Imagine an auction website where the selling prices are consistently amazingly low: customers have bought a $1,799 MacBook Pro for $35.86, a Nikon digital SLR camera for $16.03, an iPod for $15, and so on. Does the site have your attention? You are not alone. These actual winning prices from “penny auctions” do in fact attract the attention of many, many consumers.
Here is how a penny auction works. Before bidding, bidders buy credits, which allow them to make bids. At Swoopo, the German-based early mover in penny auctions, bid credits cost 60 cents each. Thus it costs about 60 cents each time you place a bid—not a lot of money if you can buy a MacBook for under $40. Credits are sold in “bidpacks” that come with perhaps 25, 40, 150, or 1,000 credits per pack. The bidding for an item starts at 1 cent, goes up 1 cent at a time, and continues until a deadline. However, any bid that comes in during the last twenty seconds extends the auction for twenty seconds beyond the current bid, and these extensions continue until no further bid is made. Still interested?
Penny auctions may sound tempting, but I would be embarrassed if any reader of this book participated in one of them after he or she has finished reading this book (or even just this chapter). Let’s take a look at the arithmetic on the MacBook. As noted in one analysis of Swoopo, at the time of auction, you could have purchased the $1,799 list price MacBook online for $1,349—obviously much more than the $35.86 that someone paid for it on Swoopo.1 However, the fact that the MacBook sold for $35.86 means that there were 3,586 bids, at 60 cents a bid, which adds up to $2,151 in bidding fees. Adding in the sales price, Swoopo collected $2,187.46 for their MacBook. Of course, it is possible that the winning bidder made only a few bids and still got a great deal. But imagine a different MacBook auction on Swoopo. As in the other auction, bidders hope to buy the computer at a low price. But in this auction, two or three bidders get into a bidding war and drive the price up to $1,000. In this scenario Swoopo collects $60,000 in bidding fees alone.
A tremendous amount of research has been done on the economics and psychology of auctions. Strange and interesting patterns emerge in auctions, largely due to their ability to push an array of our psychological buttons. As a result, auctions are also excellent for identifying what participants notice and what they miss.
Imagine yourself in a classroom with eighty-five other professionals. I’m your teacher, and I take a $100 bill out of my pocket and make the following announcement:
I am about to auction off this $100 bill. You are free to participate or just watch the bidding of others. People will be invited to call out bids in multiples of $5 until no further bidding occurs, at which point the highest bidder will pay the amount bid and win the $100. The only feature that distinguishes this auction from traditional auctions is a rule that the second-highest bidder must also pay the amount that he or she bid, although he or she will obviously not win the $100. For example, if Tom bids $15 and Sally bids $20, and bidding stopped, I would pay Sally $80 ($100 – $20) and Tom, the second-highest bidder, would pay me $15.
Would you be willing to bid $5 to start the auction?
Martin Shubik invented this auction format, often called the “two-pay auction,” and published an article on it in 1971.2 But I have often been given credit for refining some of the rules and for realizing that it would be a good idea to increase the amount of the $1 auction that Shubik described. In fact I have run over five hundred $20 and $100 auctions over the past three decades.
As any of the more than twenty-five thousand executive students who have seen me teach decision making or negotiations will recall, the early pattern is always the same.3 The bidding on a $100 bill starts out fast and furious until the bidding reaches the $60 to $80 range. At about this level, all but the two highest bidders drop out of the auction. The two bidders begin to feel the trap, but the $60 bidder tops the $65 bidder with a $70 bid. Now the $65 bidder must either bid $75 or suffer a $65 loss. The uncertain outcome of bidding $75 often seems more attractive than the current sure loss, so the $65 bidder charges forward with $75. This pattern continues until the bids are $95 and $100.
The decision of the $95 bidder to bid $105 is very similar to her prior decisions: she can accept a $95 loss or continue and reduce the loss if the other party quits. The class typically finds it very funny, and a bit awkward, when the bidding goes over $100, which it usually does. But at what point did the mistake occur?
A careful analysis of this auction game suggests that individuals who bid create a potential problem for themselves. While one more bid may get the other guy to quit, if both bidders feel this way the result can be a disaster, with both the “winning” bidder and the second-place bidder paying a significant multiple of the value of the bill being auctioned. In my experience running more than five hundred of these auctions, people always bid, no bidder has ever made money, and my students collectively have lost tens of thousands of dollars.4 I have had $20 bills sell for over $400 and $100 bills sell for over $1,000.
What advice would I give bidders in this auction, if I weren’t busy trying to trap them? As is true of war, sometimes the best strategy is to stay on the sidelines. Successful decision makers take note of environments where inaction is the right course. The key is to identify that the auction is a trap and never make even a very small bid. More broadly, a rational approach to decision making requires us to consider our decisions from the perspective of the other decision maker(s) involved. In my $100 auctions, this approach would quickly tell you that the auction looks just as attractive to other bidders as it does to you. Armed with this knowledge, you can accurately predict what will occur and stay out of the auction.
Swoopo’s business plan is an adaptation of the Shubik two-pay auction, re-created in a sinister all-pay (or pay-per-bid) format. Created in 2005 with the original name Telebid, Swoopo was quite controversial, accused of being a hustle, a devilish swindle, and even the “crack cocaine of online auction websites.”5 As of January 2012, its site had been taken over by a penny auction website called DealDash. In fact there are currently dozens of all-pay auction sites on the web, all offering unbelievable deals. And there is a reason they are unbelievable. Many critics have argued that these websites are actually gambling sites that should be subject to greater government regulation; tellingly the all-pay sites claim they are in the entertainment business.
Researchers Emir Kamenica and Richard Thaler looked at twenty-six Swoopo auctions of checks for $1,000. (Yes, Swoopo used to auction off cash.)6 On average Swoopo received $2,452 for each $1,000 it auctioned. The winning bidders did make money in all but two of the twenty-six auctions, but the rest of the bidders in these auctions (all of whom were expecting to make money) lost money. Only the final bidder was a winner. Ned Augenblick, an economist at the University of California at Berkeley, estimates that, on average, Swoopo collected 50 percent more than the value of the commodity being auctioned.7
How can we explain the mistakes of the bidders in these types of auctions? Having auctioned off hundreds of $20 and $100 bills in two-pay auctions, and having talked afterward to the executives who chose to bid, my experience is that bidders exhibit three common mistakes. First, they fail to follow the common advice to put yourself in the other guy’s shoes. In the $100 bill auction—in fact in any auction—it is essential to think about the other bidders’ motivations. In my auctions, and in the Swoopo auctions, if you think about the other bidders, the appearance of a great deal quickly disappears.
Imagine that you plan on simply making thirty bids for a MacBook in a penny auction, starting at $20. This limits your exposure to a loss of $18 in bidding credits. What about the other bidders? If other bidders also plan to start bidding beyond the $20 level, they probably have a similar strategy. Thus your most likely outcome is that you will not win the computer, but you will add $18 to the bottom line of the all-pay auction site. And if regulation is lacking, it may well be the case that a party loosely or closely connected to the website will keep bidding up to a level that guarantees the website a profit.
Second, people who plan on making only a limited number of bids in an auction often bid past their self-imposed limit. Why? To justify their decision to participate in the auction in the first place. Once the executives in my $20 and $100 auctions become trapped by the fact that the second party needs to pay his or her bid, they escalate their decision to remain in the auction. When two or more bidders escalate their commitment in this manner, the bidding can really take off.
Third, like Swoopo bidders, once in an auction, most of us feel an irrational desire to win, no matter the cost. With these three common mistakes in place, I am optimistic that I will continue to be successful in my role of auctioneer, and pessimistic about the outcomes of people who participate in my auctions and in all-pay auctions.
How can you avoid being the next victim in an auction trap? First, recall that the subtitle of this chapter is “Or, If It’s Too Good to Be True . . .” Anytime you hear about something that seems too good to be true, you should be skeptical. The next part of the analysis is to recognize that a for-profit company is conducting these auctions. Why would the company’s owners be willing to auction off a MacBook for as low as $40? Notice that penny-auction bidders can gain an extra hint by asking themselves, “Why is this site offering these items at such low prices?” One can imagine a professor being willing to lose a small amount of money for the purpose of some strange in-class demonstration, but that is far less likely to be the motivation of a for-profit website. The prompt “It’s too good to be true” should help you avoid similar wily traps that are waiting to be devised.
Imagine that you have built a career as an investment adviser. Part of your job includes identifying investment opportunities for your customers. For years one of the investments that you have recommended to clients has outperformed the market, with a surprising low level of risk. The fund performed consistently across years, even when the market was down dramatically. The head of the fund is a well-respected individual, and your customers are delighted with their returns. You earn 2 percent of all the money that is invested in the fund, plus 20 percent of the upside—more than your compensation for selling other investments. How does this sound to you?
Let’s add a few more details. Finance experts argue that you cannot significantly and consistently outperform the market with the low volatility of this fund. In addition, the SEC has investigated this fund on multiple occasions.
Oh, and by the way, the name of the man running this fund is Bernard Madoff.
It is now clear that, while at least a few of the investment firms that sold Madoff’s investments were filled with crooks who knew that the fund was too good to be true, many other investment advisers simply never noticed any problem with Madoff’s returns. Over three decades most of Madoff’s investments passed through “feeder funds,” funds of other investment advisers who proudly told their clients they had access to Madoff or claimed they used an exotic investment strategy. In reality they did nothing more than turn much, sometimes all, of the money they collected from investors over to Madoff. These salespeople simply acted as intermediaries and were extremely well paid. As Madoff claimed his fraudulent record of amazing success, the intermediaries were getting rich.
Madoff was a thief. His Ponzi scheme created enormous losses, wiping out $64.8 billion in paper profit. Were his intermediaries likewise thieves? Ample evidence shows that many intermediaries had hints that something was wrong but lacked the motivation to notice the evidence that was readily available. Consider René-Thierry Magon de la Villehuchet, the CEO of Access International Advisors and Marketers. He invested his own money, his family’s money, and money from his wealthy European clients with Madoff. De la Villehuchet was repeatedly warned about Madoff and received ample evidence that Madoff’s returns were impossible, but he turned a blind eye to the overwhelming evidence due to his desire to believe in Madoff and bank those returns. Two weeks after Madoff’s surrender made de la Villehuchet’s role clear, he killed himself in his office in New York. As this tragic story shows, human beings have an amazing capacity to ignore clear warning signs of others’ unethical behavior.
In fact the list of people who didn’t notice that Madoff was a fraud is long. It includes individual investors, many of whom had sophisticated knowledge of finance, major investment firms, and the feeder funds that sold Madoff’s investments. Many MBAs worked for these organizations, as did other individuals with impressive financial credentials. The list also includes the Securities and Exchange Commission, the government body we depend on to regulate organizations such as Madoff’s firm. None of these people or organizations noticed Madoff’s crimes despite massive anomalies, the grand magnitude of the fraud, inconsistencies in statements made by Madoff to the SEC, unprecedented secrecy in how his fund operated, and many whispers on Wall Street.
Why didn’t they notice? Many didn’t want to notice; they suffered from positive illusions, or the tendency to see the world the way we would like it to be. For those who invested with Madoff, positive illusions ran wild. We also know that people have trouble noticing the gradual decline of others’ unethical behavior. Thus when fraud occurs over time, particularly on a slippery slope, the impossibility of returns such as Madoff’s is likely to go unnoticed. Too many of us leap at the prospect of 20 percent returns year in and year out, or a $40 MacPro laptop, rather than ponder how such improbable promises are being delivered.
Those of you who followed the story closely may recall that someone did notice the impossibility of Madoff’s returns. An independent financial fraud investigator, Harry Markopolos, repeatedly attempted to warn the SEC that Madoff’s returns were not legally possible.8 Markopolos’s own account makes it clear that he didn’t know whether Madoff was sitting atop a Ponzi scheme or engaging in front-running through an arm of his organization that cleared transactions for other investment firms. Front-running is a form of insider training in which an investor uses his knowledge about trades coming in from clients. The investor puts his own trades ahead of large transactions to take advantage of how these transactions are expected to move the market. (Front-running is illegal, but while a Madoff investor likely would be harmed in a Ponzi scheme, she would benefit from front-running.)
Between 1999 and Madoff’s arrest, Markopolos approached the SEC five times. His 2005 book identifies a large number of red flags, and an appendix reproduces a thirty-five-page document that he provided to the SEC. His book also makes it clear that, by 2005, he thought it was far more likely that Madoff had constructed a Ponzi scheme than that he was front-running. So why didn’t the Securities and Exchange Commission pay attention to Markopolos?
In her fascinating book The Wizard of Lies, reporter Diana Henriques discusses Markopolos’s own bounded awareness. His presentation of the evidence to the SEC was unnecessarily complex and confusing at best, and it seemed to be aimed more at showing how smart he was than at providing evidence against Madoff. While the complex financial issues that Markopolos analyzed did provide evidence against Madoff, much simpler evidence was available: namely, the other side of Madoff’s trades.
To convince the SEC that his trades were real, Madoff provided fabricated statements he claimed were from the Depository Trust and Clearing Corporation (DTCC), which keeps a record of all trades. If Markopolos was convinced that Madoff was running a Ponzi scheme, instead of delivering a complex technical analysis he could have simply encouraged the SEC to call the DTCC to confirm that Madoff’s trades had actually occurred.9 A quick call to the DTCC would have immediately uncovered the Ponzi scheme. Moreover there is evidence that Madoff always expected to be caught based on the lack of trades documented by the DTCC. But Markopolos never made this suggestion, nor did the SEC think to make the call on its own.
Along with his preference for complexity over clarity (let alone simplicity), Markopolos was also an unfortunate messenger for the discovery of Madoff’s fraud. Most of us understand that when you want someone to pay close attention to information you have, it is important that the person has an interest in listening to you. Simply put, it helps to be likeable. While it is clear that SEC employees made tragic errors, it also appears that they wanted nothing to do with Harry Markopolos. They found him arrogant, condescending, and insulting; and female SEC employees found him sexist as well. In Markopolos’s own memoir—which is supposed to focus on the Madoff episode—he chooses to relate the negotiation that he had with his fiancée, in which he proposed buying her breast implants instead of a diamond ring, so that they could both enjoy the gift.10 It is perhaps not surprising that he titled his book No One Would Listen. His bounded awareness of his interpersonal shortcomings, convincingly described by Henriques, may have been an obstacle to catching Madoff far earlier.
This account highlights that there were at least two threads entwined in this story. One, Madoff’s returns were too good to be true, and the SEC should have noticed. Two, when you do notice and want to tell others, put yourself in their shoes—advice that might have allowed Markopolos’s message to be heard.
In the 1980s and 1990s politicians worked with mortgage lenders Fannie Mae and Freddie Mac to increase home ownership in the United States, largely by weakening the standards consumers needed to meet to qualify for loans.11 Many of the politicians behind this initiative, including Bill Clinton, had good intentions: they wanted to help more people own their own home. This was generally a fine policy, if housing prices would always go up—an expectation that is obviously too good to be true.
Fannie and Freddie are in the business of buying mortgages from mortgage lenders, which issue mortgages directly to the public. Fannie and Freddie are strange entities: they have private shareholders, but their mortgages effectively are backed by the federal government. With the combination of the government-led effort to encourage home ownership and the presumed federally guaranteed safety net, Fannie and Freddie could provide shareholders with wondrous returns and dump any significant losses from potential massive mortgage defaults on taxpayers.
With housing being a good investment historically, and with greater access to loans, more and more people with marginal qualifications purchased homes. Led by CEO James Johnson, Fannie Mae aggressively pursued changes to U.S. law that provided it (and Freddie Mac) with greater flexibility and less oversight. Johnson helped fuel the market by lobbying for regulatory reform that drove down the need for a down payment, reduced the due diligence required by lenders to provide mortgages, and created new forms of mortgages with low initial rates (which could be predicted with high probability to go up over time). The availability of easy money, creatively designed (subprime) mortgages, complex mortgage documents, and the failure of regulatory oversight contributed to a dramatic surge in mortgages that home owners would not be able to pay if bad things happened in their lives or in the economy.
Large mortgage lenders, such as Countrywide, employed hundreds of brokers who were charged with originating loans that, thanks to existing regulations (or lack thereof), did not require applicants to verify their income and did not require truly independent appraisals. These mortgage brokers were rewarded financially for their success in helping home buyers complete their purchase. And it is rare for the actual customer to see the mortgage broker as someone pursuing his or her own rewards, as opposed to someone trying to help the buyer. James Theckston, vice president for Chase Home Finance in southern Florida, and his team wrote $2 billion in mortgages in 2007. Later he said, “On the application, you don’t put down a job; you don’t show income; you don’t show assets. . . . But you still get a nod.”12 Mortgages were approved and quickly resold in the secondary market, most typically to Fannie Mae and Freddie Mac, then packaged along with many other such mortgages to create mortgage-backed securities.
Investment banks eagerly purchased such mortgage-backed securities, combined them, and sold them off as bonds. Often the mortgage-backed securities, also called collateralized debt obligations, were structured in ways that were designed to game the models that rating agencies used to rate them. Magically, in this process high-risk mortgages were transformed into AAA-rated securities.
The owners of these securities often invested in insurance in the form of credit default swaps (CDSs) issued by large and well-regarded insurance firms, such as AIG. The traders in the insurance companies happily issued hundreds of billions in CDSs without posting collateral for them. These firms avoided the need for collateral by not technically referring to the CDSs as insurance. Yet they had no means of paying off their debt in the event that mortgages went sour in large numbers.
As long as the economy thrived, all was well. Home buyers were courted by lenders and had access to mortgages that were never before available to individuals with their meager assets and low income. Employees throughout the mortgage distribution chain were thriving, particularly the executives. The banks, hedge funds, and other companies that owned the mortgage-backed securities were delighted with the high, above-market yields being offered. And these investors were happy to pay insurance companies for (seemingly) taking away the risk, without noticing that the insurance companies lacked the financial resources needed to pay for the losses in the event of a mortgage meltdown.
But then the trouble began. As the economy hit some bumps, the housing market began to unravel. Home owners affected by layoffs and/or expiring adjustable rate mortgages could no longer meet their mortgage payments, and the number of defaults began to mount. Mortgage originators were no longer able to sell off their mortgages to the secondary market, and their stock prices dropped dramatically. The mortgage-backed securities held by investors crumbled in value as it became clear that there were many loans that would never be paid back in full. As housing prices dropped dramatically, the value of many homes would no longer cover owners’ outstanding loans.
Once the holders of the mortgage-backed securities realized that their securities had dropped in value, they expected the insurance companies, particularly AIG, to cover their losses based on the CDSs. But AIG and others were unable to meet these claims due to their lack of resources. AIG went to the federal government for a bailout. The government agreed, since the failure of AIG threatened the entire economic system: the company was simply too big to fail. Too late the investment bankers realized—apparently for the first time—that the CDSs provided no real insurance and that the entire system was built on a set of mutually reinforcing and self-serving illusions.
In this process, home owners appeared to have ignored their inability to pay their mortgage if negative events occurred, such as a job loss or economic downturn. They ignored the escalation in interest rates that was predictable in their subprime mortgages. And many failed to think ahead about how their lives would be affected if they eventually faced foreclosure. Many lost their home and saw all their savings wiped out.
Most of the organizations described in this story are otherwise doing just fine. They extracted enough profit from the economic boom to compensate for their eventual losses. But these outcomes leave unanswered the question, Who is responsible for the disaster?
Congress failed; citizens failed; and most especially regulators, whose job it is to notice, failed. Had regulators applied the “too good to be true” measure, they would have caught the problem from the start. If they had applied the “wear the other guy’s shoes” measure, they would have seen both home owners’ incentives and mortgage brokers’ incentives. And we citizens continue to fail, as we do not hold our elected officials accountable for their failure. Politicians continue to blame regulators for the crisis, yet we vote against those who would institute needed regulations.
How did we not notice? To begin with, the 2008 financial collapse is a far more complicated story than I am able to convey here. This makes it tough to diagnose the specifics of the problem. Yet we did have the information we needed to know that a massive crisis loomed. If hundreds of thousands of people hold mortgages that, in the event of an economic downturn, they won’t be able to afford, it fails simple logic to believe that Wall Street can create magic by bundling and transforming those mortgages into relatively safe investments. The fact that an insurer would guarantee those investments should have prompted the question, Can the insurer make the appropriate payment if there are massive losses? But somehow it didn’t. We trusted the complexity of a system that did not warrant that trust.
Part of this trust comes from a desire to see our investments and our opportunities in a positive light and to overlook hazards. Part of the trust comes from our tendency to discount the future. People wanted to buy a nice house now, and they discounted the long-term risks. Investors wanted their above-market returns now, and they failed to think through the real hazards. The media continued to favor simple stories that could be explained in sixty-second segments rather than questioning systems that they did not comprehend.
It might be tempting to say that the collapse couldn’t have been predicted, but Michael Lewis’s The Big Short documents the handful of investors who understood the entire path the disaster would take before it struck. But because it was not their job to inform the public, they took the other side of the bet and made their fortunes as the financial markets collapsed. These stories are each complex and brilliantly told by Lewis, but the commonality is that these investors noticed a pattern in the market that was too good to be true—and they bet on that analysis.
The 2008 financial crisis was complex. Let’s consider something much, much simpler. When I meet people who actively trade stocks, I often ask them why they think they know more than the party on the other side of the trade. Most of these investors have never considered this question. They ask what I mean, and I try to clarify. When an investor is buying a stock, it is because someone else is selling it. As the buyer, shouldn’t you consider what is going through the seller’s mind?
When I ask investors why they want to buy a particular company’s stock, they tend to mention positive aspects of the company: the firm’s historic return on investment and profit, its growth potential and control of unique properties. They tend to overlook the fact that the market at large has the same positive information about the company. Indeed in most cases, someone has far better information about the company than you. Yet most investors are paying fees to make an exchange with someone, the seller, who has better information about the trade than they do. Overall, I argue, this sounds like a bad bet.
The key to such situations is to ask the most obvious questions. If an opportunity seems too good to be true, further diagnosis is necessary. We need to think through the worst-case scenarios, and we need to think through the actions and motives of all of the relevant players—the topic of my next chapter.