In a knowledge economy, a firm’s long-term profits depend on its people. That is less true in a manufacturing economy, in which routine workers are often interchangeable. The same logic applies to universities, research centers, and government agents. The concern for long-term success implies an emphasis on attracting and retaining talented and diverse people.

Firms care about profits, to be sure, but those that maintain long-term success also care about developing future leaders. An executive remarked to me that it was not a coincidence that “succession management” embeds “success.” We see these in the actions executives take. In the first weeks of the Donald J. Trump administration, Tim Cook (Apple), Jeff Bezos (Amazon), Sheryl Sandberg (Facebook), and Larry Page (Google) all posed with President Trump in support of a proposed policy change that would allow US firms to repatriate monies held overseas.

At that time, American corporations held more than $2.5 trillion overseas and sought to reduce the 35 percent corporate tax rate to 10 percent. The result of that policy change would be over $500 billion extra in profits. Apple alone had more than $200 billion overseas at that time. Their $50 billion in savings would exceed their 2016 profits.

Less than a week after that photo was taken, the Trump administration issued an executive order limiting immigration and travel to and from the United States. Apple’s CEO, Tim Cook, spoke out against this policy. He did so because Apple employs people from all over the world and needs those people to be able to travel and communicate freely. Other CEOs made similar critical comments.

These CEOs took the expected positions on both policies. A concern for short-term profits led to their support for the proposed tax changes—what CEO would not want an extra $50 billion in profits? A concern for their long-term success and its dependence on a diverse, talented workforce led them to criticize the changes to immigration policy.

Two clear impressions emerge from visiting successful firms and organizations. First, they focus on their core mission, and second, they seek and value talented workers. These are not just my takeaways from visiting hundreds of corporations, universities, and organizations. They can also be found in any number of management and strategy books.

Given those impressions, if diversity and inclusion are framed as the right thing to do, they cannot have much force. In addition to a misalignment with the main goals, that framing positions diversity and inclusion as one of many do-good policies a corporation might pursue. This places diversity and inclusion initiatives in competition with other “oughts,” such as sustainability, charity work, and service to the local community.

Thus, as lofty as the normative goals may be, we cannot expect corporate leaders to make promoting a multicultural society or correcting social disparities their primary concern. Corporations look to the bottom line. As much as advocates might want corporations to privilege normative desires for shared understandings and social justice above base concerns for profits, that will be a hard sell. Normative-based inclusion efforts not linked to a business case run the risk of degenerating into halfhearted, symbolic, feel-good gestures. For corporations to take meaningful action, they must see diversity and inclusion in their self-interest. They must believe in the business case.

An idealist might dismiss these arguments as overly cynical and argue that we should work toward creating a world in which the normative arguments carry the day, in which individuals have climbed sufficiently far up Maslow’s hierarchy of needs to commit to economic reparations and affirmative action policies, and in which we all embrace and pursue inclusion and integration.

Let us suppose we get to that place, that we live in a world where people support affirmative action on normative grounds. I will argue that even in that world—in fact, especially in that world—the diversity-bonus logic is necessary, as it shows how to make inclusion successful.

If we do not understand how diversity produces bonuses, we cannot put our diverse repertoires to best use. We might even put them to improper use. Bonus logic informs emergency room doctors to put lots of eyeballs on novel and critical presentations and to apply checklists for sprained ankles. On the difficult problems, we should seek medical professionals with diverse training, case experiences, and identity attributes. Social justice logic offers no such guidance. If anything, normative logic argues for identity-representative panels of doctors on all cases. That would waste time, cost money, and result in worse health outcomes.

Diversity-bonus logic provides insights into how to constitute teams and how large those teams should be. How many doctors should we assign to a patient? How many economists, business people, and lawyers to a federal bailout program?

The diversity-bonus logic also points to useful behaviors. On creative teams, people must be willing to share their ideas, no matter how crazy. On predictive tasks, we need people with novel categorizations and causal models. When problem solving, people must take proposed solutions seriously and look to recombine them to create even better solutions. These behaviors require trust and personal validation, characteristics found in successful groups.25

While trust and validation align with social justice thinking, other behaviors that can generate diversity bonuses need not. On critical decisions, asking someone to play the role of devil’s advocate and to challenge every idea can add value. That may not be seen as normatively inclusive.

Even worse, an overemphasis on normative concerns can result in satisficing, in which every decision or action must be approved by each individual. Thus, a team tasked with coming up with a new design or policy can wind up with an incoherent collection of parts meant to satisfy people with different goals. The saying that a camel is a horse designed by a committee is meant to capture this type of group failure. One person wants the horse to have long legs to go fast; another wants the horse to have big strong feet to be sturdy; and a third wants the horse to be able to travel long distances.

Set aside for the moment that these attributes make for an animal well suited to the desert, and focus on the fact that they make for a lousy horse. What we learn is that if people have different goals, diverse groups will not produce good outcomes. We need only look to Congress to see this to be true. Elected officials who must satisfy multiple constituencies to remain in office often fall victim to the camel problem. Diverse groups that evaluate solutions according to a shared set of criteria avoid the camel.26

The contrast should be clear. The bonus logic promotes reasoned, outcome-focused, inclusive diversity, and that can require challenging people’s ideas and sometimes making choices that favor one person over another. The normative logic promotes representative diversity and fair, rather than efficient, outcomes.

Thus, to the extent that normative arguments offer incomplete and counterproductive guidance for how to achieve bonuses, they make inclusion appear to be less in our self-interest than it could be.27 So while bonus-logic thinking may produce less immediate representativeness than normative thinking, evidence shows it generates better outcomes.28 And if people experience direct benefits from diversity and inclusion, then those policies will have more support. Thus, paradoxically, a long-term goal of inclusion on normative grounds may be better attained by also promoting inclusion on pragmatic grounds.

Colleges and universities pursue this approach. Almost all university and college diversity and inclusion statements mention some form of diversity bonuses in addition to their normative interest. They emphasize how interactions among groups of diverse students produce deeper understandings (a bonus), better solutions to problems (a second bonus), and new areas of inquiry (a third bonus). These potentials for direct benefits, even for universities and colleges that have a strong normative commitment to inclusion, endow the bonus logic with a power that normative arguments lack.