Nudge, Sludge and Choice Architecture

 

It took a little more time than I thought it would but I have finished reading "Nudge the Final Edition".  the book was originally published in Spring of 2008.  This final edition was updated to include information on "sludge, Covid 19 and Choice engines.  The book was written by Richard Thaler and Cass Sunstein.  Before purchasing the book I read a couple of the on-line reviews and that did spawn a previous post.  I have also blogged about another book by Richard Thaler; Econs and Nudges.  The basic premise of these works is that humans do not make optimum decisions.  This is contrary to standard economic theory where people will always maximize their returns.  

Thaler has coined the term "Econ" to represent this mythical character embodied by economic theory.  Econs never make an incorrect decision.  Econs do not exist.  In Thaler's world of Behavioral Economics humans often choose incorrectly.  Thaler provides many concepts and ideas for humans poor economic decision making.  These range from simple biases (i.e. favoring the status quo) to a lack of self control (coveting the flashy new car).  Also a preference for having something now versus a later benefit (saving for retirement).  So the main question is how do we help us imperfect humans make sound decisions, or should we?

To this end Thaler and Sunstein introduce the concept of Libertarian Paternalism.  In short libertarianism is a philosophy that values a persons freedom to choose there own course with little to no government intervention.   Where as paternalism is the exact opposite.  Paternalism is defined as the "policy or practice on the part of people in positions of authority of restricting the freedom and responsibilities of those subordinate to them in the subordinates' supposed best interest".  These two terms directly contradict each other.  Using them together is akin to having your cake and eating it too.  The basic idea behind Libertarian Paternalism is to still allow an individual the ability to choose a course of action but use a variety of  technics to guide them to a more beneficial decision.  One such example is saving for retirement.  Libertarian Paternalism would strongly encourage a person to contribute to a 401K savings plan but would allow them to easily opt out if they wish to.

I immediately sensed two main issues with this approach.  The first is that this concept is predicated on the idea that these decisions do not harm others.  If the decision does not harm others, a person should be free to do as they please.  How do we define harm to others?  An obvious example is second hand smoke from cigarettes.  As the negative affects of second hand smoke became more well known smoking was banned in most common areas.  No choice, no liberaltarianism.  However,  what about retirement.  There are many reports on how unprepared the population is for retirement.  One option for these people is to keep working.  This option does harm others.  However, this may not be an option for everyone.  It is likely that a number of people will not be able to continue working and will need to retire without adequate financial resources.  Will we as a society support them with increased government benefits.  If we do decide to increase benefits, who is going to pay for them.  Will my savings be taxed to support them, maybe we will increase taxes on the younger generations.  If so it could be argued that we are harming them to pay for the poor decisions of the older generations.  This train of thought could also be applied to health care costs.  What one eats could be thought of as a personal decision not affecting others.  However poor eating leads to poor health that in the end increase health care costs for everyone.  I am finishing the left over Halloween candy as I compose this!

A second issue would be how much of liberaltarianism and how much paternalism do we allow?  If we think of these as opposite ends of a continuum do we start the person in the middle and let them work from there.  Would it be better to start them at one end of the continuum (i.e. start with a very paternalistic benefit) and have them exert extra effort to work there way back liberaltarianism. This is where the concepts of Nudges and Choice architecture come into play. Choice architecture is how the decision process is structured from the perspective of the user.  A simple example offered is a school cafeteria.  Students would benefit from an offering of healthy foods, i.e. fruits and vegetables.  However, liberaltarianism would allow the choice of a variety of foods, i.e. burgers and fries.  Choice architecture would dictate where food items are placed and how they are obtained.  For example if the goal is to get students to eat healthier foods fruits and vegetables would be placed within easy reach and unhealthy foods would be placed behind a counter where a server would have to retrieve them.  According to the authors choice architecture should allow the user to easily make choices.  One well studied example is retirement contributions through an employer.  One architecture presented is to have all new employees automatically enroll in the company retirement plan.  Making enrollment the default is a "nudge" toward the paternalistic side of the continuum.  In most peoples opinion enrolling in the plan benefits the employee.  The liberaltarian side of the continuum is protected by making it very easy to opt out of the plan.  Essentially an employee can opt out with a single click on a computer screen.

How foods are arranged in a cafeteria and opt out boxes on computer screens are "nudges".  These nudges are meant to ease a user to a more beneficial decision.  By making enrollment in retirement plan employees are more likely to accept it as the default rather than opt out.  The authors offer many different types of "nudges" and how they should be used.  They also discuss when these nudges should not be used and when a nudge goes too far and becomes a mandate.  An example of a nudge going to far is "presumed consent".  Presumed consent is discussed within the context of organ donation.  Presumed consent dictates that all people are organ doners unless they actively deny consent.  In the extreme a citizen would have to go to a specific web site and actively choose to not be an organ doner.  The authors consider this too paternalistic and too invasive on ones freedom to choose.

Another concept introduced is "Sludge".  Sludge is bad.  It is when a choice architecture is so difficult that it impedes the decision process.  We are all familiar with this process.  we just know it by another name; "red tape".  Anyone who has tried to cancel a gym membership or cell phone contract is familiar with this process.  Another example of sludge is rebates.  You purchase something then you need to mail several items (copy of receipt, a form, etc.) to get the rebate.  The best example of sludge comes from South Park where Randy Marsh is Charity shamed into donating money at a high end grocery store.

All in all I found this to be an excellent read.  I thought the idea of choice architecture the most interesting.  Basically how the choice process is started and how nudges are used to guide the user.  It is tempting to look at it as a purely academic exercise however the authors do provide many real world examples where the concepts were applied successfully.  I would suggest that anyone take the time to read this latest edition.