Under-Recognition of the Power of Reinforcement and Incentives

“I’ve been in the top 5% of my age cohort all my life in understanding the power of incentives, and all my life I’ve underestimated it. And never a year passes but I get some surprise that pushes my limit a little farther.” –Charlie Munger

This article is part of a multi-part series on human misjudgment by Phil Ordway, managing principal of Anabatic Investment Partners.

Munger’s favorite cases were Federal Express, which was finally able to fix its system by paying night-shift workers by the shift instead of the hour; Xerox, which had commission arrangements that gave an incentive to an older, inferior machine; and B.F. Skinner, the eminent researcher who made many important psychological discoveries.

Update

Later called “Reward and Punishment Superrresponse Tendency.” In Munger’s revision, he added several new examples:

• Mark Twain’s cat that, after sitting on a hot stove, never sat on another stove (hot or cold) ever again;
• Ben Franklin’s maxims that “if you would persuade, appeal to interest and not to reason” and “never, ever, think about something else when you should be thinking about the power of incentives”;
• the Soviet employee’s phrase that “they pretend to pay us and we pretend to work”;
• “Perhaps the most important rule in management is ‘Get the incentives right’”;
• “Another generalized consequence of incentive-caused bias is that man tends to ‘game’ all human systems, often displaying great ingenuity in wrong serving himself at the expense of others. Anti-gaming features, therefore, constitute a huge and necessary part of almost all system design”;
• “Dread, and avoid as much as you can, rewarding people for what can be easily faked. Yet our legislators and judge…often ignore this injunction”;
• “Punishments also strongly influence behavior and cognition, although not so flexibly and wonderfully as rewards.” Price fixing was more common when it was met with fines rather than jail time. A European tribe in the time of Caesar, when the assembly horn blew, always killed the last warrior to reach his assigned place. And George Washington hanged deserters at a great height as an example to others who might think of deserting.

Wells Fargo recently demonstrated the power of incentives with its “fake accounts” scandal. The cause and effect of this situation reads like a script for this talk, with amazing parallels to the Salomon scandal of 1991. The Wells Fargo situation was a straightforward one: the legacy Norwest culture of “cross-selling” had been enormously successful, but the incentives involved got stretched past their breaking point and then denial, more incentive-caused bias, commitment and consistency, Persian messenger syndrome, and social proof, among other tendencies, combined to allow the problem to morph into a true scandal.

At the heart of the issue is that management and the board made efforts – some genuine, some half-hearted – to address the problem before it spread. Numerous employees raised the issue to the head of the community bank, Carrie Tolstedt, and to the company’s senior executives. In almost all cases the evidence was ignored or suppressed. There were massive compliance manuals and plenty of procedures and policies, but culture and leadership trumped the manuals and policies as they often do.

And the problem festered for years. The first signs came as early as 2004 when an internal investigations unit noticed a rise in “sales integrity” issues. The report said, “Whether real or perceived, team members . . . feel they cannot make sales goals without gaming the system. The incentive to cheat is based on the fear of losing their jobs.” And many people did get fired or quit. The sky-high turnover rate (42% in 2012 among all branch and call center employees) was a red flag itself, but it was dismissed internally as “slightly lower than turnover for similar types of roles in the retail industry” (emphasis added). The 2004 report “recommended that Wells consider reducing or eliminating sales goals, as several peer banks had done, and warned that the issue could lead to ‘loss of business and . . . diminished reputation in the community.’” Wells did none of that until 2016, of course. But almost a decade after the first report, a Los Angeles Time story broke the news that the issue had festered and grown over the years. And as several more years went by, management did almost nothing to address the problem or to manage the fallout when the settlement – dismissed as small relative to Wells Fargo’s financial size – was inevitably announced. The Board made some attempts that I believe were more genuine than those often seen in such circumstances, but inertia and Stumpf’s opposition slowed things down.

The parallels to John Gutfreund and the scandal at Salomon Brothers are stark. Gutfreund’s downfall resulted from his response to Paul Mozer’s misdeeds. Had Gutfreund moved aggressive to snuff out the problem, most or all of the damage could have been avoided. But Gutfreund fell victim to reciprocation and the fact that Mozer’s group had made enormous amount of money for Salomon. At Wells, John Stumpf had repeated praised Carrie Tolstedt in public as

The Board’s official report spelled it out explicitly:

“Stumpf’s long-standing working relationship with Tolstedt influenced his judgment as well. Tolstedt reported to Stumpf until late 2015 and he admired her as a banker and for the contributions she made to the Community Bank over many years. At the same time, he was aware that many doubted that she remained the right person to lead the Community Bank in the face of sales practice revelations, including the Board’s lead independent director and the head of its Risk Committee. Stumpf nonetheless moved too slowly to address the management issue.

“[Stumpf] was not perceived within Wells Fargo as someone who wanted to hear bad news or deal with conflict. In accordance with the decentralized model, a deferential culture existed whereby there was limited encouragement for the management of different businesses to challenge each other or comment on significant issues in the other lines of business. Under Stumpf, weekly Operating Committee meetings generally did not serve as a forum for discussion, engagement or challenge among its members.

“Stumpf wrote that Tolstedt ‘knows the business cold – nothing gets by her’ and that a management structure she had devised was a ‘stroke of genius.’ In 2013, Stumpf attributed the Community Bank’s success, including in achieving ‘record cross-sell,’ to Tolstedt’s leadership. He also pushed Tolstedt to work to increase cross-sell when strong growth proved more elusive. Many observers expressed their belief that Tolstedt operated the Community Bank in the way she did because she thought Stumpf would approve. This is also supported by contemporaneous emails, particularly with respect to the setting of aspirational sales goals and focus on improving cross-sell.

“[Tolstedt] was credited with the Community Bank’s strong financial results over the years, and was perceived as someone who ran a ‘tight ship’ with everything ‘buttoned down.’ Community Bank employee engagement and customer satisfaction surveys reinforced the positive view of her leadership and management. Stumpf had enormous respect for Tolstedt’s intellect, work ethic, acumen and discipline, and thought she was the ‘most brilliant’ Community Banker he had ever met. Nonetheless, Tolstedt mismanaged the Community Bank’s response to the rise in sales practice issues, failing to appreciate both the negative impact on customers and the grave risk to Wells Fargo’s brand and reputation. There is no evidence that Tolstedt showed serious concern about the effects of improper sales practices on Wells Fargo’s customers or that she initiated efforts to evaluate or remediate customer harm. Tolstedt resisted change to the Community Bank’s sales model even when confronted with evidence that it led to low quality sales and improper sales practices. She viewed the sales model as an engine of the Community Bank’s historical success and did not want to take steps that could impede its operation. Instead, she reinforced the high-pressure sales culture…. Despite the universal criticism of the [incentive] program as an incubator of low quality sales and bad sales practices, Tolstedt was ‘scared to death’ that changing it could hurt sales figures for the entire year and opted instead for only incremental changes. Numerous witnesses provided a consistent account of Tolstedt’s management style: she was ‘obsessed’ with control, especially of negative information about the Community Bank, and extremely reluctant to make changes. Tolstedt fostered an insular culture at the top of the Community Bank and had an ‘inner circle’ of staff that supported her, reinforced her views and protected her. She resisted and rejected the near-unanimous view of senior regional bank leaders that the sales goals were unreasonable and led to negative outcomes and improper behavior.”

Tolstedt instructed her team members to avoid talking to other company executives without her presence. She went so far as to actively suppress evidence and mislead the board: “Tolstedt never voluntarily escalated sales practice issues, and, when called upon specifically to do so, she and the Community Bank provided reports that were generalized, incomplete and viewed by many as misleading… By 2015, many Board members believed that she was intentionally understating the problem which she had helped to create.” ^[5]

Even if much of the problem could have been avoided with a proper response, the core problem of overly aggressive incentives is a common one. In fact, Munger himself recently argued as much:

“Wells Fargo had a glitch. They made a business judgment that was wrong. They got so caught up in cross selling and having tough incentive systems that they got the incentive system so aggressive that some people reacted badly and did things they shouldn’t. Then they used some misjudgment in reacting to the trouble they got in. I don’t think anything is fundamentally wrong with Wells Fargo for the long pull with Wells Fargo. They made a mistake. It was an easy mistake to make and the smartest man I ever knew made a similar mistake. Henry Singleton was the smartest single human being I’ve ever known in my whole life…and at Teledyne [he] also had very aggressive incentive systems like Wells Fargo. And his customer was…the government and of course it’s not that hard to cheat the government and two or three of 20 subsidiaries cheated the government. It’s not that Henry was trying to cheat the government it’s just that he got a little aggressive trying to apply the incentives and he got blinded sided. That can happen to anybody. I don’t regard getting the incentives a little aggressive as Wells Fargo did as the mistake. The mistake there was when the bad news came they didn’t recognize it rightly…How do you know [incentives] are aggressive until you try? They didn’t react enough to the bad news fast enough. And of course, that is a very dangerous thing to do. I don’t think it impairs the future for Wells Fargo. As a matter of fact they’ll probably be better for it. One nice thing about doing something dumb is that you probably won’t do it again. ”^[6]

An even more vivid example of the power of incentives – combined with social proof — might be Valeant. This has to be one of the important cases in business and investing of the past decade. To be clear, there are no stones being thrown out of a glass house – it is genuinely humbling to look at the Valeant shareholder list at the peak in 2014-15. A lot has been written and said, of course, about this episode, but it’s worth going through it and having its various attributes and the tendencies behind them at our ready disposal.

Financial institutions in general are rife with incentive-caused bias. The Global Financial Crisis (“GFC”) itself had many causes, of course, and as with all insane outcomes it was the combination that enabled the enormity. But right at the heart of the bubble was the fact that so many people had perverse incentives that all acted together. Securitization was a microcosm of the problem. It started out – like so many ideas in finance – as a novel and genuinely useful practice. But the incentives combined with other tendencies to metastasize into a monster. A good idea was taken too far, and generating more “product” became the only goal. There were other obvious examples found in realtors, mortgage originators, structured products traders on Wall Street, homebuilders, appraisers, rating agencies and many, many others. Almost everyone had incentives that either directly rewarded the ultimately destructive behavior or at least encouraged people to look the other way.

Consider also the fees charged by investment funds. Many complain about closet-indexing, but what do the incentives really encourage? The same is true of the capital allocators at pensions, endowments, and funds of funds. Career risk – incentive-caused bias – far outweighs the risk of underperformance. An investment officer at a major U.S. endowment once told me: “If I hire a brand-name fund and it blows up, that’s bad luck or somebody else’s fault. If I hire a small fund run by a less prominent manager without pedigree or prestige, and then he has a bad year, I get fired.” Jeremy Grantham wrote that “the central truth of the investment business is that investment behavior is driven by career risk.”^[7] There he’s in agreement with Keynes who said, “Worldly wisdom teaches that it is better for reputation to fail conventionally than to succeed unconventionally.”^[8]

Self-interest in politics combined with business can create a potent brew. Consider ethanol – it can’t make sense to even the most narrow-minded Iowa farmer to take his precious topsoil and use it as a substitute for readily available alternative fuels, but a higher price for his crops and the quirks of the American primary voting system have given us ethanol mandates nonetheless.

The Costco business model and its membership fee is a great example of this process put to good use in a business. Amazon Prime is a more modern incarnation. Few things are more powerful than a business model with built-in incentives that also make its own business stronger. The incentives and the reinforcement work to everyone’s mutual benefit. By charging a “membership fee,” Amazon and Costco customers have a subtle incentive to shop more to “earn back” the fee. That volume creates a high-turnover business that can use the savings and purchasing advantages to reinvest in lower prices. On and on it goes.

Speaking of Costco, I recently had my one and only bad experience when I had to buy a new air conditioner. I didn’t know much and hadn’t done my homework yet, but I knew Costco air conditioners through a local dealer in each market. The salesman I got – who was working on commission – tried to sell me a 26 SEER unit that would have been appropriate had I lived in Death Valley instead of suburban Chicago. But he kept trying to convince me that if I wanted to make the best investment in my home – at one point insisting that if I really loved my wife and kids – a $22,000 A/C unit that could cool an aircraft hangar in Panama City was the only way to go.

[5] Sales Practices Investigation Report and “How Wells Fargo’s Cutthroat Corporate Culture Allegedly Drove Bankers to Fraud”

[6] Charlie Munger at the Daily Journal Annual Meeting

[7] Jeremy Grantham: “My Sister’s Pension Assets and Agency Problems (The Tension between Protecting Your Job or Your Clients’ Money).” April 2012.

[8] John Maynard Keynes: The General Theory of Employment, Interest, and Money.