Latticework of Mental Models: Feedback Loops

Few weeks back, Vishal was conducting the Art of Investing workshop in Bangalore and he asked me to speak on the subject of human behaviour and how it relates to investing.

Now, unlike him, I am not used to speaking in front of a large audience. Although it was a room full of around 50 people, but for me anything more than two is a crowd. So to say that I had butterflies in my stomach would be an understatement. It felt like flying horses in my belly.

Finally when Vishal handed over the mic to me, one part of my brain was trying hard to control my shaking legs, the other part was busy adjusting (with trembling hands) the microphone attached to my shirt.

Just when the first word came out from my mouth, a loud screeching noise echoed in the room. For a moment I thought that my heart, which by that time was pounding hard against my chest, had come out. Then I realized that the loud noise was coming from the speakers. I immediately cut it by covering the mic with my hand.

It’s a pretty common problem (sound from the speakers feeding back into the microphone, and then going in an unending loop between speaker and mic, often causing a loud squeal) and most of you must be familiar with this kind of audio feedback noise. This acoustic problem can be easily corrected by tuning the instruments and using few filters. But what’s fascinating about this feedback loop phenomenon is that it pervades almost every other natural as well as man-made systems.

So let’s explore this mental model i.e., Feedback Loop, and see how it can help us understand the world around us.

Here is a working definition from Wikipedia –

Feedback occurs when outputs of a system are routed back as inputs as part of a chain of cause-and-effect that forms a circuit or loop. The system can then be said to feed back into itself.

In other words, in many systems, the output re-enters the system as another input. So a feedback loop is like an input, but its origin is from within the system itself, not from outside the system.

Positive and Negative Feedback Loops

As we saw that systems adjust in response to feedback. Feedback Loops can be positive or negative. A positive feedback amplifies an effect, while a negative dampens it.

A thermostat is an example of negative feedback. In fact, the human body has an in-built thermostat. The tendency of human body, and for that matter all living beings, to maintain orderly conditions, including body temperature, is called homeostasis.

Victor Niederhoffer, in his book[1], elaborates on this:

In system theory, it is called negative feedback … A common homeostatic behavior in humans is temperature regulation. If the temperature rises above the 98.6 F optimum for normal human activity, sensors on the skin detect it and signal the brain that a rise has occurred. The brain relays the information to the effectors that increase blood flow to the skin. This induces perspiration….[which on] evaporation lowers the body temperature. When the body cools below a certain point, a comparable mechanism is set off, this time reducing blood flow and causing shivering. This activity generates heat through physical activity thus raising the body temperature.

A negative feedback loop usually keeps the system in equilibrium whereas an uncontrolled positive feedback loop can cause a system to reach a critical state, a tipping point[2], and then fundamentally change it.

If you want to see positive feedback in action, stock market is great place. We will come back to stock market shortly.

Learning and Feedback Loops

Human behaviour can be improved to a great extent with the power of feedback loops. Experiential learning is one area where feedback loops play a terribly important role.

In fact, the whole philosophy of deliberate practice is strongly based on the process of getting frequent quality feedback.

Elon Musk says –

I think it’s very important to have a feedback loop, where you’re constantly thinking about what you’ve done and how you could be doing it better.

In investing and in life, what makes all the difference is an ability to make right decision. Right decision may not guarantee the desired outcome every time but they increase your odds of success. Perfect decision-making is next to impossible, and what matters more is forward momentum and a tight, fact-based feedback loop to quickly recognize and reverse bad decisions.

We have to constantly incorporate feedback generated by our decisions and keep improving the process.

Let me take a moment of yours and share my personal experience.

My primary purpose for writing the Latticework series, as I have been claiming in almost all my previous posts, is to enhance my own learning. Now if you think about it, the feedback loop model is very much applicable in my learning process also. How?

In order to write, I need to read a lot. The more I read about a chosen topic, the more insights and ideas I stumble upon which results in an ever increasing list of topics to learn (and write) about. So more reading results in more writing. A virtuous cycle of learning, a nice feedback loop.

Complexity of Feedback Loops

So far, we have seen examples where the feedback produces somewhat predictable outcome (positive or negative). But in many cases, especially in complex adaptive systems[3], feedback loops can result wild unpredictable swings in system behaviour.

George Soros writes in his book –

I envision reflexivity as a feedback loop between participants’ understanding and the situation in which they participate, and I contend that the concept of reflexivity is crucial to understanding the situation that have thinking participants. Reflexivity renders the participants’ imperfect and ensures that their actions will have unintended consequences.

What it means is that changing one variable in a system will affect other variables in that system and other systems. In systems such as economies, stock markets etc., behaviour often changes as a result of the behaviour of others and experiential learning (feedback loop). These systems therefore become difficult to predict as unexpected (emergent) properties may emerge.

It is unlikely therefore that history will exactly repeat itself as participants keep modifying (learning with feedback loops) their behaviour.

So it’s important to make the distinction as to what kind of systems are we dealing with. Is it a system prone to cascading effects of positive feedback loops? Or is it a system which tends to remain at equilibrium by virtue of negative feedback loops? Or is it a complex adaptive system prone to unintended consequences

Here is a nice video illustration about different types of feedback loop systems.

While trying to predict the behaviour we must not only consider particular elements of a design, but also their relation to the design as a whole and to the greater environment.

Another important factor, while thinking about feedback loops, that needs to be considered is time lag between a feedback and its effect on the system. Feedback is a circle and it takes time to travel round a circle. This means that effects can appear some time after their cause.

In their book The Art of Systems Thinking[4], the authors, Joseph O’Connor & Ian Mc Dermott, write –

Think of appetite as part of a feedback loop…there is a time delay between the stomach being full and the sensation of satisfaction…the feeling of fullness is not directly related to how much food is in your stomach at that time, but to how much food was in your stomach several minutes ago. This delay between being full and the sensation of being full means you may continue to eat past the satisfied stage to the uncomfortable stage. If you are still eating when you feel full, you have gone too far. The way to avoid this is to eat more slowly, chewing well to speed the digestion of the sugars. Give yourself time for the feedback to appear.

Not being aware of this time delay can lead to faulty understanding of cause and effect in a system. When we do not take time delays into account we evaluate the success of our strategies too soon, long before the full consequences are observed.

In Business and Investing

The heart of value investing and long term wealth creation is the idea of compounding.

What makes compounding interesting is that every small incremental change (interest on principal amount) goes back in the system (the principal amount) and becomes the part of it which in turn generates a slightly bigger feedback (bigger interest). Allow this to happen for a long time and what you have effectively is a positive feedback loop in action, a snowballing effect.

On the contrary, dancing in and out of the market in response to every wiggle in the stock prices is akin to creating a negative feedback loop for your wealth. The frictional costs (taxes and brokerage) coupled with interrupted compounding brings back your system (wealth) into a stable (non-increasing) equilibrium position.

So what do I do when my broker pushes me to trade? I look at it as a trigger for negative feedback loop and I simply ignore him.

While trying to understand a business, this mental model can be of tremendous use. If you can figure out the kind of feedback loops operating for a business or an industry, it can give you invaluable insights about the company.

Look at the Berkshire Hathaway’s (BRK) model. The humongous cash generated by BRK’s businesses is in the hands of world’s best capital allocator, Warren Buffett. He uses this cash to purchase other companies which in turn produces earnings which are then used for buying other companies. This feedback loop has been happening at Berkshire for a very long time.

Prof. Sanjay Bakshi, in his interview[5] with Safal Niveshak, mentioned –

Vicious circles can turn into virtuous ones and if you can get in before they do, on very favourable terms, then good things should happen, if you keep on doing it repeatedly.

A vicious circle is positive feedback loop in a negative direction and a virtuous cycle is a one in positive direction causing snowballing effect.

You can easily visualize a positive feedback loop in action during bull (or bear) markets. The stock market falls causes a sell-off. This creates a ripple effect of further sell-off and price declines. The opposite occurs in a stock market bubble.

I found an interesting insight about nested positive and negative feedback loops in this[6] email exchange –

…you’re on the right track when you visualize a positive feedback loop as a mechanism which is nested inside a negative feedback loop. To illustrate, why do bear markets follow bull markets? Because over the long run, markets operate inside a negative feedback loop with built-in corrective mechanisms. When prices run too far away from underlying values, there are forces that pull them back.

So over a long term the stock market shows characteristics of negative feedback loop i.e. the markets gradually go upwards with periodic corrections but in short term, especially during bubbles and crashes, it seems to be operating under the influence of positive feedback loops.

There are many other aspects of financial market which can be evaluated using the lens of feedback loops. Legendary value investor, Seth Klarman, used this mental model to explain the fallacy of indexing[7]. He says –

I believe that indexing will turn out to be just another Wall Street fad. When it passes, the prices of securities included in popular indexes will almost certainly decline relative to those that have been excluded.

More significantly, as Barron’s has pointed out, “A self-reinforcing feedback loop has been created, where the success of indexing has bolstered the performance of the index itself, which, in turn promotes more indexing.”

When the market trend reverses, matching the market will not seem so attractive, the selling will then adversely affect the performance of the indexers and further exacerbate the rush for the exits.

Conclusion

I have been extremely fortunate that I have this opportunity to expose my thoughts to the Safal Niveshak Tribe. This is turning out to be my greatest learning experience so far.

You can also harness the power of feedback loops for your learning just by a click of a button. I mean the comments section below. 🙂

Take care and keep learning.