"Risk-Benefit Analysis" and Newton's Third Law of Motion

Every action (solution) has a reaction (consequences). It's the consequences we intentionally or unintentionally ignore which can cause the greatest harm.

My good friend “Visceral Adventures” wrote a great article recently on Problematic Probabilities.

I’m always happy to see one of her articles drop into my inbox. They are always insightful and entertaining.

The example discussed in her recent article is important because it is related to what is required to perform a good risk-benefit analysis.

I believe the reason we are in the mess we are in today is because people no longer know how to perform a risk-benefit analysis, let alone a good one.

We don’t know how to ask hard questions.

We don’t know how to “weigh-up” inconvenient answers.

And… it quickly becomes uncomfortable. Even potentially offensive. (Which explains a lot, because we are not allowed to offend anyone anymore. To be clear, a risk-benefit analysis isn’t offensive per say, it’s just that some people will take offense. Constantly self-censoring ourselves in order to not offend anyone, even when we mean well, will be our downfall. Being able to speak and express ourselves is an important process in forming our identity. If we no longer can do that, we become nobodies.)

The insight from her article is that people have forgotten (or haven’t been taught) about Newton’s third law of motion, which is crucial to performing a proper risk-benefit analysis.

“Newton’s third law states that when two bodies interact, they apply forces to one another that are equal in magnitude and opposite in direction”.

This might be obvious to some but it very counter intuitive to others. It even takes Engineering students a while to wrap their heads around the implications. (It’s hard to grasp that for every force that interacts with an object, there is an equal and opposite force interacting with something else. It’s unavoidable. However, many people don’t know or ignore that fact, and yet it is the “opposite force” and the “something else” that can end up having the greatest impact on their everyday lives.)

Metaphorically, this law is universal in its application:

Any action will result in one (or a multitude) of intentional AND unintentional reactions (or consequences).

Taking the time to think about what the unintentional consequences might be is extremely important.

For every proposed action (or solution) we must identify the expected reactions (or consequences) and brainstorm and discuss the unexpected ones. (We need to learn how to think about the things we know nothing about. Imagination and creativity are crucial here.)

Furthermore, every problem and every solution has a “dual”. (In the sense of a mathematical dual. My favourite is the controllability-observability dual. These are extremely beautiful and profound in their implications.)

We don’t really understand a problem, nor its solution if we don’t identify and fully grasp its dual. (i.e. In some ways the “reaction” in Newton’s 3rd law of motion is the dual of the “action”. One can not exist without the other. And yet too often we neglect to give it proper consideration.)

Not only does every “action” result in a “reaction”. There are in fact many possible “actions” to choose from. All of which will help us achieve the same desired outcome. Yet, all of which will result in different “reactions”.

The problem is two fold:

1. Not only do we neglect to properly analyze the “reaction(s)” to our “action”

2. We fail to properly explore alternative “actions” altogether

Example

Let’s image there is a perceived threat, and a perceived solution. (The threat is purely fictitious.)

The proposed solution is to manufacture “special” cookies with “special” ingredients that will protect people from the threat after they consume it.

Here are the odds:

• 970 out of 1000 people who consume the cookie will be completely protected

• 20 of them will get sick but fully recover (i.e. the cookie will not protect them from the threat.)

• 9 of them will develop a permanent debilitating condition after ingesting the cookie but will not die

• 1 person will die within a week after eating the cookie

(NOTE: These are honest odds. In a not so honest society we can imagine the cookie manufacturers, (who have a lot to gain from the sale of their cookies), could manipulate their trial results to show that no one died and that no serious injuries were recorded.

As a thought experiment, (which we know will require a lot of imagination): Under what conditions and assumptions would a cookie manufacturer feel justified to behave dishonestly? How could they get away with it?)

There are at least three questions we need to ask ourselves:

1. The “dual” principle: Do we understand the problem and the potential solution well enough that we have identified its dual? Manufacturing cookies for the entire world population is expensive. (Very expensive). Are there other things we could be spending our money on that would provide more benefit to society as a whole?

   • Exploring these possibilities is crucial in order to properly “assess” the proposed solution. In most cases better ways of spending money will become apparent. (But they won’t if no one takes the time to perform this exercise. And we can’t compare the “goodness” of the proposed solution if we don’t have other solutions to compare it to).

   • Being clear on our goals and objectives is also fundamentally important. (We can’t “assess” anything if we don’t have an objective. Choosing a good objective matters because it will affect our whole judgement system. It’s the yard stick we use to measure everything against, whether we realize it or not.)

   • When people are faced with the choice of forfeiting “universal healthcare” (as an example) for a “one off” cookie, the proposed “cookie” solution becomes much less attractive. (We now have a choice that allows us to “weigh-up” solutions against each other. This is crucial.)

2. The action-reaction principle: What are the unintended consequences?

   • Even though the odds were honest. Some unintended consequences only became apparent years later. The main one being that half the people who ate the cookie became infertile.

   • The odds we listed above are honest:

     • Would you eat the cookie if you knew you had “1 in a thousand” chance of dying? or had 9 in a thousand chance of being seriously and permanently injured?

   • Always look for the negative consequences, even when they are not explicitly stated or known. In fact, if there are none, it is a clear sign the “risk-benefit analysis” is seriously lacking.

3. The “correct” problem principle: Are we even addressing the correct problem? What is the severity of the threat?

   • The threat will only kill 1 out of 1000 people. Mainly the older vulnerable population.

   • Exposure to sun light was shown to reduce the risk of the threat by a factor of 100. (i.e. 1 out of 100,000 vulnerable people will die. The odds for healthy individuals is 1 in 10 million).

   • In light of this new information, would it change your assessment of the “risk benefit analysis”?

     • Withholding important facts such as this, is used to manipulate us.

4. Extras to ponder:

   • Have we identified all the assumptions?

   • Have we identified all the biases? (i.e. the conflict of interests for all the stakeholders).

   • Have “stopping-conditions” been included in the analysis? (They must be included.)

   • Have feedback loops been designed and discussed? (What is the target? What will be measured? How often will it be measured? How will the trajectory be adjusted when it get’s off target?)

Questions

The questions below are the bare minimum we should be asking ourselves for every new threat and proposed solution we are faced with moving forward. (Let us know if we missed an important one in the comments.)

• The risk of the threat:

  • How fast does it spread?

  • How serious is the threat?

  • How long will it last?

  • Does it affect everyone equally?

• The risk-benefit of the solution

  • Is it safe?

  • Is it effective? ( Both from a temporal and magnitude point of view. How fast does it act? How long does it last? Is the protection complete?)

  • What are the consequences? (i.e. the “reactions”. They ALWAYS exist whether we correctly identify them or not.)

  • Does it work the same for everyone?

  • How much will the solution cost to implement?

  • Could the money be used to achieve a better OVERALL outcome in a different domain?

  • What are the alternatives (There ALWAYS are. A proper risk-benefit analysis can NOT be done without them. Period. We must demand to see them and they need to be properly analyzed as well.)