When I read Supervisor Zack Scrivner’s April 24 paean to Lean Six Sigma (“Turning to a new way to run county government”) as a management tool to help address the budgetary issues of Kern County, bells went off in the back of my head. A little history is in order.
Six Sigma is nothing new in systems management. It was developed by Bill Smith at Motorola in the mid-1980s. It was later adapted by GE, and its success in the business world caused it to spread to many industrial producers. Practitioners then combined it with “lean manufacturing” techniques to emphasize manufacturing quality and efficiency.
Three of the main salient features of Six Sigma are the focus on achieving measurable and quantifiable returns, personal emotional buy-in by management (hence such jargon as “deploy,” “empowering,” “strike teams,” “black belts,” etc.), and a commitment to making decisions on the basis of verifiable metrics, often statistics, rather than intuitive knowledge. And therein lies the rub.
Let us be frank. We have government provide services because the market, employing business techniques, is either incapable of providing those services or provides them in a way that is societally unacceptable. County government is not a business and provides services that are inherently resistant to metrics. For many of the services, measuring by verifiable metrics is positively unlawful, unethical and misleading.
Think of the area I am most familiar with, the courts and county lawyers. Who wants to incentivize county lawyers to efficiently produce lots of motions, appearances and interviews? Good metrics, bad policy. I would prefer lawyers who exercise good judgment and discretion, even if it is unquantifiable.
In fact, in a lot of areas we would all prefer fewer measurable statistics, even if that leads to government inefficiencies. I would be perfectly happy if firefighters had no fires to put out, but I want them in my neighborhood anyway. I wish no crimes were committed, but I like deputies patrolling. It would be wonderful if no inmates misbehaved, but corrections officers should run the jail. We would all prefer probationers to never violate their terms, but probation officers should still supervise. Who would complain if KMC had fewer sick people show up?
This all sounds a little silly, at first, but the point is that in most public policy areas success is not subject to efficiency analysis. How do you measure the bad things that don’t happen? How do you measure a community’s peace of mind? A little conceptual clarity and frankness is in order.
Contrary to Supervisor Scrivner’s jargon, county government has very few “value-added” programs. Almost all of the money goes to ameliorate social problems, maintain critical infrastructure, or just preserve a livable community. Law enforcement, fire, probation, corrections, medical care, public assistance, public administration, public health, sanitation — these are where the big money goes, and these are precisely the areas where a market efficiency analysis can do the most damage to real human beings.
Efficient manufacturing management processes may be great if you want to reduce your statistical expectation of defective features to .00034 percent. They aren’t so great if you want to ensure that the disabled elderly get the intense personal attention they need. They are positively destructive if you think that a neighborhood library branch creates a positive community ambiance.
Budget decisions are always tough for politicians; they constantly face the practical reality that there is an infinite demand for services, and finite resources with which to satisfy that demand. That being said, I really worry when manufacturing concepts are applied to human services.
On Dec. 28, 1957, Whitaker Chambers read Ayn Rand out of the conservative movement by pointing out that her fictional technocratic elite, supervising through a managerial political bureau, inevitably leads to dehumanization. That impulse remains, and it concerns me.