As Neel Doshi and Lindsay McGregor share in Primed to Perform, “no topic raises passion like pay.” And as they reflect, “pay-for-performance is neither inherently good nor inherently bad. Depending on the circumstances, it can be either, both, or neither.” There! Problem solved? Clearly not, and the authors go on to describe how the key to great performance “is knowing when pay-for-performance works and when it doesn’t.”

Indeed, corporate incentives have rightly taken a central role in discussions of risk, compliance, and ethics. Incentives are an inherent feature of every corporation, and a key part of their governance structures. They are designed to drive good behavior and performance, but they can also invariably drive conduct beyond what was intended, and sometimes far worse than top management could have envisioned. Perverse incentives can be so insidious that they can drive good people to do things that they look back on with surprise and regret, usually when it’s too late. But such incentives lurk in every major corporation, waiting to be triggered by the right set of circumstances, like hidden land mines ready to explode, plastering the results on the front page of major newspapers.

We recognize the general feeling that a workforce with ethics in their DNA can’t be tipped by incentives to behave badly. Neither should we have to incentivize people to behave ethically; however, this perspective ignores the reality that incentives can also penalize people for behaving well. Yes, cheating is a choice, and a discussion of incentives isn’t a war on business objectives. Nor is this an exercise in scapegoating. But the same incentives that can potentially reward bad behavior can also effectively penalize good behavior, which then undermines compliance programs, and erodes corporate culture. Still unconvinced? Well, perhaps we should take Wells Fargo TV commercials touting that they have stopped “cross selling” as a clear signal that harmful incentives are no longer a niche compliance conversation; it’s literally a prime-time challenge.

The good news is that the growing focus on incentives has forced a recognition by top management and boards of directors, that hard-won corporate reputations are too easy to lose over conduct in what’s often a small corner of a vast enterprise. These problems are avoidable if you know what to look for. A good start is a holistic, systematic, and dispassionate review of incentives, compensation, and controls.

Carrots, controls, and culture. When Wells Fargo imposed a target of eight accounts per customer on its branches, they created an aspirational goal. Cross-selling is a proven method of encouraging growth and maximizing upside, with little incremental expense. But when executives made managers’ bonuses dependent on the degree to which that sales goal was being achieved, and the managers, in turn, began checking their people’s progress towards those goals twice a day, they created a pressure cooker. When employees began losing their jobs for falling short of those goals, and observed the termination of their peers, the ethical fabric of the organization began to fray. And when management began sidelining people who complained about those pressures, and failed to conclusively shut down what they knew was occurring, they knocked away controls that might have contained a system headed towards catastrophe. It was an outcome that could have hardly been imagined when the goals were designed, approved, and articulated. And it was certainly an outcome that no one wanted—not management, employees, shareholders, or customers.

Perverse incentives can be so insidious that they can drive good people to do things that they look back on with surprise and regret, usually when it’s too late. But such incentives lurk in every major corporation, waiting to be triggered by the right set of circumstances, like hidden land mines ready to explode, plastering the results on the front page of major newspapers.

When does it start to go wrong? One may ask where in this progression does an incentive to perform transform into an incentive to cheat. The answer is: the moment the incentive is put into place. There are innumerable ways to achieve any goal, especially a goal like production, sales, or anything else that can be captured in a pay-for-performance metric. By design, such incentives will generally align with the health and culture of the organization, but only up to a point. While initial results might demonstrate success, later incremental growth might start to come from maladaptive strategies that diverge from the greater organizational ethos. When goals linked to reasonable sales growth start to “stretch” and run ahead of market realities, then cutting a corner looks like a reasonable and morally acceptable way to get “home” in a tough quarter or pay period.

Incentives are a loud, unspoken message as to “what management really wants.” Putting them into place requires thinking about them the way doctors think about drugs. You can’t just consider their intended effects; you also have to consider their side effects and interactions. And it’s not just the formal, financial rewards embodied in a commission schedule or bonus plan that motivate employees. Informal incentives such as the promise of promotion, the threat of losing one’s job, or the prospects of special recognition, are also potent drivers of behavior. Not all measures can be found in an employment contract. Incentives have to be viewed—and reviewed—holistically.

When controls and culture aren’t enough. Companies can contain the excesses of their incentive programs with controls. That works well most of the time. But controls come at a cost. A robust compliance and control environment might often mean that some individual or business unit gets poorer results now, in exchange for longer-term or bigger-picture gains that they may not see in their paycheck. That can pit business executives against compliance personnel in a struggle for “whose voice is the loudest.” We have seen a number of cases of corporate misconduct where controls that existed were ignored or overridden, which can be more dangerous than no controls at all. That’s compliance without consequences.

People engage in channel stuffing, holding back orders, or not writing down bad debt to meet their objectives, often with knowledge of supervisors, because the design of their incentives relies on the existence of controls that are either not taken seriously, or distorted to the needs of business objectives. Here, we often see middle level management playing the odds of either not getting caught, or thinking they are acting with a managerial “wink and nod.”

When that happens, organizational ethical tone and compliance gets discarded during sales roll-up and status calls, to the temporary improvement of financial results. But the benefit quickly wears off. Worse, channel stuffing in one current quarter merely robs from organic sales in subsequent periods and can result in unnecessary warehouse fees (if the customer is not ready to take possession) and discounting. Here’s yet another unintended consequence of a bad incentive system: the squeezing of margins to make short-term targets. In such cases, the underlying bad (and perhaps illegal) behavior hides behind the “good” performance. As Warren Buffet once noted, “only after the tide rolls out do we find out who was swimming naked.”

Compensatory cliffs, caps, and unrealistic goals. The root causes of perverse incentives often hide (in plain sight) in the most common variable compensation structures, which include payout floors, thresholds, payout caps, and stretch goals.

The seeds of the Epipen scandal were planted when management became eligible for millions of dollars of a one-time bonus payout, if they hit 90 percent of their cumulative earnings target. Such “threshold” performance levels are a feature found in most incentive plans. But this particular element meant that management stood to lose out on millions of dollars if they merely achieved 89.9 percent of their goal. What would you do if your team were $89 million toward a $90 million threshold with just a few weeks to go? What would you do for that last million? The last $100,000?

A large segment of all corporate scandals in the last couple of decades have been associated with steep incentive plan thresholds. What makes them land mines is not their ubiquity, but that they are invisible until the company steps into the wrong circumstances. Think of them as compensatory cliffs, where a minor misstep at the end of a long road can lead to devastation. If management comfortably soars beyond the threshold, nobody notices the peril ahead. If performance is looking well below threshold, with no chance for awards, the cliff is too far to be seen. But what happens when businesses find themselves perilously close to the cliff, which is the area just short of threshold performance. In other words, almost at the finish line.

It might be that a manager never had to face the temptation created by a steep payout cliff. And because organizations can live with these dangerous features for so long without incident, they are lulled into thinking that the plan structure is sound. But when they find themselves facing that cliff, certain behaviors that they would have condemned as shortsighted, such as channel stuffing, accounting manipulation, or bribery, suddenly seem worth considering. It’s the moment of “what does everyone really want,” success or compliance, because “now I can’t deliver both.”

Avoid the cliff. The good news is that steep thresholds are almost always avoidable. While an oft-used compensatory practice, companies can significantly reduce or eliminate these unnecessary features without any loss of motivation or accountability. A simple proportional payout to plan targets can eliminate the risk. Even if you wanted to ramp up incentives as the goal gets closer, let’s call it a “hockey stick” payout, that still has less behavioral hazard than an all or none “cliff.”

Beyond the threshold, “high leverage” plans—whereby incrementally higher performance results in very high additional rewards—encourage a focus on performance with a sense of urgency; that is the intended effect. The sooner you make plans, the sooner the higher payout structure takes effect. But this can also encourage behavior that is riskier than the company would prefer. It also encourages significant dishonesty during the planning process, as leaders will try to “sandbag” their goals as much as possible to get to the next level of payout. Thus, organizations would be well advised to look at scenarios during the planning season in order to identify where incentives may need to be de-leveraged, in order to create a better balance between motivation and risk.

What about bonus caps? Can they create danger? Consider a salesperson that has reached his bonus cap three months before fiscal year-end. Any achievement after that is not only unrewarded, but is actually penalized in the sense of creating higher future performance expectations and targets. That’s called ratcheting. A creative person with good relationships can conspire with their customer to squirrel away sales into the following year, i.e., the opposite of channel stuffing. Thus, a sales person may end up in an informal and unreported financial agreement with a customer to delay recording a sale that may run afoul of revenue recognition rules. Material? Maybe. A breach of ethics and compliance? Definitely.

Sandbagger! With regard to target- or goal-setting, sales executives have long responded to this process with sandbagging, i.e., pushing the easiest targets possible to insure “achievable” goals. Junior sales managers play this game with those senior sales executives as they roll up their forecasts and targets. Business unit leaders respond by implementing “stretch goals,” which run ahead of market realities, and eventually a budgeting tug-of-war ensues. This multilayer organizational dance distorts the sanity of the resulting goals. It is inherently dishonest. If you are a compliance manager and have not witnessed one of these roll up meetings, now is a good time to start. After all, how can we expect integrity in incentive plans if the process is based on deliberate hedging and stretching during the forecasting period?

Let’s remember, there’s nothing wrong with stretch goals per se. Wells Fargo expecting their branch teams to cross-sell might have compelled employees to be legitimately creative in cross-selling their services. If you belong to a gym, you have probably been incentivized to bring in potential members through gift cards, free training sessions, and the like. That’s cross-selling at its designed purpose, i.e., to use existing customers as a base to grow the enterprise. But communicating to employees that they will be sharply penalized for failing to achieve those goals creates a strong likelihood of risky behavior. So, while a stretch goal that has some relationship to existing market potential is realistic, harsh consequences for not achieving those targets can endanger ethics, culture, and compliance.

Identifying unrealistic goals is a key element of an incentive mine sweep. You might start by asking for unfiltered, and perhaps anonymous, information from your line personnel about their perceptions of existing compensation structures and engage a market study to better understand industry growth patterns. Indexing your goals to the market, and getting unfettered data from those whose compensation is linked to “pay for performance,” might provide surprising results, which allow you to disarm those mines before they are front-page news.

It’s risky business, but stay the course. Searching for incentive land mines means messing with people’s pay. You are not just looking for potentially dangerous design elements, but also identifying changes to variable plans that preserve as much of the positive power of your incentives as possible, while addressing the risks. This requires a level of trust by plan recipients that the people doing this work understand the pressures faced by front-line employees and have experience with incentives and related governance structures.

This can present an unusual challenge for financial, legal, and compliance leaders. The process forces them to ask: “How do I contribute to the planning and execution of these plans before and after they hit the front-lines of our business?” Too often, compliance, ethics, and even HR leaders don’t have a seat at that table, so expect push-back with any change. A good incentive package benefits everyone; however, some might not give it time, and leave the organization, wanting more lucrative and short-term upside, especially where competitors flout such packages. Preparing for those challenges in advance helps to insure a process that embraces all organizational perspectives as well as individual concerns.

A critical analysis of incentives isn’t an attack on goal-setting or the rewarding of performance. It’s recognition that with incentives you don’t just get what you pay for, you can get a lot more, and some of that ‘more’ might not be welcomed or intended. Remember, nearly all of the recent scandals were in organizations that had robust and well-articulated codes of ethics and compliance programs. But their incentives and controls combined to create a lethal unspoken message that cutting corners was tolerable, even desirable. Think and ask, is it worth considering an investment—internal and/or external—toward making sure that your incentives do not create hidden land mines?

This article was co-written by Richard Bistrong and Marc Hodak.

Richard Bistrong is the CEO of Front-Line Anti-Bribery LLC,  and a contributing editor of the FCPA Blog.  He was a former International Sales VP,  FCPA violator,  and FBI/UK cooperator, having served fourteen months in Federal Prison for violating the FCPA, returning home in December 2013.  He now  consults, writes and speaks about FCPA, corruption and compliance issues  from his front-line perspective. In 2015 he was named one of Ethisphere’s 100 Most Influential in Business Ethics. He can be contacted by email at richardtbistrong@gmail.com, or his website at www.richardbistrong.com  He regularly tweets on anti-bribery & compliance issues at @richardbistrong.

Marc Hodak is founder of Hodak Value Advisors, a corporate advisory firm specializing in corporate performance measurement, management incentives, and related governance issues for investors, boards, and senior executives. Before forming Hodak Value Advisors in 2002, Marc led value-based management implementations and related projects for Stern Stewart & Co. efore that, Marc was a Chairman’s Award winning manager of a major business unit at Conrail, one of the largest U.S. transportation companies at that time. Marc is an accomplished author and blogger who has been published in both academic journals and popular magazines, including Forbes, Chief Executive, and NACD Directorship. He has been quoted in the Wall Street Journal, Bloomberg, BBC, and numerous other outlets. Marc has taught corporate governance as a professor at NYU’s Stern School and as visiting lecturer at the University of St. Gallen in Switzerland.