Outcome versus process
In any public-private partnership, the government will seek to accomplish certain tangible public policy goals through the imperfect mechanisms of motivating the private sector. In such a program, the government regulator will be seeking to measure the private partner’s performance, and then to use the resulting measures to incentivizes (reward and punish) the partner’s performance.
In this, government can choose from among two competing theories of measurement:
- Process measurement: did you follow the procedure?
- Outcome measurement: did you deliver the results I wanted?
That these two are quite different is self-evident, but often when government is establishing a program it conflates them, under the plausible but naïve theory that government knows the right process, and that this process will always produce the desired outcome. Conflating the two is bad because:
1. It prevents the private sector from innovating a better or cheaper process (since that is now a violation of the rules).
2. If reality has diverged from the baseline, the prescribed process will fail.
“
3. Should a bad result occur,
… the private sector partner will defend itself by arguing that the procedure was deficient. The resulting administrative dispute rapidly diverges into a you-did-I-didn’t debate about what was done rather than about how to fix it?
“Off with their heads!”
4. Goldbricks and shysters (yes, every business has them) pour their creative energy into figuring out how to maximize profit by minimally complying with the rules, rather than by producing better outcomes.
5. As a result, administrative guidance continually increases (the law of administrative entropy).
“As you can see from Section 32-12(d) second of Chapter 32 of Handbook 4350.1 …”
The law of administrative entropy
As a program ages, the rules governing any particular program will always grow in scope and complexity. Pruning them requires conscious effort and seldom accomplishes as much as it should.
The original 1913 tax code was 2 pages long. Today it is 6,000 pages long.
And its corollary:
Regulation follows catastrophe
The acts establishing the Securities and Exchange Commission (SEC) and giving it jurisdiction over new issues and resales of stock were passed in 1933 and 1934, after a certain Stock Market Crash.
Spectacular catastrophe (fiasco, scandal) often provokes over-regulation.
Thus, a simple moral: Outcome compliance is better than process compliance
When you go to a restaurant, you apply outcome compliance:
- Outcome compliance = did the food taste good?
- Process compliance = did the chef follow the recipe?
Other important lessons regarding outcome compliance:
1. Good systems have both upside and downside; but more upside. Human behavioral psychologists have discovered that when teaching animals, children, and affordable housing developers, the proper ratio of praise to blame is 5:1.
2. When designing incentives, money is the best plus, getting fired the best minus. It’s really very straightforward. Embarrassment and exposure are also powerful downsides, and public praise and recognition are terrific upsides. Compliments cost nothing and work wonders. So does threatening to publish a list of tax cheats in the local newspaper.
3. Measure early, measure often. A steady, frequent, and reliable stream of information is essential.
“I’ll finish my coffee before checking.”
4. Act immediately, act consistently, and scale the response. An administrator is only as good as the last decision. Game theory (as tested in Prisoner’s Dilemma competitive simulations) shows that tit-for-tat, prompt and commensurate direct response, is critical to sustaining credibility. This works in geopolitics too.
“Take that!”
5. Rules you don’t enforce are worse than useless. How many security council resolutions does it take to disarm
“Let the punishment fit the crime, the punishment fit the crime.”
The corollary to this one is, never threaten, never bluff. Do or do not. There is no try.
6. When trying to pull an underperformer into the fold, (a) always allow for absolution, and (b) always have something worse to escalate.
“I’ve got a little list! I’ve got a little list!
Of society offenders who might well be underground
And who’d surely not be missed!
Who would surely not be missed!”
