant improvement in fairness and outcomes. We did not adopt the majority of what they asked for, however.

Why? Well, the biggest reason was that faithfully executing their vision would have required a number of additional staff. (At the time, this program was running on probably half the number of people who reasonably should have been working on it.) It also would have required contracting for new software, since the existing platform (essentially Excel) was not capable of doing what they proposed without requiring many, many more staff. Besides not having a budget, procurement and onboarding new software would have taken a year or more because of rules around competitive contracting, negotiating the contract, etc—all things outside of the power of my department, let alone this program team, to change on a relevant timeline. Once onboarded, the software would have required IT staff to manage. We were not authorized to hire any of these people. And no one was going to bid on a contract that had no funding behind it anyway.

All that said, we could have adopted the advocates' asks in the actual official program policies! No one would have stopped us. And, in the short term, we would have gotten a pat on the back from people who, on the whole, we wanted to be on good terms with—maybe even a good press release.

In the medium and longer term, the policies and procedures in the program document simply would not have been adhered to, for all the reasons above. This would have been extremely confusing and frustrating for third party application processors and the applicants themselves, and led to a lot of additional work for the already-understaffed program team to try to explain things. It is not unlikely that at some point the processors, applicants, or advocates themselves would have tipped off media about the fact that the written rules were not being followed, and there would have been bad news stories—or, say, an Office of the Inspector General report—with negative repercussions not only for the department but potentially for the professional futures of the staff who were already working long hours and doing admirable work keeping things running.

Nor do programs that get a lot of bad press or critical OIG reports tend to do well in the next round of budget requests, which is what we would have actually needed to stand up the kind of infrastructure they wanted.

At one point, an advocate told us, about what they wanted: "This is very easy to do." And from one perspective, they were right! If you were starting from scratch, and had a minimal number of smart, dedicated people, it would not have been too difficult to pull off most of what they were asking for. But we were not starting from scratch. We were starting from inside a very cumbersome machine, with work obligations we were already struggling to keep up with, in a system where taking seemingly obvious steps would require permissions that required months or years to get an answer to (and the answer may have been "no") and resource allocations in both time and money that would have required arguing that our colleagues' requests for resources for their own programs—many of which had their own constituents and advocates making solid arguments behind them—should be denied in favor of ours.

At least one of the advocates, when we tried to explain this, told us that we must just not take our jobs very seriously, or care about housing after all.

I don't want to leave you with the impression that the answer is just: The machine sucks, submit to the machine. In fact, many of the most rewarding things I got to work on involved changing how the machine worked precisely to change the kind of structural relationships of work inputs and outputs that were misaligned here.

But for both bureaucrats and advocates, it's critical to understand all the different pressure and failure points in the long series of pulleys and levers between one person agreeing to your demands and the final outcome of that agreement. In organizing terms, you need to be very careful about identifying your target (the person or people who need to say "yes" for you to get what you want), or identifying when there are actually multiple targets who all need to be aligned. In the anecdote I've told here, my team was certainly one target, but it was not the only one, and no amount of pressure or persuasion would remove the need for other pieces of the machine to fall into place.