One size fits all?

In a recent Washington Technology column, titled "Beware of the tyranny of one size fits all," Stan Soloway warns against the tendency -- in government procurement and by extension in other areas of government management and even policy -- to encourage or even mandate a single approach to various challenges the government faces.

I have known Soloway around the government procurement world since the 1990s, when I was in government and he ran one of the professional services trade associations. He later went on to lead acquisition reform at the Defense Department in the late 1990s, and then to run, for 15 years, the Professional Services Council, which he left last year to start his own consulting business. Reading the column was a trip down memory lane from procurement reform and the "reinventing government" movements of the 1990's, bringing back discussions and arguments from that period that are very much still relevant today, but which have in recent years faded from attention.

Soloway notes that a defining feature of reinventing government in general, and 1990s-era procurement reform in particular, was to move away from one-size fits all to a situation where career managers and frontline employees were encouraged to figure out what decision made most sense in a specific situation. The idea was that a general rule produced bad decisions in many specific cases.

Also, there was a feeling that the tyranny of the general rule was insufficiently respectful of the abilities and desires of people to decide for themselves -- that it took away from their humanity. One of the first reform steps in DOD, Soloway notes, was a memo from then-Secretary of Defense William Perry stating there should no longer be a presumption that government-unique military specifications should be used for weapons systems. Instead, Perry call for a case-by-case determination of whether milspecs were needed or commercial specs were more cost-effective. The words we used then to describe what we wanted the federal workforce to do when making decisions were "judgment," "discretion," and "using your head."

Soloway sees a move in recent years back to one size fits all. "Unfortunately," he writes, "specific, mandatory certifications and specifications, many of them government-unique, have actually been making a worrisome comeback in recent years. …. From cost accounting to capability maturity models, from contract type to acquisition strategies, we see too many procurements that send a clear message: Those who haven't done it our way, or are not doing it a certain way, need not apply."

He also quotes Jennifer Pahlka, founder of Code for America, warning against "the tyranny of misapplied doctrine" -- in this case, a one-size-fits-all attitude toward agile development. "Agile," according to Pahlka, "is one useful doctrine, not the doctrine."

The view we had at the time, though, was controversial. Many in politics, the public and the media were what might be described as government workforce pessimists, who believe the typical fed is neither very smart nor very motivated, and may even be venal.

Asked to allow decisions that depart from one-size-fits-all by exercising judgment, the workforce pessimists worry that feds would make poor decisions that would be worse than those dictated by one size fits all. They believe it is less dangerous to put up with the shortcomings of one size fits all than to accept the results of greater discretion.

Workforce pessimism is associated with what New York University Professor Paul Light calls the "watchful eye" approach to government management, where inspectors general and auditors must check what feds do to catch bad behavior. A heavy reliance on rules, including ones that are of the one-size-fits-all variety, is part of the watchful eye toolkit.

We in the 1990s reinventing government and procurement reform movements were something of workforce optimists -- have auditors and IG's to patrol against venality, like the view that laws should generally seek to stop bad behavior but not necessarily to require what some would see as good behavior. Absent venality, bet on the ability of the workforce -- especially with encouragement and training, and especially as education levels in the workforce rose -- to make more decisions that were improvements over one size fits all than those that made things worse.

These debates have receded in recent years. The workforce optimists have recently chosen to focus on encouraging people to use their heads to develop innovations in government practices and not spent much energy on encouraging people to use their heads to make better decisions in their regular jobs. Since innovation, unlike decision-making, is not going to be a constant activity, eliminating the earlier focus on one-size-fits-all decision-making gives the discussion a less-central importance in organizational life.

One way out of a human judgment/one-size-fits-all dichotomy is provided by artificial intelligence, as discussed in my last blog. If you have enough data, decision recommendations can be developed through artificial intelligence that are adapted to the facts of the individual situation, avoiding one size fits all -- though not having people make the decision themselves. (It is possible of course to frame artificial intelligence solutions as mere suggestions or alternatives. However, enough data for AI is very often not available. Another a way out of this dichotomy is to develop some simple guidelines for a few kinds of situations where one approach should be used and ones where another should be. The Federal Acquisition Regulation tries to provide such guidelines to allow for both direction and discretion in whether to award a fixed-price or cost-based contract. However, there is a tendency in these kinds of situations, especially if the cultural expectation is one size fits all, for such guidelines to degenerate into one-size-fits-all rules.

In the absence of the kind of data that makes artificial intelligence or other computational solutions to decision problems possible, we are left with a choice between promoting, or even requiring, some version of one size fits all on one side or promoting the ability of civil servants in more situations to use their judgment, based on their experiences but with very imperfect information. Which way we go will depend on some pretty fundamental judgments about the knowledge and motivation of the government workforce. I suspect that deep down I am inclined towards workforce optimism because it more respects the opportunity for employees to be fuller human beings than an order that chains them to one-size-fits-all rules. Soloway deserves praise for bringing this question back on the governance agenda.