Consider two hypothetical fires. A garment factory in a Third World country with minimal governmental regulatory oversight burns down, killing half of the three hundred women and children employed there; it subsequently becomes clear that the factory’s owner failed repeatedly to spend money on meeting minimum safety standards. A second fire kills a suburban dad who filled his lawn mower with gas too close to a recently tossed cigarette; the fire raced up the flow of gas and into the can, which exploded. Who is to blame for these deaths?
Let’s shift our focus. If you live in the United States, you probably have shopped at Walmart at least once or twice, and you likely are aware of their “Everyday low prices” tagline. Have you ever thought about the connection between their everyday low prices and the safety of the products Walmart sells, or the connection between their everyday low prices and the safety of the employees who make the goods that the retailer sells? No need to feel awkward if your answer is no. Most people do not think about the harms created by indirect actions, that is, behaviors that hurt others indirectly, such as buying a low-price product from a company that skimps on safety. But perhaps armed with more data, you will.
Blitz USA was once the largest manufacturer of gas cans in the United States, with approximately 80 percent of the gas can market. Cy Elmburg, the chairman of Blitz USA, has testified that in July 2006 he sent a letter to the CEO of Walmart asking Walmart to get involved in a national consumer awareness campaign aimed at protecting consumers from gas can explosions. Elmburg felt he needed Walmart’s cooperation because the gas cans produced by Blitz and sold through Walmart had been connected to dozens of explosions, serious burn injuries, and deaths. In their contracts with Walmart, suppliers must agree to accept any financial or criminal liability resulting from the sale of their products. Elmburg felt that Walmart, given its size and as the point of purchase, had an ability to influence consumers in a way that Blitz could not. Perhaps because it was protected from liability, Walmart failed to act on Elmburg’s proposal. The explosions continued.
The basic problem with the Blitz gas cans is that when gas is poured from them, there is a risk that gasoline vapors will connect with an ember or other fire source, and the fire will run up the gas flow into the can and explode.1 This had occurred in many dozens of cases, and a large number of lawsuits against Blitz had followed. A former Blitz employee has testified that Blitz presented a revised gas can design to Walmart that would prevent the burn injuries by installing an “arrestor,” a device that would prevent a flame from flowing into the can, at a cost of between 80 cents and $1 per can.2 According to this testimony, Walmart rejected Blitz’s design on the basis of the price increase, and Blitz halted its redesign project because it would be difficult to launch a national product that Walmart refused to purchase.
The flip side of Walmart’s policy of providing everyday low prices to its customers is its goal of securing everyday low costs for Walmart. The guideline given to Walmart buyers is to achieve low costs, a motto that its buyers are encouraged to live and breathe.3 Court testimony provides extensive evidence that Walmart places extreme price pressures on its suppliers. This can translate, as it is claimed to have with Blitz, to a supplier realizing that adding commonsense safety features to a product can prevent it from acquiring Walmart’s business. My wife, Marla Felcher, is a product safety expert, and we have a shared interest in what keeps safer products from reaching the market and what keeps less safe products on store shelves. In 2002 she wrote:
As the world’s largest retailer and the nation’s largest toy seller, Wal-Mart could take the lead in ensuring the products we buy for our kids are safe. But the company does not require manufacturers of toys, carriers, high chairs or other children’s products to demonstrate the products are safe before they wind up on a Wal-Mart shelf. The retailer does, however, flex its market power to insist that manufacturers cut costs. . . . Wal-Mart has enormous clout with manufacturers. The retailer should use this clout not only to insist its suppliers cut costs, but also to insist that manufacturers safety-test their products. A solid first step would be for Wal-Mart to require manufacturers of children’s products to certify that their goods have been safety tested by a truly independent third party, and that the products comply with meaningful safety standards. For the world’s largest retailer to take a bold position on safety would set a strong precedent for other retailers to follow. It is time for Wal-Mart to be part of the solution, rather than part of the problem.4
More than ten years later, not only has Walmart failed to lead on product safety but the Blitz cases indicate that little has changed.
Based partially on legal fees and settlements with plaintiffs from exploding gas can lawsuits, Blitz went bankrupt. Consequently many of the plaintiffs turned their attention to Walmart and sued the retailer as a causal agent in the deaths and injuries. Given that more than one party was involved in Walmart’s sale of unsafe gas cans, who is to blame?
A logical strategy for analyzing the role of different possible causal agents in gas can injuries would be to assess what the likely outcome would have been if one of the agents didn’t exist. Consider the counterfactual in which Blitz did not exist as a company during the years in question. Without Blitz, would Walmart likely have sold a gas can without an arrestor? Without Blitz, would Walmart have engaged in an effective communication campaign on the safe use of gas cans? My assessment is that Blitz would have been replaced by an alternative manufacturer, that Walmart would not have engaged in a safety campaign, and that little would have been different in terms of the safety of gas cans sold at Walmart.
Consider the alternative counterfactual, namely, that Walmart did not exist during this time. Would Blitz have created a safer gas can to sell to other buyers? Based on Blitz’s behavior, there is evidence that Blitz was concerned about improving the safety of its gas cans. Thus it is quite likely that Blitz would have brought a safer gas can to the marketplace.
This comparative analysis of these two counterfactuals suggests that Walmart was the driving force in unsafe gas cans being sold to consumers. While I believe this is the correct analysis, Walmart has yet to be found guilty in any such product liability suit.
A similar indirect effect of Walmart on safety—this time, the safety of those who make products sold by Walmart—can be found in the case of the 2012 garment factory fire in Bangladesh. On November 24, 2012, a fire broke out in the Tazreen Fashions factory in Dhaka, the capital of Bangladesh. At least 117 people died and at least another two hundred were injured, making this the deadliest factory fire in Bangladesh’s history. Subsequent analyses document that the factory failed to meet any reasonable set of safety standards.
Who is to blame? The owners of the factory, or the retailers that demand price levels that cannot be met if reasonable safety standards for factory workers are in place?
Let’s step back and consider this account of an interaction between garment manufacturers and more than a dozen Western retailers, including Gap Inc., Target, and JCPenney, that took place just a year and a half before the fire:
At the meeting in Dhaka, the Bangladesh capital, in April 2011, retailers discussed a contractually enforceable memorandum that would require them to pay Bangladesh factories prices high enough to cover costs of safety improvements. Sridevi Kalavakolanu, a Wal-Mart director of ethical sourcing, told attendees the company wouldn’t share the cost, according to Ineke Zeldenrust, international coordinator for the Clean Clothes Campaign, who attended the gathering. Kalavakolanu and her counterpart at Gap reiterated their position in a report folded into the meeting minutes, obtained by Bloomberg News.5
My argument is not intended to acquit factory owners of running unsafe facilities in order to generate greater profits. But like the gas can story, the root of the problem is that price pressure from Walmart and other retailers can lead to unsafe decisions by gas can manufacturers and factory owners. This pattern is being repeated across product categories in many nations.
When a company refuses to accept price increases to create a safer product, to educate consumers about product safety, and to pay extra to participate in making factories safe, it is a causal actor in creating harm. But as we will see throughout this chapter, people fail to hold organizations accountable when they are the indirect cause of harm. By definition, indirect harm often goes unnoticed and is particularly hard for people—manufacturers, retailers, and consumers—to see.
In the early 1990s Dr. Ladislas Lazaro IV, a Louisiana-based rheumatologist, used to occasionally prescribe an anti-inflammatory drug named H.P. Acthar Gel for the treatment of gout, as he told the New York Times. At the time, a five-milliliter vial of the drug cost patients $50. The drug disappeared from the market for a while, and Lazaro more or less forgot about it. When it became available again, the price had gone up a great deal. How much do you think H.P. Acthar Gel costs now? I assume that you inflated your number to adjust for the fact that I told you it went up a great deal. But did you come even close to the correct answer of $28,000 for the same five-milliliter vial?6
I have worked as a consultant for many pharmaceutical organizations and believe it is appropriate for drug companies to earn healthy profits when they create new solutions to our health challenges. Nevertheless I find this number amazing. How did we get from $50 per vial to $28,000?
The extremity of this price increase aside, the pattern of this story is surprisingly common in the world of “orphan drugs,” drugs that treat rare diseases and thus are produced in only small quantities. This particular orphan drug was owned and sold by the pharmaceutical company Rhône-Poulenc, which became Aventis following a merger. Rhône-Poulenc and Aventis were somewhat constrained from instituting dramatic price increases due to the public attention that large pharmaceuticals receive from the media. After selling only about a half a million dollars a year of Acthar, Aventis sold the intellectual property for the drug to Questcor, a much smaller pharmaceutical company, for $100,000 plus profit sharing. Because of its size, Questcor did not have much of a brand image to protect and was less concerned about publicity than Aventis, as is often true of the buyers of orphan drugs. Questcor immediately raised the price of Acthar to $700 per five-milliliter vial and continued to raise the price over the next decade until it reached $28,000. After taking over the sale of Acthar, Questcor began marketing it for a host of conditions, including multiple sclerosis, though there is little evidence that Acthar is more effective at treating these conditions than much cheaper drugs. Questcor CEO Don M. Bailey has told analysts that the new uses for Acthar amount to multibillion-dollar opportunities for his company.
Most observers, upon hearing that Questcor raised the price of a $50-per-vial drug to $28,000, have blamed Questcor and let Aventis off the hook. Few noted Aventis’s role in the price increase: that Aventis had knowingly sold the drug to a company that could be predicted to dramatically increase its price and then share the profits with Aventis. That is, Aventis got Questcor to do its dirty work.
My research on the topic of indirect causes of harm predates the Questcor story. In 2006 I read another New York Times article, this one about Merck, a major pharmaceutical manufacturer, which had produced a cancer drug called Mustargen.7 Merck faced the problem of having a small market share of loyal patients taking cancer medications. The profit-maximizing decision for this drug in particular would have been to raise prices, but that would have created negative publicity for Merck. So Merck sold the rights to the drug to Ovation Pharmaceuticals, a much smaller company that specializes in buying slow-selling medicines from big pharmaceutical companies. Ovation quickly raised the wholesale price of Mustargen by roughly ten times. Ovation made no investment in R&D for Mustargen. In fact it did not even manufacture the drug; Merck did, on a contract basis. Because Ovation is a much smaller company than Merck and lacks a well-known brand image, it was able to raise the drug’s price without attracting much attention. In this indirect manner, Merck was able to avoid the negative publicity of having directly raised the price of the drug tenfold.
The Merck story led Neeru Paharia, Karim Kassam, Joshua Greene, and me to team up to study the role of indirect dramatic price increases in a laboratory setting.8 We were interested in examining whether study participants, after careful reflection on their initial assessments, would hold an indirect actor more responsible for unethical behavior. In one of the six laboratory experiments reported in our paper, we described to participants a situation in which a major pharmaceutical company is the sole marketer of a particular cancer drug. All study participants read the following:
A major pharmaceutical company, X, had a cancer drug that was minimally profitable. The fixed costs were high and the market was limited. But, the patients who used the drug really needed it. The pharmaceutical was making the drug for $2.50/pill (all costs included) and was only selling it for $3/pill.
One group of participants also read the following:
A: The major pharmaceutical firm (X) raised the price of the drug from $3/pill to $9/pill.
A second group read about a different course of action:
B: The major pharmaceutical firm (X) sold the rights to a smaller pharmaceutical firm (Y). In order to recoup costs, company Y increased the price of the drug to $15/pill.
Our results show that most people do not notice the originating company’s culpability in the price increase. People who read Action A judged the behavior of firm X more harshly than did those who read Action B, despite the fact that Action B resulted in a larger price increase for consumers than Action A.
Across multiple studies, people fail to hold actors accountable for the indirect harms that they cause. My colleagues and I assessed this argument by presenting a third group of participants with both of the possible actions, A and B, simultaneously, and asked them to judge which was more unethical. The logic behind adding this third condition is that our past research has found that people act more rationally and reflectively when they compare two or more options at the same time.9 We found that when participants were able to compare the two scenarios, they reversed the preferences of those who looked at just one option and judged Action B to be more unethical than Action A. In short, when the indirect action was brought to their attention—when they were made to notice—they saw the reasonable causal link and assigned blame accordingly.
More broadly, when people are brought to think carefully about such ethical dilemmas, they hold indirect actors more accountable than they would otherwise. Reflection clarifies the obvious role of indirect actors in influencing the direct actor toward ethical misconduct.
I have heard many tough-minded executives use this phrase to sum up their management approach. Incentives are a key driver in a market economy, and one that I generally value. However, it is important to think through how people will actually respond to incentives, and too many executives do not do this often enough.
Take the retail market for computers and computer equipment. Due to product overlap among retailers and tough competition from Internet sales, the market is cutthroat and generates small profit margins. For many retailers, the real money comes from selling add-ons—peripherals, warranties, and service agreements—all of which have higher profit margins than the advertised laptops and printers. So the solution that many retailers have hit upon is to reward salespeople for selling these extras. What is the indirect result of this strategy, and should executives have anticipated it?
One organization that has rewarded salespeople for selling add-ons is Staples, the large U.S. office retailer. “The average needs to be $200,” Staples manager Natasja Shah told reporter David Segal for his New York Times feature “The Haggler.” That is, as part of an incentives system called “Market Basket,” each time a Staples employee sells a computer, he must sell about $200 worth of other stuff to meet his target. According to Shah, this average is carefully tracked. Staff who don’t meet their goals are coached. They can end up with lots of night and weekend shifts, a reduction in scheduled hours, and even disciplinary actions that can lead to termination. “If you can’t do the job, you can go sell fries at McDonald’s,” a Staples supervisor told one store manager who had concerns about the Market Basket system.10
Given that this chapter is about the indirect effects of unethical behavior, you already have a hint about how such a policy might affect Staples customers and the company’s reputation. But let’s take a look at this posting from an unhappy Staples customer from the consumer website www.my3cents.com, as reprinted by Segal in “The Haggler”:
I had an appalling experience in not one but TWO different Staples stores yesterday. . . . Staples featured an Acer laptop advertised in the Sunday newspaper insert yesterday (3/8) for $449. . . . Upon my arrival, I found an associate who informed me that the laptops were in stock. However, before he would get me one, he proceeded to try to sell me his “protection plan.” . . . I declined this. The sales associate indicated that it was ok, and then walked away to, I assumed, get my computer. He returned with the store’s general manager who again proceeded to aggressively push the protection plan onto me. He was EXTREMELY rude, implying that I was “cheap” for not adding the plan. He walked away when I finally maintained I did not want it. . . . Moments later, the sales associate informed me that the laptop was not in stock after all. Just to summarize the timeline. . . . It was available. . . . I declined the approximately $150 protection plan that would have boosted the price of the laptop to the regular sale price. . . . Then suddenly it was unavailable. . . .
I then went home and called a 2nd Staples store. . . . I asked the gentleman who I was transferred to in Electronics if the Acer laptop was in stock. He checked and came back to say they were in stock. . . . I arrived exactly 8 minutes later at the location. I walked in and found the exact person who I had spoken to 8 minutes earlier. The store was virtually empty. I asked him if I just spoke to him about the Acer laptop and he confirmed that he was the person. I asked him if they were indeed in stock, and he indicated that they were. I then asked if he could please go get one, because I definitely wanted one. And then, before he goes back to get one, he asks me if I want the service plan. . . . “No thanks . . . I went over this all at the last store, I know what it covers, and I am not interested.” . . . He said “fine” and walked away. 2–3 minutes later . . . he walks back and, just like the last store, the inventory suddenly has vanished.11
Several Staples employees confirm what this story and others like it suggest: that the company’s upper management expects store managers to tell employees who are unable to sell $200 worth of add-ons to tell the customer that the computer is not in stock. The policy is called “walking the customer,” Shah told Segal.
For decades, goal setting has been promoted as an effective means of managing and directing employees. Goals work well in many contexts, but it is critical to think about—and to notice—their broader impact. In particular, organizational leaders have a responsibility to think through the indirect effects of goals, such as how a policy such as Staples’s could affect customer satisfaction.
The problem of leaders failing to notice the indirect effects of organizational policies is hardly new. In the 1960s the Ford Motor Company was losing ground to foreign competitors as the market turned to smaller, more fuel-efficient cars. CEO Lee Iacocca announced the goal of producing by 1970 a new car that would be under 2,000 pounds and cost less than $2,000. To expedite the development of the Ford Pinto, managers pursued this target at all costs, including signing off on unperformed safety checks. In an effort to cut corners, engineers placed the fuel tank behind the rear axle of the car in less than ten inches of crush space. Lawsuits later documented that it should have been obvious this design could cause the Pinto to ignite upon impact. Yet even after Ford identified the hazard, executives remained committed to their goal and failed to repair the faulty design. They calculated that the cost of lawsuits associated with Pinto fires (which went on to cause fifty-three deaths and many injuries) would be less than the cost of improving the design.12
Similarly consider Sears, Roebuck and Company’s experience with goal setting in the early 1990s. Sears set a sales quota for its auto repair staff: $147 in sales per hour. This goal led employees to overcharge for their work and to sell unnecessary repairs to customers. Eventually Sears chairman Edward Brennan admitted that the goal had motivated Sears employees to deceive customers.13
In a December 2012 New York Times editorial, Northwestern University professor Carolyn Chen argued that Asian Americans are being discriminated against by the admissions offices of the top U.S. colleges and universities. She notes that Asian Americans make up anywhere from “40 to 70 percent of the student population at top public high schools,” where admissions are largely based on objective criteria such as exam scores and class grades.14 Nonetheless white students were three times more likely to be admitted to selective U.S. universities than Asian Americans with the same academic record, according to a 2009 study by sociologists Thomas J. Espenshade and Alexandria Walton Radford.15 Asian Americans must score 140 points higher, on average, than whites on the math and verbal sections of the SAT in order to have the same chances of admission at the most selective universities, Espenshade and Radford concluded.
If applicants were reviewed solely on objective criteria (grades, test scores, academic honors, and extracurricular activities), then many more Asian Americans would be admitted to top schools, Chen and others have argued. Of course, Asian Americans might vary, on average, from white applicants on softer criteria, such as letters of recommendation and interviews with staff and alumni. If this argument sounds familiar, Chen reminds us of what happened in the 1920s, when Jews first began to compete with prep school students for admission to the top universities (including the one where I teach, Harvard). To cap Jewish enrollment, the universities started asking applicants about their family background and “sought vague qualities like ‘character,’ ‘vigor,’ ‘manliness,’ and ‘leadership,’ ” writes Chen. Significant evidence suggests that informal versions of the quotas against Jews at Harvard, Yale, Princeton, and other Ivies lasted into the 1960s.
These statistics and others have raised suspicions about whether top colleges and universities are deliberately limiting the number of spots that they give to Asian American applicants. Do the Ivies and other elite schools have quotas for Asian Americans, just as they instituted quotas on Jews in the 1920s? Personally I do not believe that deliberate quota systems exist. But I do believe that the type of implicit racism that Chen describes—discrimination based on biases such as stereotypes and in-group favoritism—does exist and that it contributes to qualified Asian Americans being rejected from elite schools in favor of less qualified whites.
I find the issue of implicit discrimination fascinating and disturbing. We commonly fail to notice when our behavior or that of others hurts people indirectly. More specifically, I believe that elite universities are arriving at perverse outcomes without having any deliberate intention to discriminate. As part of a series of New York Times commentaries that accompanied the Chen piece, scholars John Brittain and Richard Kahlenberg argue that one source of discrimination against Asian American college applicants comes from the indirect effect of a different (implicit) form of racism, namely legacy admissions.16 As I discussed in Chapter 7, most of the top U.S. universities have programs that give preferential admission to the children of alumni. This practice advantages the already advantaged. When legacies who are barely qualified are admitted, they take spots away from superior applicants. Notably the legacies at top universities tend to be Caucasian, given the past predominance of whites at these schools. Meanwhile the applicants who are rejected at the margin are much more likely to be of other races, including Asian American.
When Harvard was investigated in the 1990s for discriminating against Asian American applicants, the U.S. Department of Education’s Office for Civil Rights concluded that one of the reasons Asian Americans fared less well than similarly qualified white applicants was that they were being squeezed out by legacies.17 The discrimination against top Asian American high school students is in part an indirect effect of the policy of providing advantage to the children of those who attended our top universities.
Walmart buyers are doing their job when they insist on rock-bottom prices from their suppliers. Rhône-Poulenc executives were meeting firm objectives when they concluded that selling H.P. Acthar Gel to a smaller firm would be more profitable than keeping the drug in house. Staples employees who sell unneeded extras are doing something that salespeople do quite regularly. And the admissions officers who give extra consideration to the children of wealthy alumni are following procedures that have been in place for decades. Yet the indirect effects of these behaviors are highly problematic.
The harms created by indirect effects are a leadership challenge. Leaders need to think beyond the moment to anticipate the problems that their organization’s procedures could create. When an organization sets up processes that lead people to unintentionally discriminate against a particular group, the organization itself is engaging in discrimination. The organization’s leaders are responsible for noticing this and for making the changes that are needed to prevent indirect harms from occurring.
Interestingly, if a group of executives had been charged to think about the potential downsides of the policies and practices described in this chapter on their organizations or on society, they would easily have identified the indirect harms that emerged. Unfortunately organizations rarely conduct such exercises. There is no lack of ability to solve these problems in our leading organizations but rather a failure of leadership imagination to anticipate their logical consequences.