Excerpts from "What's Mine Is Yours: The Rise of Collaborative Consumption"
by Rachel Botsman and Roo Rogers
Get Out of the Way
In the same way that shared commodities such as information, knowledge, and media have become increasingly cooperative, the reusing or passing on of stuff is becoming a vehicle for forming trust between strangers. Beal has a passionate philosophy that “you bring out that goodness in people if you create organizations that are fundamentally built on values of honesty.” Another pioneer in redistribution markets who shares this view is the person Time magazine referred to as “the wizard of the local web,” Craig Newmark, the founder of craigslist. As Newmark puts it, “Trust is the new black.”
As most people know, craigslist is essentially a community marketplace devoted to serving the everyday needs of people—whether you’re looking for furniture, a job, an apartment, a nanny, or a pet. Beyond its community features that include “discussion forums,” “lost and found,” and “job postings,” craigslist functions like Freecycle, where members form local “hubs” and post what they need or what they want to sell or give away. The key difference is that people generally charge for stuff on craigslist. It’s not usually free.
Newmark, just like Beal, created one of the Internet’s most popular sites almost by accident. He started it in 1995 by sending his friends and coworkers a list of cool art and technology events in San Francisco. Gradually, it spread into the wider community. “People started sending me more and more stuff, such as job listings, stuff to sell, and apartment rentals to add to the list, and more and more people asked to be added to it,” Newmark says. Within two years, he had thousands of readers, most of whom he didn’t know.13 From there it has grown into the world’s most popular Web site for classified ads, with more than seven hundred local sites across seventy countries from Romania to Kenya to Canada. The site has 47 million unique users every month in the United States alone—nearly a fifth of the nation’s adult population.14 Worldwide it receives approximately 20 billion page views per month, making it the seventh most viewed site on the Internet. Newmark observes, “This all happened without my intervention and when I saw what was happening most of what I did was just keep out of the way.” Jeff Jarvis, author of What Would Google Do? refers to the strategy of “Get Out of the Way” as “Craig’s Law.” As Newmark explained to Jarvis, “If you make a great platform that people really want to use, then the worst thing you could do is to put yourself in the middle, getting in the way of what people want to do with it.”15
Beyond defining some basic underlying rules, both Beal and Newmark placed the control and decision making in the hands of members. On craigslist it is as simple as “Treat other people as you want to be treated.” Freecycle groups around the world all adhere to this rule: “Keep it free, legal, and appropriate for all ages.” Other than that, local groups designate their own moderators to resolve issues that may come up and to self-police the community. Of course, a small percentage “cheats” the system with spam and false postings. Newmark himself dedicates much of his time to catching the “bad guys out there.” Early on he introduced a flagging system where tens of thousands of users identify spam or inappropriate postings. Instead of sending offenders nasty e-mails, Newmark and Jim Buckmaster, the current CEO, have introduced a reproach that involves a Japanese form of poetry, haiku. If you try to double- or triple-post the same message, something like this might appear: “A wafer thin mint that’s been sent before it seems one is enough, thanks.”16
Considering the size and diversity of exchanges occurring within these networks, instances of fraud or misbehavior are remarkable anomalies. Ori Brafman and Rod Beckstrom write in the The Starfish and the Spider about the “unstoppable power of leaderless organizations” and say that “craigslist members feel that the site is a community and tend to put more faith in a fellow craigslist user than they would in a person off the street. Members assume the best of each other, and generally that’s what they get in return.” Both Beal and Newmark were revolutionary in thinking of their users from the outset as interconnected members or a community of stakeholders, not isolated consumers.
Freecycle and craigslist show how the Internet can be used to create vast decentralized systems of redistribution that are predominately self-organized. Beal and Newmark both believe that their networks have grown so rapidly because they are simple enough to be self-managed. They provided the minimum infrastructure to empower members to work things out among themselves. To paraphrase a point made in an article in Wired magazine on craigslist, it’s actually the lack of helpful features on these networks that “signals us to activate our own methods of reassurance.”17 As with so many other forms of Collaborative Consumption, by pushing power back out to its users, these redistribution markets encourage people to manage their own actions and the actions of the entire community. This in turn creates high degrees of trust and reciprocity to efficiently move surplus stuff from nonuse to reuse.
Our Inner Ledger of Fairness
Every second of every day, buyers and sellers trade $2,000 worth of goods, a majority of which are used, refurbished, or vintage, on the grandfather of reuse marketplaces, eBay.18 Every minute, a secondhand vehicle changes hands on eBay Motors, without ever being taken for a test drive. All this happens between complete strangers with no guarantees the goods they pay for will even arrive, let alone be in the condition they expect. And yet most buyers get what they pay for. “To some economists,” writes Christoph Uhlhaas in his article for Scientific American Mind, “this is a borderline miracle because it contradicts the concept of economic man as a rational, selfish person who single-mindedly strives for maximum profit. According to this notion, sellers should pocket buyers’ payments and send nothing in return. For their part, buyers should not trust sellers—and the market should collapse.”
In traditional shops, before we buy something new, we go through an in-person examination. Whether it’s the fruit we pick up and squeeze, the labels we inspect, or the fabrics we touch, these experiential tools allow us to determine the desirability of a product. We also get a good indication of the reputation and quality of the goods being sold just by walking through the shop door. Even in “old-fashioned” secondhand marketplaces such as yard sales or flea markets, we can look a seller in the eye and ask, “Does it work?” or “How long have you had it for?” or even “Why are you selling it?” And if you decide to make a purchase, it’s usually a simultaneous action of handing over payment and walking away with the goods. In online markets such as eBay that sell secondhand goods, these “tools of reputation” do not exist.20 You can’t physically inspect an item; you may be buying from a seller for the first time; users are free to choose a trader identity, even a pseudonym; and on top of all these points of doubt, you pay without any guarantee your goods will be shipped. How do I know you will actually ship the Tiffany bracelet in time for Valentine’s Day? Yet even if you just go by sales volume alone ($100 billion over ten years), eBay follows what Harvard political economy professor Richard Zeckhauser coined as Yhprum’s law: “Sometimes systems that should not work, work nevertheless.” (It’s the reverse of Murphy’s law, the popular adage that “Anything that can go wrong will go wrong.”) How does a system that shuttles vast quantities of used goods among distant strangers work so well?
For decades behavioral economists wondered whether our sense of fairness is as much a part of human nature as is our obvious sense of competition. The answer to this question is critical to redistribution markets. People’s ability to determine what is fair and what is not plays a big role in making these peer-to-peer reuse systems work. In the “ultimatum game,” the classic experiment in behavioral economics that attempts to understand how fairness works, two people are paired together. Player A, randomly assigned, is known as the proposer. He or she is handed a sum of ten $1 bills (the total amount is irrelevant). The proposer’s job is to decide how the $10 should be split and to make an offer. No dialogue or negotiation between the players is allowed. Player B, the responder, is then given the chance to accept or reject the offer. The twist is that if the responder rejects the sum of money offered by the proposer, both walk away empty-handed.
Does the proposer lowball? Does he split the money evenly? Is he overly generous? When Werner Güth first played the game in the 1980s, he and his colleagues predicted that the proposer would offer the lowest possible amount—in this case one dollar—and the responder would always accept. Why? Player A has the control of the split, so the rational and selfish thing for him to do is propose an offer that gives him the highest amount, in this case $9. For Player B, a dollar is better than nothing, making it in his or her rational interest to accept whatever is offered. The game was played over and over all around the world, from Taiwan to Israel to Pittsburgh to France to Java. The degree of cooperation varied, but researchers observed two consistent outcomes. First, any lowball offers perceived as unfair were almost always rejected by the responder. People will turn down free money rather than let someone else walk away with too much. They don’t care that they would be better off with something rather than nothing. They care about what is fair.21 Second, proposers almost always made fair offers, close to a fifty-fifty split, to ensure the offer would be accepted. In most games, the average amount offered was $4.71. This scenario is like giving a child a packet of M&M’s and telling him it’s up to him to choose how to share it with a friend, with the caveat that he and his friend must both be happy with what they get, or no candy. And the child tips almost half the bag of M&M’s into his friend’s hand. Behavioral economists later discovered that the primary reason a proposer makes a fair offer is that he or she can imagine how the responder will feel and anticipate rejection and anger. Jonah Lehrer writes in The Decisive Moment, “That ability to sympathize with the feelings of others leads to fairness.”
Our inner ledger of fairness is critical to why eBay works so well. For the most part, sellers offer their secondhand goods for a fair price based on what they think the item is worth. They can protect themselves by setting a minimum price, but regardless, buyers usually make a fair bid (whether it’s through an auction or the “Buy It Now” format) that they feel will have a good chance of getting an “accept.” Economist Axel Ockenfels has spent years investigating patterns of trading, negotiation, and cooperation. He explains that “buyers generally will not accept unfair offers and sellers seem to realize that.”23 The same applies to sellers, who for the large part won’t accept offers when bidders are lowballing. “It is not a question of doing something to benefit the other party,” notes Ockenfels, “but a reciprocal behavior: as you treat me, so I will treat you.”
Shadow of the Future
Like so many other forms of Collaborative Consumption, eBay is successful because of the bottom-up community ecosystem that founder Pierre Omidyar calls an “of the people, by the people, for the people” environment.25 In February 1996, when eBay was just six months old, Omidyar wrote a letter to all users: “Most people are honest. And they mean well. . . . But some people are dishonest. Or deceptive. . . . It’s a fact of life. But here, those people can’t hide. We’ll drive them away.”26 Omidyar was referring to what has become the most well-known Internet reputation system, which he first introduced to help members assess and share feedback from their selling experiences.
After any transaction, both the buyer and the seller have the opportunity to rate each other’s performance. There are three types of feedback a user can leave: 1 (positive), 0 (neutral), or –1 (negative). Users also have just one line to write a short comment or to explain why they gave this rating. Typical comments are along the lines of “Item received as described. Thanks.” Or “Seller was extremely helpful. Would highly recommend.” On the flip side, if the buyers did not receive their goods or the goods were not as they had expected, they post comments such as “Paid through PayPal but never received items. Seller is a rip-off.” Once users have reached a certain amount of points, they receive a star indicating their trustworthiness. The “Red Shooting Star” is the very highest, for 100,000 points or more, and the “Yellow Star” is the starting point, for 10 to 49 points. Just as we rate songs on iTunes or films on Netflix or write reviews on TripAdvisor and Amazon, these individual ratings are pooled. Pooling empowers users to take on the responsibility to self-manage the community. As Professor Ockenfels puts it, the reputation system on eBay “functions like an organized rumor mill.”
The striking thing about the feedback on eBay is that it is so positive. Studies have shown that sellers receive negative feedback only 1 percent of the time, and buyers 2 percent.28 This would indicate that 99 percent of the used goods traded on eBay are at the minimum satisfactory. So why do people behave so well? It’s simple. Users know their behavior today will affect their ability to transact in the future. People want to buy from sellers with positive ratings, and studies show that items sold by sellers with a strong track record fetch an 8 percent premium over identical items sold by sellers with low or few ratings.29 As a result, users will often go to great lengths to protect their reputation. Take Doug McCallum, managing director of eBay in the UK. He had sold without a hitch a car, hi-fi speakers, and a fireplace on eBay and had clocked a positive feedback record of about 150. But when he sold his computer, a disappointed buyer wrote to him complaining, “It’s missing the dongle that connected it up.” McCallum explains how he “scoured the universe for this thing to get it to him, finally finding one in America which meant I made practically nothing on the sale but I did avoid negative feedback.”30 It’s incredible customer service on a peer-to-peer level.
An online peer-to-peer exchange for used kids’ clothing, thred-Up, was started in the spring of 2010 by James Reinhart, along with his Harvard Business School classmate Chris Homer and former Boston College roommate, Oliver Lubin. The entrepreneurs successfully launched a similar model in September 2009 for women’s and men’s shirts. Four thousand shirts were exchanged in the first two weeks. But the founders quickly realized that shirts were a niche; the bigger hand-me-down market was in children’s clothing. Young children grow out of clothes on average every three to six months, having worn outfits only a couple of times. “Currently parents are spending upward of $20,000 just on the staples of kids’ clothing by the time their child is seventeen. And they’re retiring some fourteen hundred items!” explained Reinhart.
The entrepreneurs discovered that the problems plaguing existing online children’s swap sites were twofold: inconvenience and poor quality. The team had already solved the first problem with the original thredUp model, building an easy-to-use interface and prepaid envelope system that made the experience feel similar to Netflix. But ensuring quality was a big concern, especially for children’s clothing. The challenge was to design a reputation system that would encourage members to send only what they themselves would want to receive. The team now refers to this benchmark as the “Golden thredUp rule.”
Reinhart explained to us how most reputation systems from Yelp to Amazon don’t seem to decipher the difference between what people actually care about and what they think they care about. They also seem to focus on building a high level of trust or loyalty but don’t pinpoint the single most important behavior the reputation mechanisms need to encourage or the most “unforgivable thing.” “For us it is about how you incentivize quality exchanges. Dirty or poor-quality clothes are not acceptable. Nothing else really matters,” explained Reinhart.
The thredUp founders also recognized that ranking systems were gaining traction on increasingly popular real-time platforms, such as Foursquare, where people are motivated to participate and socially obligated to behave in a certain way to accumulate points and earn reward “badges.” It is part of a human’s innately competitive behavior to want to be at the top of the scoreboard.
The thredUP team ended up implementing three different tools in its reputation system: The first is a star rating similar to eBay where people are rated on a scale of four stars based on the quality of items sent; the second is “Stylie Points,” a subjective zero-to-ten rating users give, based on the stylishness of the items they receive; and third, users are reviewed based on the punctuality of their shipments. Each reputation mechanism is targeted at the core promises of the thredUp offer: quality, style, and speed.
Even though the system is still in its infancy, clothes offered from members with a four-star rating and high number of Stylie points are the first selected. On the flip side, senders who break the quality “golden rule” are being banned from the thredUp community as other members simply refuse to transact with them.
Both thredUp and eBay are examples of reputation systems that fill the critical trust gap needed to make secondhand product markets work. As with McCallum’s “missing dongle,” it’s in the seller’s best interest to resolve the problem. In the case of thredUP, it’s in the members’ best interest to send high-quality clothes on time. Robert Axelrod, a political science professor at the University of Michigan, posits that “people cooperate not because of friendship or trust in each other, but the trust in a promise of keeping a durable relationship that could benefit them in the future.” This tendency he refers to as “the shadow of the future.”31 That shadow creates good behavior in the present, as there are clear incentives for honesty and trust about the price and condition of what they are selling.
No comments:
Post a Comment