Creative Concepts
Invention Ideas

Economic Theories
Political Theories

Deep Thoughts
Story Ideas

New Business Ideas
Problem Solving

What Are Self-Correcting Systems?

What happens when you swing forward? You then swing back, of course, unless you fall off the swing. Eventually you stop right at the bottom of the arc. And when you drink too much alcohol you pass out, therefore stopping the ingestion of more. The hangover that results might even make you not want to drink for a while. Self-correcting systems like these are systemic processes which naturally return toward a "normal" state when they have moved away from that state by a significant distance or degree.

For an example from nature I can look out my window. When it gets too hot in the afternoons in Colorado (where I first wrote this: now we live in Florida), the clouds form from the hot air rising up over the mountains and cooling, and the rain comes, reducing the temperature. This general principle operates in small and large ways, and over short and long periods of time. In arctic areas lemming populations rise until the over-population causes an increase in foxes which feed on them and the lemming population crashes and the whole four-to-five-year process starts over. But despite the many examples, scientists and others often forget to take into the possibility or reality of this long-term tendency toward self-correction in some systems.

For example, when Paul Ehrlich’s book, "The Population Bomb," came out in 1968, it opened with the following lines: "The battle to feed all of humanity is over. In the 1970s and 1980s hundreds of millions of people will starve to death in spite of any crash programs embarked upon now." Erlich later wrote in Ramparts Magazine, "Millions of people will soon perish in smog disasters in New York and Los Angeles… the oceans will die of DDT poisoning by 1979… the U.S. life expectancy will drop to 42 years by 1980, due to cancer epidemics."

His mistaken predictions were based in part on straight-line projections of the type scientists make too often (or perhaps it is the cause-advocates who do this more than true scientists). The process goes something like this: you plot data, identify a trend, and assume it will continue in the same way. Of course, that ignores the changes which might be made by random chance, or human intervention, or those that might occur naturally in the case of self-correcting systems.

Seal populations in a given area, for example, may grow in numbers in a neat slope on a graph, making it look as though there will be ten times as many of them in just a decade. But at some point the availability of prey is diminished by their own feeding in ever larger numbers, and so more seals begin to starve and the birth rates decline. That's when you see a population crash on the graph, and the cycle repeats itself all over.

In human economics we can see the same kind of tendency of systems to self-correct. Real estate prices, for example, start rising faster at some point, and ever more people buy homes not just to live in, but as investments. Builders looking to cash in on this trend build more and more houses. Banks start making it easier and easier to borrow to buy a home so they can process more loans. Then, when most people have bought a house, and there are too many still being built, and prices are getting too high for the average incomes families have, the whole thing crashes and prices fall. There are other factors, and this particular phenomenon is not always as dramatic as the real estate bubble that popped in 2006, but the cycle is evident when you look back over the years.

Still, when we left in Tucson in early 2006, investors there believed that prices would continue to go up forever. If you looked at the charts of prices in the previous few years it was hard to argue against the idea. But they were making a common mistake. Suppose a man is depleting his savings at a steady rate and so an "expert" predicts bankruptcy in two years based on a straight-line projection. But of course as he gets nearer to this fate, his fear of losing everything could become stronger and more motivating, so he might make the necessary changes in order to avoid going bankrupt. If not, the bankruptcy itself forces changes, so his spending will not continue to rise or even stay the same. It's tricky at best to use existing trends to predict the future when we are dealing with self-correcting systems.

It was 46 degrees Fahrenheit outside my window this morning at six o’clock, and, as I write this around noon, the temperature has risen to 70 degrees. That means the temperature is rising at a rate of four degrees per hour. At that rate, within a few more days the temperature will be over 450 degrees and trees will be bursting into flames. That may sound silly, but similar straight-line projections are sometimes used when studying self-correcting systems or systems that have other influences which will affect what actually happens.

If you are interested in the concept, and want to explore it further, here are two questions to get you thinking:

1. What other examples of self-correcting systems can you think of?

2. If straight-line projections do not work well with such systems, how do we make predictions about them that are more accurate?

If you like what you see, please share...


Self-Correcting Systems