Behavioral Economics and Digital Institutions: A View from the Field

The Gruter Institute for Law and Behavioral Research is holding a conference at Lake Tahoe next week. The theme is “Law, Behavior & the Brain.” I have been asked to give a talk on digital institutions. You may be asking, “What is a digital institution and what does it have to do with my brain?” For those answers and more, here is the abstract for my talk.

Abstract

Digital Institutions allow us to “reframe strategic interactions,”[1] incorporate externalities, and leverage behavioral economic tendencies toward more positive outcomes. Compared to traditional institutions they are relatively quick to establish, cheap to operate, and easy to adapt or discard. They can even be self organizing and in the extreme lead to a vision of “governance by algorithm.”[2]

Today’s most innovative corporations, though founded as traditional institutions, are both witness to and participant in the emergence of their digital counterparts. These same companies will also be deeply changed by the phenomena. This talk provides a first person view from within IBM’s famed T.J. Watson Research Center. With 60 years of history, over 3,000 researchers, and multiple Nobel Prizes, IBM Research provides an exceptional environment for identifying and making sense of change. The discussion covers three topics from the perspective of a Digital Institutions “practitioner.” First, a survey is provided covering several technologies developed at IBM Research that enable Digital Institutions; technologies that include 3D metaverses, real time speech-to-speech translation, and semantic web reasoners. Next, we examine a specific instance of Digital Institution formation involving a volunteer team leading IBM’s World Development Initiative (WDI) to address the needs of those people at the “bottom of the pyramid” living off less than $5 per day. Finally, the talk covers a problem space for future investigation in intellectual property licensing where Digital Institutions might produce better outcomes through overcoming psychological bias in decision making.

[1] Oliver Goodenough and Monika Gruter Cheny, “Is Free Enterprise Values in Action?” Preface to “Moral Markets: The Critical Role of Values in the Economy,” Edited by Paul J. Zak, 2008
[2] John Henry Clippinger, “A Crowd of One: The Future of Individual Identity,” 2007

Penny for your Thoughts: Revisited

Here is a paper that might put a new lens on the previously discussed Alter and Oppenheimer money familiarity study (or perhaps vice versa). In "The Dishonesty of Honest People: A Theory of Self-Concept Maintenance," Nina Mazar, On Amir, and Dan Ariely discuss the results of a series of cheating experiments. In the experiments students are paid 50 cents for each question they get right on a test. Some of the experimental conditions provide an opportunity to cheat by allowing students to self report their scores. The surprising finding is that cheating dramatically increases when tokens are given to students instead of cash, even though these tokens can be exchanged for cash at a station only a few feet away. The authors of the Dishonesty study believe there is something special about cash. It is why people might take home a few office supplies but they are much less likely to take the equivalent value out of petty cash. Moral ambiguity is erased.

Perhaps there is a related effect at play in Adam Alter and Daniel Oppenheimer’s research. $2 bills, Susan B. Anthony dollars, and altered bills play the role of tokens. Like the tokens in the Dishonesty study people rationally know they have equivalent value to cash (ironically a token of value in its own right). However, in practice tokens and cash are not the same since people behave differently by say, increasing their cheating or having a willingness to part with a token at a discount. Of course we beg the question of why $2 bills are tokens and $1 bills are not. We may be back to familiarity.

Another take is that familiarity is playing a role in the Dishonesty study. There could be a discounting of the unfamiliar tokens taking place. People are cheating the same “amount” but they need more of the discounted tokens to steal the equivalent perceptual value.

Penny for your Thoughts

Into the growing body of research on “predictably irrational” behavior comes a new study from Adam Alter and Daniel Oppenheimer of Princeton, to be published in the Psychonomic Bulletin & Review. Like others before it, this study examines the great plasticity of perceived value. In their experiments, Alter and Oppenheimer’s subjects value a $1 Susan B Anthony coin less than they do a traditional $1 greenback bill. Similarly, subjects also seem to value $2 bills less than they rationally should as compared to the more frequently encountered “ones.” The authors argue that this is a result of familiarity which subjects value more than they rationally should.

Initially, the authors’ explanation of the results did not strike me as correct. Generally rare things are the most valuable. Even with money, the units we see the least frequently (like $100 bills) are more valuable than those we see all the time ($1 bills). However, I have tried to think of alternative theories and I have not hit on anything that holds together as well as the familiarity explanation.

One alternative explanation is that subjects are using touch stone categories as heuristics to determine value. Imagine that we place amounts we encounter into one of three buckets valued lowest to highest: pocket change, wallet, and bankable. These categories are formed around how people use money, where they physically store it, and perhaps some kind of “purchase ambition.” Subjects would place a Susan B. Anthony coin into the pocket change bucket. The $1 bill would be placed in the higher category of “wallet” and thus be valued more. Of course the $2 bill falls into the same “wallet” category as well yet subjects value the $2 bill significantly less than two $1 bills. Something else must be going on.

Perhaps there is a glass half full or half empty framing effect with value. Here are some possible subconscious monologues. I have a single coin and I feel poor because you can’t get much for a coin (half empty). I have a dollar bill and I feel relatively wealthy because one dollar is a psychological threshold to having something of real value (half full). The same effect may happen with two $1 bills. I separate the money into two units and frame each on the threshold of value and feel even wealthier (half full). On the other hand, the $2 bill may inspire visions of higher value bills. I now frame on the wish that I had a $20 bill and I then feel poor (half empty).

Neither of the two theories I put forward can account for a third experiment run by the authors but it does prompt a new alterative. In the third experiment subjects effectively compare the value of one real $1 bill and one that has been slightly altered by reversing the image of George Washington’s head (the fake). The fake should appear less “familiar” than the real bill and therefore be valued less, which indeed was the case.

Is it possible that subjects are not valuing the familiar itself but instead discounting the unfamiliar? Unfamiliar money could somehow make subjects less comfortable, perhaps because they fear that it could be fake? They may worry it will be more difficult to trade it to others in the future even if they understand that it is the equivalent to the more frequently encountered money. Maybe, but to risk a bad pun this is probably just the other side of the same coin. I think I will stick with the authors on this one after all. They present the simplest explanation to all the observations.

Neuroeconomics: Testosterone and Trading

The Wall Street Journal recently published an article on the possible impact of testosterone on stock traders. The study is by John M. Coates, a senior research fellow at Cambridge. His general finding is that higher levels of testosterone are correlated with higher levels of risk taking and (during the course of the study anyhow) better trading results. Increased risk taking behavior seems intuitive; however, the one percentage point gain in financial performance is not as clear cut. One could dig into the original study to see how Coates tried to account for this but I do not see a clear way to determine if the trading decisions were good or bad ones based only on the outcome. Taking really risky bets can sometimes yield high returns. However, could those same returns be generated (on average) with more certainty and less risk? Would they have a higher risk adjusted rate of return?

The more interesting study mentioned is that of MIT Sloan’s Andrew Lo. He wired up traders to monitor their psycho-physiological state in real time while executing real trades. It seems he would have much tighter paring of the independent variable to the actual decision making moment. However, this study may suffer the same issues with the outcome measure. Some time ago Professor Lo presented his experimental design in a class I was taking at the MIT Media Lab (Sandy Pentland’s Digital Anthropology Class). I was an MBA student at the time and playing a lot of Texas Hold’em with buddies from the Muddy Charles -- so perhaps that was the inspiration but I suggested he run a similar experiment on Black Jack players. Unlike with the stock market, the expected value of given Black Jack hands can be precisely quantified. On a hand by hand basis you could determine if the subject made the decision that maximized his or her expected value from playing the hand. I would guess that a higher testosterone level and/or greater excited psycho-physiological state would cause subjects to make suboptimal decisions, even if on occasion they got lucky and won.

Framing Effects and IP Licensing

Marketing Professor John Gourville has produced some very interesting work on how framing effects contribute to new product failure. In my opinion Gourville’s primary contribution is to extend the analysis of framing effects to include the influence it has on product teams creating the innovation as well as the more traditionally studied consumer. He speculates that product innovation teams eventually get so engrossed in the new features of their innovative product that they reframe to the innovation as status quo - while their potential customers of course to not share this new frame. Since individuals value gains less than they do comparable losses the new framing leads to substantial misjudgment of customer willingness to adopt the new product. Product innovation teams see “living without” their fancy new features as a loss to great to bear whereas customers view these same features as small gains. The work is published in a few places including as a working paper from the Marketing Science Institute called The Curse of Innovation: Why Innovative New Products Fail.

I believe a different but related analysis can be carried out in the realm of IP licensing. Generally speaking, in the corporate world research investments are made with the intent of creating valuable IP for a company’s own use in products it brings to market. However, at the other end of the research process companies often find they have some subset of IP that is valuable but does not necessarily fit the company’s business plan. In this case the company is faced with a decision: a) license/sell the IP to another company who wants to use it in the market in exchange for a fee and/or royalty stream or b) hold onto the IP “for now” as an option just in case the company later decides it wants to make use of it.

Many argue that far too often the choice is “b” for various reasons including inertia, fear of creating a competitor, inability to find a suitable licensee or to properly value the asset, etc. I believe a substantial contributor to under licensing is the frame in which the eventual performance of a licensed asset is viewed. Many in the licensor company will view licensee performance from a perspective of “it could have been ours.” They take on a frame that had their company made use of this IP they would have 100% of whatever benefit is derived by the licensee. The entire gain of the licensee is framed into a “loss” for the licensor. Even if the licensor received a substantial royalty in the arrangement and thus shared materially in the success, this royalty would be viewed in the realm of gains or netted out entirely as a loss if it is considered at all. This creates a paradox in that the greater the success of the licensing arrangement the more licensors will perceive a costly loss. It is no wonder in the face of this paradox licensor decision makers often wrongly choose to let the asset “rot on the shelf.” They are able to get away with this because most companies do not seem to assign the cost of spoiling IP inventory to the licensor decision makers. That true loss needs to make its way into the mental calculus if we are to make better licensing decisions.

Moral Reasoning: Why do people perceive stealing intellectual property as different from the theft of physical property?

An interesting new collaboration may happen out of my visit to the Berkman Center. Oliver Goodenough, who was mentioned in the last post, is embarking on a project to better understand why some people who would never steal something as mundane as office supplies are somehow able to self justify their theft of intellectual property. The IP itself could take many forms from song or movie downloads to knowing infringement of a business process patent and can be quite valuable. Yet for some reason, who knows maybe even good reasons, people view some IP theft as more minor than shoplifting a t-shirt. Professor Goodenough and I are hoping to work together to address some of the interesting questions this paradox creates.

Below are a few initial concepts for exploration, several of which came out of a conversation I had with a good friend Ryan Ismert.

Perception of Amoral Act

• Expectation -- We have been trained since childhood that content can be "free" because of advertising supported TV and radio. Not many would say that someone is "stealing" because they step away during the commercial break, even though these ads are paying for our content. This "used to getting it for free" sentiment may compel people to avoid IP payments when they can obtain content by other means.
• Metaphor-- When we are taught the nature of theft; we are most typically provided with material property examples. If the crime doesn't fit neatly into these core examples we may more easily rationalize our theft.
• Victim Impact -- Crimes have victims. Stealing physical property is a zero sum game. In most cases only one owner can benefit from it at a time. As such, stealing physical property deprives one party for the benefit of another and creates a clear victim. With content and ideas, copying someone else's bits or ideas does not directly take from their owner. Their owner is still free to use the content/idea. Instead the theft circumvents their monopoly right to that IP.
• Prevent v. Produce -- Patents provide the right to prevent someone else from producing. To the owner they do not create the right or the obligation to produce. Overcoming someone else's right to prevent me from doing something (practicing an idea or down-loading free content if I am able to) does not feel like stealing in the same way that home invasion does.
• Problem with Disregarding -- Once you learn something it is very difficult to neglect that information. If someone is exposed beneficial IP, such as a superior business process, it will be very difficult for them to disregard it. That person may feel compelled to follow a more efficient/superior process if one is known.
  

Willingness to Commit Act

• Tangible Risk -- Physical property must be absconded. The act of stealing in the material world is observable, as is the evidence embodied in the object itself. Intangible property theft and retention is perceived (probably falsely) to be less observable.
• Victim Empathy -- In many cases, the victim of IP theft is less directly observable. Physical property theft victims can usually be specifically identified and therefore have a greater likelihood of creating empathy. IP theft victims are more often unsympathetic corporations who are even sometimes cast in the villain role. It may even create a Robin Hood effect if the IP thief is seen as stealing from the rich for the benefit of the "poor."

Berkman Center Visit

A few colleagues and I paid a visit to the Berkman Center for Internet and Society at Harvard Law School this past week. They are doing some fascinating work in a variety of areas. The topic of the day was how to bring sustainable development to people living at the “Bottom of the Pyramid” – those living on less than $2 per day. Berkman projects on Identity and Reputation in scale systems are especially relevant. At the village level a person at the bottom of the pyramid can only effectively do business with people they know and trust. They have little access to the modern infrastructure wealthier people rely on every day such as ID cards, credit ratings, etc. Finding ways to expand trust infrastructure is critical to development projects of all kinds.

After the meeting I spent some time with Oliver Goodenough, a Professor of Law at Vermont Law School. I did not realize this until after the meeting but it turns out Goodenough worked with Richard Dawkins including co-writing the Nature paper “The 'St Jude' Mind Virus.” Now he is doing research which applies neuroscience to problems in business/law. This includes using fMRI to study moral reasoning when it comes to legal subjects.

It is amazing to me to see how rapidly the cross pollination of social sciences research with neuroscience is taking place. Our increased understanding of the brain is going to remake social science as we know it. We are living in exciting times.