The more senior a leader becomes, the more carefully other people nearby filter themselves. Directors soften criticism, executives dress up bad numbers, and candidates rehearse their way past hard questions. Jamie Dimon, the chairman and CEO of JPMorganChase, has spent his career figuring out how to surface realities regardless and how to encourage his employees to do the same. 

As part of Columbia Business School’s Distinguished Speaker Series, Adjunct Senior Research Scholar Pedro Moreira Salles moderated a discussion with Dimon that covered corporate governance, retail competition, AI, and credit risk. Bad actors turn up in every company and every system, Dimon noted, and it falls to leaders to set the tone for candor.

He cultivates an honest, effective culture by practicing these three habits.

1. Vacate the room

At JPMorganChase, Dimon leaves every board meeting toward the end so the lead director can finish up without him. He has departed early from every meeting since his first year at Bank One, and he calls it the single most important governance practice he knows.

Once he has left, those who remain can feel free to be more candid: “They don’t have to be worried about hurting my feelings,” he said.

When Dimon arrived at Bank One from Citigroup in 2000, he told that first Bank One board what to expect from him: “I’m not going to wine and dine you. I have a job to do.” They would get the truth, he said, the good, the bad, and the ugly. 

Once, quarterly earnings came in far below expectations, and a director asked about a significant writedown on used car leases. The answer the board got from the then-CFO was technically accurate, but told him nothing, and Dimon felt that the CFO was only dressing up numbers for the board.

“People lie about numbers,” Dimon said.

At the next board meeting, he walked the directors through the math himself: The company was carrying billions of dollars in used cars on its balance sheet at values the market would not support, and the writedown was only what had already been sold. He told the board the next quarter would bring a massive loss, which it did.

Doing so helped establish Dimon as a truth-teller who could be trusted. He had promised the board the good, the bad, and the ugly, and this was it. 

He built his later working relationships according to the same principles. Senior executives present their own businesses to the board directly, and there is no vetting of presentations beforehand, so that executives can say what is on their minds. 

2. Do your homework before a job interview

The errors that have cost him most, Dimon said, have involved hiring people. “It’s the absolute hardest thing you ever do,” he said. “If you’re wrong, it hurts the company.”

So he gathers a lot of intel before the first conversation with a new potential employee. “I could know all about you and never talk to you,” he said. He wants to hear from a candidate’s friends, bosses, peers, and subordinates. Does the candidate show up on time? Do people trust them? Do they work hard?

That picture, Dimon said, can tell him more than his own read of someone sitting across a table. It does not make him right every time, but it drops his error rate dramatically. He is also careful not to oversell a job to someone he wants to hire.

Good leadership, Dimon said, includes having hiring managers tell the truth, treat people right, and have a purpose.

“There are bad people who don’t care what you think and they get away with cutting corners,” he said. “It happens in every business, in every society. Setting the standard, and holding to it, falls to whoever is in charge.”

3. Don’t make big decisions on Friday

Two important leadership rules Dimon follows helps him avoid regrets: “Don’t make big decisions on Friday when you’re tired,” and “Don’t make decisions when you’re angry.”

Past that, his method is iterative. He works a problem over and over until an answer emerges, ensuring there is a process and the right people in the room.

Dimon calls himself a fanatic about reporting, learning, and simply raising your hand. At Bank One, he asked each executive what they needed to run their business, then asked them to send him reports. The reports didn’t come, so he asked again and set a deadline. When the reports arrived, they were poor, and the executives running those businesses could not identify their marginal costs or explain how anything was allocated. 

The rules, the process, the reports: All of it exists so a decision can rest on something solid, he said.

None of it makes mistakes disappear, however. “You’ll minimize errors,” Dimon said. “You still make them, but it’ll be much smaller.”