Create a Model

From SystemsWiki

There are numerous references that provide guidelines associated with various aspects of creating Causal Loop Diagrams (CLDs). Though what none of these references seem to provide is actual guidance as to how to go about identifying the elements, their relations and actually developing the model. In this article we're going to use the four primitives, stock', flow, variable, link, to develop a model, while pointing out the relevant questions and conventions employed along the way.

Getting Started

There are two points to remember as you're beginning:

• It doesn't matter where you start because all the pieces are related in one manner or another. As such, with the appropriate rigor applied to the development of the model, you should end up with an appropriate model regardless of where you start.

• You're developing a model, a simplification of reality to promote understanding, of some perceived situation. You are not developing a model of "the system". The problem with developing a model of "the system" is that the model will tend to become overly complicated and make it difficult for the model to fulfill its intended purpose; to promote understanding.

If you're developing a model to promote understanding of a perceived situation then a great place to start might be to describe the perceived situation. And, it's best if you describe the perceived situation as it has unfolded over a period of time rather than just the current state. Formulating Questions might be an interesting read at this point.

OK, so here's the description of a perceived situation we can investigate in a rigorous manner.

If one can put money in an investment account and it grows over time, and it grows even faster with regular deposits, why aren't more people rich and ready for retirement? I've started numerous retirement programs through the years though for one reason or another they've all evaporated in time. What is the basis of this sad state of affairs?

Now we have a place to start. The previous description identifies two elements; an investment account and deposits. Let's start with Investment Account. This sounds like a quantity of money so it must be a stock, which we'll label Investment Account.

Traditional wisdom says that when naming stocks you should use a noun, and a positive and directionless one at that, and you'll see why later. Also note that nothing changes in a stock unless it is increased by an inflow or decreased by an outflow. Neither hand waving nor smoke and mirrors have an effect within a model.

A deposit is something that increases the value of the Investment Account so it must be a Flow into the Stock. Fig. 2 depicts deposit as a Flow into the Stock. That the Flow is blue is a personal convention. The Flow is labeled as a noun; a positive one at that. The "regular" modifier was dropped because, while it would be nice, the deposit doesn't have to be "regular". If deposit increases it adds to Investment Account. If deposit decreases it still adds to Investment Account, just not so rapidly.

Initially deposit begins in a cloud as we're not concerned with where the deposit comes from, though this could change as we continue to develop the model.

We're now at a point where we've exhausted what was provided in the situation description. To progress from this point we have to think and ask questions. Though what are the appropriate questions? There are really only two questions.

• What else affects what's already part of the model, and

• What else does the existing parts have an effect on; that's relevant to the situation we're considering?

And, it's pretty much the same questions over and over and over and over. Get the point? And don't forget the part that says "that's relevant to the situation we're considering". If you forget this part you end up modeling forever because things endlessly connect. Continue adding primitives to the model that are necessary for you to develop the appropriate understanding, and STOP when what you have developed is sufficient to support that understanding.

Generally an Investment Account pays some amount of interest based on an interest rate and the amount of money in the Investment Account. In Fig. 3 we've added interest which is a flow and adds to Investment Account and the interest is a function of Investment Account and interest rate. At this point interest rate is indicated to be a constant (which isn't really true though we'll start there).

The "R1" notation is an indication that the loop (Investment Account -> interest -> Investment Account) is a reinforcing loop. Whether a loop is a balancing or reinforcing loop can be determined by counting the number of "-" signs around the loop. If the number is 0, or even, then it's a reinforcing loop; an odd number and it's a balancing loop.

Admittedly the Investment Account could be a stock portfolio in which the stocks periodically pay dividends or it could be a mutual fund with variable return on investment, though for now we've just chosen to use interest as in a savings account.

If you get to a point where you can't quite figure out another relevant connection asking others for input is a wise move. I find it continually amazing where insights come from at times.

At this point we have an Investment Account that continues to grow based on two additive flows. Though there must be an outflow otherwise we'd not have the question posed at the beginning.

In Fig. 4 Withdrawals has been added as the means by which the Investment Account is decreased. That withdrawals is red is a personal convention making subtractions easier to spot in a model.

At this point there seems to be several aspects of this model which directly relate to the initial question:

• Starting the Investment Account is a necessary first step, otherwise nothing happens.
• Choosing an Investment Account with a good interest rate will ensure it grows faster.
• The larger and more frequent the deposits the faster the Investment Account will grow
• Withdrawals are contrary to the intent of the Investment Account.

As for the first item, it's just a matter of choice. If you're able to stay healthy and just willing to work until you die then you probably don't need an Investment Account, though this isn't the option most would prefer. As such, a long term perspective would seem to imply that starting the Investment Account would be a good idea.

Regarding finding a good interest rate, it depends a lot on one's tolerance for risk. If you're looking for a guaranteed return, with no possibility of loss, then you have to settle for a lower interest rate. If you can tolerate the possibility of losing your money then certain stock investments have a high yield potential, along with the potential of loss. And, there are lots of vehicles in between these two options.

Deposit was initially placed in the diagram with a cloud as its source indicating the source of the deposit wasn't relevant. You know the deposit has to come from somewhere that's relevant to the person making the deposit, and that would be you. Fig. 5 adds some clarification around the source of the deposit.

The money available for deposit is the difference between income and expenses, which usually turns out to be about zero.

In order to encourage people to make deposits, at least in the US, there are types of accounts into which you can put pretax dollars. Depending on one's tax bracket this could be a 15% to 35% bonus. And, one can set it up as a direct deposit so one never sees the money, and therefore avoids the risk of spending it. Also one isn't taxed on the interest until they take the money out of the account.

As an additional discouragement to withdrawing money from the Investment Account there are penalties incurred if one removes the money before reaching the age of 59 1/2. Though even with the disincentives for withdrawal the attractiveness grows proportional to the Investment Account. Fig. 6 adds elements to the model to depict these aspects of the model.

We now have a model which provides some incentives to start and continue to deposit in an Investment Account, and some disincentives toward the withdrawal of funds, though have we really addressed the initial situation posed? Not really. As far as starting the Investment Account and regularly depositing money, there are incentives, and for many these incentives were enough to get them to invest. For many the incentive, for one reason or another, has not been sufficient. And, any more strict incentives would likely be looked on unfavorably. People do not like to be manipulated, even when it is for their own benefit. The penalty for withdrawal is a deterrent in some respects though as the Investment Account continues to grow its attractiveness in terms of what it can purchase continues to entice. The best answer for this situation is to legally tie up the withdrawal process so it's only an option in the case of dire emergencies. Though as much as people find being manipulated by others distasteful, being controlled by themselves is just as distasteful.

Is the model done? As usual, the answer is; "It Depends!" If it has provided sufficient understanding to address the situation posed then it is sufficient. If not then it should be taken further, though once it is sufficient you should STOP!

Summary

There are several items it is hoped that you will take away from this article:

• A Model is a simplification of reality intended to promote understanding. If the model does that then it's a good model, otherwise work on it some more. All models are false, some models are useful.

• The situation under consideration should be described in terms of how it has evolved over a period of time, and possibly how it's expected to evolve in the future.

• A model should be developed only for those interactions which are relevant to the situation being considered. Model what is necessary and only what is sufficient to address the situation described, and no more.

• Using the stock, flow, parameter, and link primitives causes one to think more explicitly than when using just parameter and link primitives to construct a model.

• Any influence which is the basis for a change must be explicitly represented (no smoke and mirrors or hand waving allowed).

• There are some good guidelines which improve the model's potential for promoting understanding.

• The resulting model is qualitative. There may be times when the real implications of the interactions are simply not intuitively evident even after the model has been developed.

• For a quantitative understanding of the interactions within a model one needs software, such as Insight Maker, that enables a simulation of the model with explicit values.

Additional Examples

• Unleashing Understanding: The Essence of AND
• Melting the Tip of the Iceberg
• Project Systems: Distinguishing Fact from Fantasy
• Outline for a Morphology of Modelling Methods: Contribution to a General Theory of Modelling by Tom Ritchey
• Model Building

References

Additional Resources Systems Thinking World Discussions Systems Thinking World Q&A * Gene Bellinger