Why we aren't the "profit only" animals we think we are
Suppose I was given $1.000 by someone with the only condition that I should share it with you. I would be totally free to choose any division I like and make you an offer. In turn, you would be able to either accept or decline that offer. If you accepted the offer I made, we would split the money as I chose, if you refused both of us would walk away empty-handed. Now, assume I came to you offering that I keep $900 and give you $100. Would you accept?
Most people presented with this deal refuse it. Now what is interesting is to look at why. From a perfectly financial point of view, refusing the offer isn't strategic. The BATNA (Best Alternative to a Negociated Agreement) here is $0 if you refuse the deal. Thus, if money were the only element people evaluated when making decisions, everyone would agree to the deal up to the $999.99/$0.01 division. That's not the case. Which division people agree to varies from individual to individual with factors such as economic status, cultural background, age and others impacting the decision, but few people agree to the $900/$100 division.
Two factors come into play here. The first being the perceived gain and the second the representation of fairness.
The perceived gain reflects the fact that humans do not process absolute values and proportions in the same way (see Jacob, Vallentin & Nieder's 2021 paper Relating magnitudes: the brain's code for proportions). When asked whether you would accept a $900/$100 refusal is likely. Asked if you would agree to a $9.000/$1.000 the refusal is still probable. However, if presented with $90.000/$10.000 or $900.000/$100.000, would you still refuse it? Chances are your willingness to accept the deal increases with the absolute value of the part you are offered. While the proportion is strictly identical, refusing $10.000 or $100.000 is harder than refusing $1.000.
Now the second more interesting feat is the fairness factor. The fact that varying proportions can alter our response is real, but if we think about it our response should be a straightforward "yes" for anything offering us more than our BATNA, i.e. any amount superior to $0. Game theory developped in its modern form by Von Neumann in the 1940s concerns itself with the modelling of decision making of actors and posits that in any given situation, any given actor will always act according to its best interests. But how does that equate with the response to this offer? Why would people turn down an offer that gives a better expected return than their BATNA? Because money isn't the only factor at stake here.
The question of the place of fairness in economy isn't a new one. In 1986, Kahneman and Knetsch raised the question in their paper "Fairness and the Assumptions of Economics", demonstrating the question wasn't a feat of the 21st century. Similarly in their 1999 paper "A Theory of Fairness, Competition, and Cooperation", Fehr and Schmidt looked at how the economic environment impacted the type of strategies that dominated. Even accross species, non-human primates have been shown to develop responses to certain instances of inequity (see Price and Brosnan's 2012 paper To Each According to his Need? Variability in the Responses to Inequity in Non-Human Primates).
So what does our example suggest? If you conduct the experiment with your friends and colleagues, you'll find that the common answers for declining the $900/$100 converge along the lines of unfairness. Most people simply don't accept that such an unfair deal be imposed on them, especially taking into account that the deal-offered (the unfair one) would be much better off than them. Which means that when conducting negotiations and engaging in decision-making regarding financial quantities, economical value is not the only concern. Other elements are taken into account when making the decision, including fairness.
Recommended by LinkedIn
Does this mean game theory is dead? From the excellent game-theory course I took with Bruno Ziliotto in Ecole Polytechnique, I forged a strong belief in the game-theoretic framework to model decision-making, which makes me biased in its favour. I would thus argue that the game-theoretic approach can still account for the motivations of such choices as long as the models do too. And this is where most people fail to understand game theory. The framework isn't designed to model human behaviour as heartless robots with only profit as drive and no notion of ethics or morals. It is designed to represent the coherent and rational course of actions given a model of the actors. And that's where the subtlety lies. Game theory assumes the model of the actors it's fed is accurate. The core stake for the true modelling of economical actors is thus to take into account elements of social psychology such as the perception of fairness.
In a world where profit is often pictured as the only sensible goal to be pursued in a free market, it is important to raise the question of its coexistence and inter-dependence with other factors such as fairness. I encourage the readers to read more extensively on the subject of whether fairness matters in business and economics, a quick research with those keywords on google scholar should provide the needed help, but I enclose a few recommended readings for those interested.
Recommended readings