Subject: Is this a causal loop?

Posted by John Gunkler on 3/9/2004
In Reply To:Is this a causal loop? Posted by Steve Kipp on 2/27/2004

Message:

Really helpful thoughts from Steve Kipp. Thanks.

Just a general comment, then its application to Steve's explanation of the interest-earning bank account.

According to Jay Forrester, and the logic of stocks and flows, one never needs auxiliaries or converters (like Interest Rate) in any SD model. They can all be included within the rate equation. So, why should we ever use such things (and it's often, I admit)? Only when making them visible will help people better understand the dynamics.

In the case of a bank account, it is not the interest rate that creates the accumulation of "Money in the Bank," it is the flow of "interest payments."
If interest payments are more than zero, "Money in the Bank" will grow. It really doesn't matter why the interest payments are more than zero. Sure, we (the all-wise SD modelers) know that the interest rate determines whether interest payments are zero or positive (and how much they are), but if the point is simply to show how reinforcing feedback works, it may be better not to include "interest rate" as a separate item. Simply diagram one stock and one flow, with the size of the stock being a positive influence on the amount of the flow. [You only need to make "interest rate" visible when you're trying to understand why "interest payments" are as big or small as they are or, equivalently, why "Money in the Bank" grows at the rate it does.]

This re-absorption of auxiliaries into rate equations is a trick that I believe lies behind some of the best advice I've got on creating SD models.
As Barry Richmond described, when people first attack a dynamic problem they're inclined to engage in "laundry list thinking." We create a large list of "factors that influence" the outcome we're interested in understanding. Many of these are exogenous and may or may not truly affect the dynamics -- a point Barry made so well -- but some are also "factors" in the sense that they affect the rate equation. Until we focus on what is accumulating and what is flowing into and out of the accumulations, we'll be at sea. I try to put all of these auxiliaries to the side (to contemplate later when I'm writing the rate equations) in order to keep the model as simple as possible at first.

I think the same advice applies well when we're teaching others to think dynamically. Keep it simple by focusing on just the stocks and flows, at first.

Follow Ups:

Is this a causal loop? - Steve Kipp 3/11/2004