Datafy that Nudge

 

In the emotional world ruled by story telling (Mackay in Madness of crowds); psychology and emotion remain the new bestsellers. This, of course, is a change from the economists who were forced to concede ground to the psychologists after a few hundred years when Daniel Kahneman became the first psychologist to get a Nobel Prize for economics in 2005. The Psychology story is slowly gaining ground with the fundamentalists and technicians. A host of economic historians like Joseph Schumpeter (Kitchin, Benner, Juglar, Kuznet) came and went away explaining the structure underlying the story. But the intra-discipline push and pull continues. Coming to look at it closely, it seems like a full circle. Louis Bachelier took the bell curve and fitted it to finance; first 60 years his work was ignored, later he was celebrated and recently sidelined just because the bell curve he used seemed to be fatter at the tails. Pareto, the father of microeconomics introduced statistics to economics. Pareto’s 80-20 principle is an unchallenged universality but still ignored by the main stream.
The statisticians might just be ready to take the baton back over. In any case even if they do (statisticians, mathematicians) and find a few more universal laws, or redefine them, like they did a few 100 years back, that will not change or stop the society from listening and believing to cause and effect stories. The society loves them, especially the ones which they can relate too. This is why we have a market, a buyer and a seller of stories, eating into the popularity pie, turn by turn, irrespective of content.
This is not my subjective opinion, this is history, which is, fortunately (unfortunately) bigger than societal memory. A month old HBR Blog update by Justin Fox summarizes my view very well. “The idea of a paradigm shift comes from Thomas Kuhn’s 1962 book The Structure of Scientific Revolutions. Kuhn, a physicist turned philosopher of science, had spent a year in the late 1950s at the then-new Center for Advanced Study in the Behavioral Sciences at Stanford and been struck by how the assembled psychologists, economists, historians, sociologists, and the like often disagreed over the very fundamentals of their disciplines. Physicists, in his experience, didn’t do that. This wasn’t because they were any smarter than social scientists, Kuhn concluded. It was because they had found a paradigm within which to work.”
I don’t necessarily agree (I have my share of disagreements) with the physicists part regarding the framework, but I think Kuhn’s opinion about intra-subject conflicts (inter-subject opinion is more divergent) clearly illustrate how subjectivity is all around us; a better story. Thaler’s and Sunstein’s Nudge is another good story; a psychological story; which I also like. As I think it’s helpful in decision support for individuals and the society at large. It’s the underlying assumptions, which I feel looks at psychology as a primary cause standing on its own, never requiring a framework to rest on.
What is Nudge?
Nudge is a mild psychological stimulus. Nudge is not a mandate. It’s a delicate approach towards the society which abhors commands. It’s a way to help society and people make better decisions. It is definitely a noble cause, very actionable, very popular. A few examples; clocky is an alarm clock that runs away and hides if you don’t get out of bed on time; the black house fly etched in urinals across airports to reduce spillage; Gambling self-bans where gambling self-addicts put themselves on a list that bans them from entering casinos or collecting gambling winnings; stickk a website that helps in sticking to goals; no bite nail polish; donating organs; saving the planet; save more tomorrow; avoiding the right to sue medical practitioners; Texas anti-littering campaign and recently It’s Colombia, NOT Columbia campaign.
Is there a Universal framework societal behavior stands on?
According to the authors, what is true for diet is true for other risk-related behavior. In other words, what’s true for your mortgage also works for your addiction for soft drinks. This means every decision making is connected, be it small or big, be it financial or non-financial, be it today or tomorrow, be it donating organs or saving the planet, be it loss on personal assets or tragedy of commons. Even regarding Thaler’s popular save more tomorrow; barring money illusion every other element in the campaign is universal, elements like Inertia, loss aversion, self-control and follow through.
Elements linked with decision making have a universality about them. Why don’t we address the commonality in behavior compared to just psychological biases? The authors of Nudge acknowledge the dynamism of changing temporal context of our choices, the 80-20 principle; citing Camere and Hogarth 1999 who suggested that higher the stakes, the less often we are able to practice decision making. There is after all a proportion rule connected to decision making. The authors even go ahead and cite measurement as a key nudge, influencing decision making. Measurements are also referred to as a proportion linked to frequency of decision making. “Some of the life’s most important decisions do not come with many opportunities to practice; this is why there is little opportunity to practice marriages”. The Behavioural commonality could help define a universal choice architect which could work irrespective of context or irrespective of changes in the context. Maybe this is outside the scope of Nudge, but could this be possible?
The need for Rules
Nudge is clear about rules owing to human limitations. The visual system is flawed (Roger Shepard 1990). This could mean that discretionary systems are flawed; a milder opinion could be that discretionary systems have limitations. Humans are mindless passive decision makers, another reason why the society needs rules to prod them into activity. There are also self-control issues which arise because we underestimate the state of arousal (George Lowenstein 1996, hot-cold empathy gap). Then there is pluralistic ignorance (herding), which creates a need to cater to and profit from human frailties. Self-control issues are also most likely to arise when choices and their consequences are separated in time. Then there are choices that have delayed effects, those that are difficult, infrequent, and offer poor feedback need more nudges. And then the decision of which choice architect to follow. Then there are incentive conflicts i.e. is the Nudge ethical or not. The idea of Nudge to offer incentives, understand mappings, set defaults, give feedback, expect error, structure complex choices are about a framework of rules so that the society can move beyond the herding and extrapolation risks.
Nudge is the milder version of self-control. Unlike Ulysses strong will to tie himself up for self-control. If only rules matter; universality and framework could help better define the problem of decision making. The rules and framework have to be better understood before we let technology infiltrate deeper into society, bombarding us with Nudges which create more complexity than simplicity. Therefore faster we realized the rules in psychology, better solutions we can build.
Need to datafy
Rules can’t happen before we datafy psychology. There are inconsistencies in the performance of Nudges. There is a need for objective feedback (did the nudge work? how much % did it work? Is this % effective?), which again is a datafication process. If the contexts for Nudges are dynamic, so is the data related to that context. A lot of Nudges like the post-completion errors (The ATM should prompt you to pull out your card before it gives you the cash) lend themselves to objective framing and help formulate objective rules. Moreover, humans can’t process data and work with elimination by aspects and need simplifying strategies. This is why there is a need for psychology systems (Nudge systems) which can sift through data and simplify. Also the fact that consumer choice has been made simpler by the web, we already have the data we need to build more objective dynamic Nudge systems. Dynamic because we need to add personal data, define proxy data, make sure that we show the light turns from orange to red to an alarm and a technology override, and because we need to keep adding feedback. If societal laws for behavior embody judgments over history than how can psychological inputs and outputs resist datafication?
Assumptions
Just like the authors chose to ignore universality and rule based datafication to develop dynamic nudge systems they also make a few assumptions. “The costs of saving too little are greater than the costs of saving too much”. This is an assumption which does not consider inflationary times and considers hyperinflation a no-case. In case, the decade ahead turns out inflationary with double digit inflation, saving for tomorrow in cash is a disastrous idea. We need to be storing hard assets like gold to conserve. The equity premium is also assumed, as a rule, as a long term valid assumption. Just because equity premium has worked does not mean that it will continue to work. Equity premium has been a key tenet for behavioral finance and a decade (or two) long underperformance of equity vs bonds would challenge that assumption. And just because equity premium exists does not mean that we should assume equity as a long-term investment solution.
This is why what seems like a heuristic might just work out well. Till the time we have a universal framework which can supersede a psychology framework, we have no way of considering psychology based nudges as objective rules. Psychological biases and heuristics have to be statistically significant to claim to be valid or redundant. This is why it’s hard to verify the effectiveness or ineffectiveness of diversification heuristic or any other thumb rule. Whether a bias will turn out better in a certain time or worse in another is for time to tell. It’s specific to the dynamics of the holding period or the decision-making duration. This is why it’s hard to categorically label Nudge for a mortgage as more relevant compared to a Nudge for a health issue.
Divergence and outliers
The author’s suggest that the American society is simply too diverse, individuals are simply too creative, circumstances change too rapidly, the government too fallible. And that even with free markets, government intervention cannot be avoided. This again is an implicit acceptance that Nudges work in a macrocosm which itself is complex and just like the society needs simplifying solutions, Nudges have limitations. Does this mean the author’s admit that it could be good to have a framework to work around Nudges?
Society learns from outliers. The in-extremis; losing a house to foreclosure teaches us about evils of debt; falling sick tells us about evils of lack of exercise. Lung cancer tells us about evils of smoking. Losing a fortune on gambling tells us about the evils of speculation. Vices teach us about virtues. The loss is at the heart of understanding. Failure is at the heart of success. And humans are about vices till at a certain point in time, either the vice kills them or the vice gets transformed into a virtue.
We smoke a cigarette because we like the idea of killing ourselves a bit. We overdo with alcohol because we like the hit more than the long-term addiction it can bring. The gambler likes the thrill of self-destruction, even if you teach him about probability. Risk taking is natural and vices transforming to virtues have a threshold. More people want to be nudged when the pain becomes unbearable. Assuming Nudgeble humans is like assuming docile virtuous people who want to become better, who don’t want to speculate, who don’t want to risk carelessly, who don’t want to kill themselves. How do you Nudge a suicide bomber to change his course? How do make people consume carefully? How do we make people less greedy? Nudge seems too mild a word for change. Change needs dramatic effort. And there is nothing more dramatic when the society topples on its own. Outliers is another universality which drives inefficiency towards efficiency. This does not need a Nudge.
Could a Nudge assist to stop herding? Society will always have self-control problems. There is no total solution. If an individual wants to change he will find ways, a society which is progressive find ways to progress. Nudge is more of a course on self-improvement. But can self-improvement be labeled as strength, or is it just a step in the direction of virtue? Steps are good, but I doubt if steps can transform inefficiency to efficiency and save the world. Inefficiency like a vice is a necessary evil. Utopia is not livable, it’s an illusion. We cannot control or harness divergence, diversity, change. This is why on one side we Nudge people away from cigarettes and on other side legalize marijuana or keep Guns legal.
What’s the solution? Keep nudging is just a solution. The other solutions involve understanding the universality beyond psychology, beyond cause and effect, datafying psychology, objective rule-based dynamic systems, which tell us that the Nudge of yesterday is not relevant anymore; we need a nudge in a new direction. Even psychology like Morning Star rankings suffer from extrapolation, psychology is not a fool proof solution.