What A CEO Cannot Delegate

Famed University of Chicago economist Richard Thaler ran a thought experiment with a large company, asking executives to evaluate an investment scenario: Suppose there existed an investment opportunity in a corporate division that would yield one of two payoffs. After the investment was made, there was a 50 percent chance that it would make a profit of $2 million (an expected gain of $1 million) and a 50 percent chance that it would lose $1 million (an expected loss of half a million). How many of the managers would go for it? Thaler threw in an additional safeguard: The company was large enough to absorb a million-dollar loss. Even when several projects didn’t pan out, none would threaten the company’s solvency.

Of the 23 executives Thaler asked, only three said they would make the investment. The reason for the almost unanimous rejection? If the project was a success, the managers believed they would get a small bonus, but if the project failed, they would probably get fired. They all liked their job, and none of them wanted to risk it. But from the point of view the CEO, each project was worth an expected profit of half a million, so the logical answer was a no-brainer, that the company should invest in all those opportunities to maximize potential gain. Hence, even with projects that don’t require “bet-the-company” resources, the CEO must sometimes absorb career risks that individual midlevel general managers would shun. This is exactly what Jeff Bezos, Jonney Shih, and Steve Jobs have done.

In an increasingly turbulent business environment where companies need to act fast, a purely decentralized innovation model won’t suffice. Corporate leaders must lend their positional power to the rank and file who are equipped with market knowledge. This doesn’t make the case for mindless micromanagement or corporate interference, but there are times when strategic intervention from the very top can prove critical. Jack Welch, former chairman and CEO of General Electric, understands this intuitively: “One of my favorite perks was picking out an issue and doing what I called a ‘deep dive.’ It’s spotting a challenge where you think you can make a difference . . . then throwing the weight of your position behind it. I’ve often done this—just about everywhere in the company.”

Once you become aware of the “deep dive,” it becomes visible everywhere.

THE CEO DEEP DIVE

Situated between Maiden Lane and Water Street near downtown Wall Street is the Community Solutions headquarters—a spinoff Common Ground. Today, Community Solutions is an entirely independent entity. In fact, in 2011, Rosanne Haggerty left her position as the president of Common Ground after promoting Brenda Rosen, the former director of housing operations and programs, to the position.

Rosanne decided she needed to dedicate her full-time efforts to the development of Community Solutions, which focuses entirely on knowledge dissemination without owning and running its own housing unit—a light asset model, many would say. “To let go of Common Ground was not as difficult a choice in the end as for many years that I thought it might be,” Rosanne told me. “We were doing wonderful work [at Common Ground], operating and building housing for those we can help directly. But I was becoming, over many years, more and more concerned about those we are not able to reach through that mechanism, so my excitement at finding an approach to this issue through our innovation work [knowledge dissemination] really made that transition pretty straightforward for me.”

“That entrepreneurial spirit and the corresponding behaviors exhibited at the apex of the enterprise remain, perhaps, the critical functions of the CEO role, which cannot be delegated.”

Still, hearing that from Rosanne was a shock to me. I couldn’t help wonder how many CEOs of corporate America would be willing to throw away their past successes to embrace a future with a clean slate, however promising it might seem. It is human nature to treasure familiarity and cherish legacy. What I’ve learned further was, before Rosanne left Common Ground for good, there had been deep dives all over the organization before Community Solutions was able to stand on its own.

When Becky’s team first speculated the importance of targeting the “chronically homeless,” it was no more than a hypothesis based on close-up observation of a small sample. There was no proof of the strategy’s effectiveness, and no one, except perhaps Rosanne herself, was ready to bet on the idea. Unlike any other routine projects, Rosanne did not delegate the implementation of housing the chronically homeless to a staff member, even the project concerned only eighteen individuals. Compared to the three housing buildings that Common Ground was operating throughout New York City, servicing eighteen individuals could have seemed a meaningless distraction for a CEO.

But instead, Rosanne hastened the process by changing Common Ground’s existing housing policies to cater to those few chronically homeless individuals. She called up Common Ground’s long-standing social services partner, the Center for Urban Community Services (CUCS), to enlist its support in assessing applicants. CUCS was far more experienced at handling psychiatric patients, expertise that was especially useful for the new target group. Rosanne then mandated that Common Ground’s housing operation to become accountable for filling up a certain quota of chronically homeless people.

That directive did not sit well with her staff from the housing operation. “Everyone was skeptical. People were saying things like ‘There goes the building. They are going to be terrible tenants. They are going to mess things up. They are going to be crazy!’” Rosanne said with a chuckle. Managers resisted Rosanne’s idea on the ground that these potential tenants would likely to miss rent payments, or worse, vandalize the housing facilities—a nightmarish scenario as each housing manager was accountable for the annual budget.

To allay fears and minimize resistance, Rosanne agreed to keep track of the impact as new individuals were steadily accepted into the buildings. Additional resources from CUCS were lined up in case of more intensive psychiatric interventions were needed. Rosanne even made a personal commitment to honor any requests for new funding if additional financial needs arose.

After eighteen months, there was almost no negative effect. “Once again, it was one of those situations where what we think might happen did not pan out. We thought these [chronically homeless] people would need many mental health services to help them adjust to the new environment. We were just wrong,” explained Rosanne. “People who have lived on the street for years are instinctively adaptive. They have learned how to survive anywhere. The only extra work was for the office manager who needed to help people set up a bank account and manage their money. This is where they need some help,” she recalled.

This singular focus on eighteen individuals proved that housing the chronically homeless would have a transformative effect on a neighborhood. It was no longer just eighteen fewer individuals on the street; homelessness in Times Square actually began to disappear. This precise understanding of why “housing first” was important laid the foundation for the wildly successful 100,000 Homes Campaign. Targeting the most visible homeless individuals and offering them the housing they wanted made it possible to persuade others who had been on the street for shorter periods to accept help.

BUT WHAT IF?

Now let’s imagine a parallel universe where Rosanne or, earlier, Jonney Shih did not intervene. Rather, the two leaders simply walked away after their initial splashes and left lower-level managers to tackle the implementation realities.

At Common Ground, the manager in charge would undoubtedly have continued to lobby, haggle, and negotiate with those from the housing operation before any of the chronically homeless could be housed—a situation that could easily have derailed the entire initiative.

At Asus, the manager in charge would undoubtedly have gotten dragged into endless debates regarding myriad corporate policies—using a software supplier outside Asia, selling a laptop at an unacceptably low price point, violating corporate assurance by giving the first products away to end users for testing. It would have been months, if not years, before the Eee PC could be launched, thereby missing a critical time window.

As one product manager recounted, “There were times when people outside the Eee PC team saw the project as a failure. . . . We also required a lot of technical expertise from the laptop unit, and the resource demands created a lot of tension. But Jonney personally quieted those ‘noises.’” Again, playing the counterfactual game is fun, and it can be illuminating—and in case you were wondering, the initial Eee PC taskforce had become a new business unit, which expanded and continued to make consumer electronics other than high-end laptops. At that point, Shih felt he no longer needed to supervise its day-to-day operations and named two executives to oversee further business development. He moved on to dive deep elsewhere in the company.

So, there we have it. When the required innovation becomes disruptive, staffing the team appropriately and giving it autonomy are necessary but insufficient. Every time an organization successfully leaps into a new knowledge base, as with Novartis and P&G, the senior leaders need to go beyond formulating a strategy and should get their hands dirty during its implementation. Corporate leaders need to absorb career risks that individual midlevel general managers normally shun. Success requires an exact combination of knowledge power and positional power. That entrepreneurial spirit and the corresponding behaviors exhibited at the apex of the enterprise remain, perhaps, the critical functions of the CEO role, which cannot be delegated. These are the chief functions of a top executive.

Read more: Sonnenfeld: DACA Fallout Damaged Trump’s Relationship with CEOs, but it’s Not Unfixable