Saturday, April 25, 2020

Thinking in Systems

By Donella H. Meadows.

Systems exist wither we like it our not. From Meadows perspective, everything is part of a system. The end of one system, and the beginning of the next is an arbitrary point that we choose for our convenience, and not because that's the actual reality.

When you analyze a system, don't zoom in too much and loose perspective, or zoom out too much and end up with something academic. Scale the system definition to the size of the problems you are engaged with.

Problems and the way out...

I like the way that Meadows calls out "Problems and the Way Out" not "Problems and Solutions." The solution to many system solutions isn't a black or white flip of the switch, but a lot of work to slowly change the path of the system.

Problem-- Policy Resistance-- when the actors try to pull the system in a direction other than the desired new goals.
The way out-- Let go. Bring in all the actors and use the energy formerly expended on resistance to seeking mutually satisfactory ways for the goals to be realized.

Problem-- the tragedy of the commons.
The way out--
1. Educate and exhort everyone so they understand the consequences, or
2. Strengthen the feedback link, regulate, or,
3. Privatize

Problem-- Drift to Low Performance
Allowing future performance to drift based on past performance, especially if there is a negative bias in perceiving past performance.
The way out-- Keep performance standards absolute

Problem-- Escalation. There is a reinforcing feedback loop carrying the system into an arms race.
Solution--
1. Avoid getting into this trap
2. Refuse to complete
3. Negotiate a new system with balancing loops to control the escalation.

Problem-- Success to the successful. Winners have a reinforcing feedback loop that allow them to keep winning more.
The way out--
1. Diversification. The system has to support diversification.
2. Strictly limiting the faction of the pie any one winner my have.
3. Policies that level the playing field. Policies that devise rewards that do not bias the future competition.

Problem-- Shifting the burden to the Intervenor. Superficial solutions that do not solve the underlaying problem.
The way out--
1. Avoid getting into this situation.
2. Take the focus off short term relief and put it on long term restructuring. Do so quickly. The longer you wait, the harder the withdrawal process will be.

Problem-- Seeking the Wrong Goal.
Systems are very sensitive to the goals of the feedback loops, and not necessarily the goals of the designers.
The Way out-- Specify indicators and goals that reflect the real welfare of the system. Be careful to not confuse effort with result or you will end up with a system that is producing effort, not result.