by Eamon Javers
The tricky part of TBA is that any one of these indicators alone doesn’t mean much. Maybe the guy just has an itch and needs to scratch. That’s why agents trained in TBA watch carefully. They’re looking for clusters: indicators that come all at once. If the subject shifts his anchor points, grumbles about how long this is going to take, and then takes off his glasses while scratching his nose, he’s probably a liar. BIA’s trainers teach clients how to take shorthand notes as an interview progresses. They tell trainees to note down a number for each question asked in the interview, and then mark a dot next to that number for every indicator they spot during the answer. The advantage of this note-taking system is that you don’t have to look down at your page very often, since you’re only making dots, not writing down in words what the person says. You can keep your eyes off your notebook and on your subject.
And it doesn’t take long to put all this together. Houston tells colleagues that the crucial instant comes within the first five seconds after a question is asked. That’s when a person can’t keep himself or herself from exhibiting the signs of a lie. BIA training, in fact, teaches agents to ignore any events that come more than ten seconds after a question. It’s in the first moment that a liar reveals everything to a trained observer.
Once they spot a lie, agents shift into what they call “elicitation” mode, which is the art of getting someone to confess something even if it is against his own best interest. Mike Floyd, a veteran of the CIA and a colleague of Phil Houston, was known for his extensive preparation before conducting an interview with a suspect. Typically, he’d bring in another expert to help with the interview, so one person could ask the questions and keep the conversation flowing while the other just focused on the suspect, watching for indicators of dishonesty.
They’d ask “bait” questions, such as, “Is there any reason why we might find your fingerprints at the scene of the crime?” They’d phrase the touchiest questions in a neutral way. So they wouldn’t ask, “Did you embezzle the money?” Instead they’d say something like, “Tell me what happened to the money.” They’d ask “presumptive questions” in which they showed the suspect they already knew something about what happened. So they wouldn’t say, “You’re guilty, aren’t you?” They’d ask something like, “Weren’t you worried that the SEC might be able to trace the funds?” That question assumes that both the interviewer and the suspect know the suspect is the perpetrator, and this conversation is simply clearing up some small details.
That kind of questioning is based on the assumption that people want to confess their crimes. Somewhere deep in their souls, they can’t stand holding on to a secret. And if they’re questioned in a careful, nonconfrontational way, they’ll spill everything.
BIA’s training begins with an introduction to Phil Houston’s techniques, followed by a series of videos of executives being interviewed on CNBC. The goal is to learn how to separate the truth tellers from the liars. Then come so-called red-letter drills, in which staff volunteers are organized into a team. One of the volunteers is told to steal a pre-placed red letter from an office. The “thief” and a group of innocent volunteers are brought into the conference room, where the trainees put their new skills into practice. They grill the volunteers to find which one is lying about the theft and is therefore guilty.
Trainees are taught how to position themselves in a room during an interview. Don’t sit across a table from the subject; sit on the same side of the conference table. If you’re sitting at a corner, push your chair back from the table until you can spot movements of the subject’s legs.
One person familiar with the sessions says many of the students can spot the liar after just a day or two of training. But often, it is the students in the back of the room who identify the liar first, not the student designated to ask the questions. For a beginner in these techniques, it can be tough to both quiz a suspect and watch for clusters of indicators.
At many of their stops, BIA trainers distribute laminated cards with deception-detection tips for trainees to remember after the session. The cards, headed “Tactical Behavior Assessment and Strategic Interviewing Pocket Guide,” urge students to “Be in L2 Mode—Look and Listen.” Over time, BIA’s clients have included several hedge funds and investment banks.
PHIL HOUSTON WAS convinced that there was nothing anyone could do to defeat his system. That is, even if an agent was trained in the TBA method, and knew what another trained observer would be looking for, he still couldn’t lie without getting caught. Houston and his colleagues from the CIA had tapped into something deep within the human psyche, and they knew it was powerful stuff. People trained in TBA often make terrible liars themselves. Once you’re convinced that lying is always transparent, it becomes almost impossible for you to carry off a successful lie.6
Phil Houston and his colleagues thought that selling this timeless TBA training to law enforcement agencies across the country might be a good business. At first, it was. After 9/11, local police forces realized they needed to beef up their interrogation techniques on the front lines against terrorism. Police officers signed up for the training, and Houston and his colleagues traveled the country evangelizing about TBA. But although Houston is a brilliant interrogator, his colleagues say he’s not a brilliant businessman. His company kept bleeding money. Police forces were late paying their bills. Houston and his colleagues were more interested in delivering the training than in following up on invoices.
Soon, the struggling little firm faced an impasse. It needed a buyer to inject business discipline and make it thrive. That’s when Houston met Liam Donohue, a venture capitalist based in Boston who was then running an investment fund called Arcadia Partners. TBA and the CIA’s “elicitation” techniques captivated Donohue, and he saw a wide-open new market for them: corporate America.
Arcadia bought a large percentage of Houston’s firm, and the CIA veterans retained a significant equity share in the reconstituted new company, which was now called Business Intelligence Advisors, or BIA. The acronym was a deliberate reference to the CIA.
Officers of BIA declined to be interviewed during the writing of the book. But the firm describes itself, succinctly, in a document distributed to clients and signed by its current president, Cheryl Cook:
Business Intelligence Advisors Inc. (BIA) adapts techniques developed in the context of international intelligence gathering and national security to enhance high-value, high-risk decision making in the private sector. BIA’s services are targeted to, and most valuable in, information driven processes, where the reliability of information can make a crucial difference in outcomes.
BIA’s employees and network of experts include the world’s leading intelligence resources, with unparalleled expertise in strategic interviewing, intelligence gathering, risk assessment and security. BIA combines these resources with experienced analysts, consultants and project managers to ensure that resources are deployed to maximize return and business impact.7
Liam Donohue eventually became chairman of the board of the new company. Under his leadership it began to thrive. As an MBA from Dartmouth’s famous Amos Tuck School of Business Administration, he brought in a series of managers who shared his background. He reoriented BIA from an enterprise primarily focused on law enforcement to a business focused on generating corporate clients. He hired a CEO, Don Carlson, who was a lawyer and had been with the investment banking firm Goldman Sachs. By 2005, Carlson says, BIA was generating between $10 million and $11 million in annual revenue.
Donohue is extremely secretive about BIA, and the firm has opened up to the press about its services only once, for an astonishing article in Barron’s by Jonathan R. Laing, “Is Your CEO Lying?”8 The article laid out BIA’s deception-detection techniques and set off a flurry of interest in CIA-style interrogation within the hedge fund community.
The CIA didn’t invent these techniques. Although it spent years working on tactics based on the best information modern science has to offer, at most i
t was just reinventing something that had been developed more than 100 years ago by the first private eye, Allan Pinkerton. All the biographies of Pinkerton mention his ability to get criminals to confess. In the 1850s Pinkerton wrote that there was a reason for this astonishing success: “Criminals must eventually reveal their secrets,” he noted. “And a detective must have the necessary experience and judgment of human nature to know the criminal in his weakest moment and force from him, through sympathy and confidence, the secret which devours him.”
Allan Pinkerton and Phil Houston, separated by more than a century, came to remarkably similar conclusions about the human condition.
Pinkerton said that his technique was based on judgment of human nature, and that is also true of the TBA technique. Pinkerton noted that detectives should use “sympathy,” which is what “elicitors” trained in TBA do as well. And Pinkerton talked about using “confidence,” just as the CIA’s elicitation method uses presumptive questions, in which the interrogator projects confidence that he already knows something crucial. Finally, Houston and Pinkerton agree on the most important point: the secret devours the criminal. Criminals all want to confess. The interrogator is just helping them along.
Pinkerton probably didn’t invent deception detection or elicitation, either. These may be among the things that smart, observant people invent for themselves. What’s remarkable, actually, is that every generation produces people who think that the best way to get information out of someone is to beat it out. In the face of so much evidence that deception detection and elicitation work, there are still interrogators who don’t want to be polite to a suspect. And usually, those are the interrogators who can’t get a confession.
IN THE SUMMER of 2005, BIA was on a roll. On July 14 Phil Houston and Patsy Boycan led a BIA team listening in on a Southwest Airlines earnings call with investors. Southwest was coming off a boffo quarter in which it beat Wall Street’s all-important expectations with an earnings spike of 42.9 percent over the same quarter the year before. BIA’s interest in Southwest was prompted by a client, Ziff Brothers Investments, a private equity fund controlled by the three billionaire sons of William Ziff, Jr., who built the Ziff-Davis publishing empire. Ziff Brothers wanted to know whether or not Southwest could continue its excellent run into the next quarter. Did Southwest’s executives believe their own rosy forecasts?
After a preamble, Southwest’s CEO, Gary Kelly, and its chief financial officer, Laura Wright, began to take questions from analysts. The back-and-forth was cordial and warm. Many of the analysts had been covering the company for years, and had spoken with the management dozens of times. Some analysts interspersed their questions with a few words of congratulations for the great quarterly results. But BIA’s spies on the call weren’t into schmoozing. They didn’t say a thing. They were in “L squared mode.”
BIA’s report on that call is not available, so it’s difficult to say exactly what conclusions the analysts reached. But a look at the transcript of the conversation of July 14 reveals several moments that may have stood out for the BIA team.9 First, J.P. Morgan’s Securities analyst Jamie Baker asked a question about potential fare increases—which could be deadly for a company like Southwest that makes its name as a discount airline.
“I know AMR [the company that owns American Airlines] put a $2 to $3 one-way increase into the majority of your markets yesterday, but would you characterize Southwest as still in the study period or have you definitively chosen not to match that?” Baker asked.
“Well, we haven’t changed our fares,” responded Kelly. “We want to be a leader, but we want to be the low fare leader.”
“Mmm hmm.”
“This is actually a perfect environment for us where all of our competitors are raising fares and it really helps us differentiate who we are,” continued Kelly.
On the basis of his noncommittal response, “Mmm hmm,” we can guess that Baker might have been skeptical about this point, too, but a trained TBA analyst might home in on specific phrases. Kelly responds to a direct question about raising fares in the present tense with information that’s not an answer at all: “Well, we haven’t changed our fares.” He doesn’t say what the company plans to do in the future, which is what Baker asked. BIA would call that a nonanswer.*
Kelly follows his nonanswer with what BIA calls a protest statement: He says, “We want to be the low-fare leader.” Again, he doesn’t answer the question, which was whether Southwest is still considering a fare increase. He describes what Southwest wants to be, not what it is going to do. Were the BIA team’s pens scribbling furiously during this exchange?
Then, as a closer, Kelly adds what BIA would call a qualifying answer: “This is actually a perfect environment for us.” By using the word “actually,” was Kelly trying to persuade the listener of something that on its face wouldn’t make any sense?
Was Kelly trying to duck the question about whether Southwest would have to raise its fares? Soon enough, it did just that: at the end of 2005, Southwest bragged in its annual report to shareholders that the company had raised fares only “modestly” throughout that year. The report noted that the average passenger fare increased from $88.57 in 2004 to $93.68 in 2005.10
Later in the earnings call, the trained BIA observers listening in might have concluded that the executives were uncomfortable with their own earnings predictions. Lehman Brothers’ analyst Gary Chase tried to pin Kelly down, asking, “Can you just sort of help us think through why you think 15 percent is an achievable growth goal for next year?”
“Well, first of all, it’s a goal,” responded Kelly. He added several caveats about what it would take for the company to hit the number: the economy would need to remain healthy, competitors would need to remain predictable, and maybe some improvements in the Baltimore market would be needed.
Still, he said, “We are not conceding that we cannot improve our earnings next year by 15 percent. And we just want to make that very clear because there are already reports out there that are suggesting that our earnings are going to decline, and that is not acceptable to us.”
But then he concludes on this tepid note: “At this point, it certainly looks to us like a reasonable goal.” Notice Kelly’s use of several qualifying phrases in a row: “at this point,” “certainly looks to us,” “reasonable goal.” That’s far from a full-throated endorsement of the company’s own revenue projections.
Certain statements in the Southwest earnings call contrast with the positive buzz the company received that day. Even though the company has an excellent reputation and even then was receiving kudos for an innovative hedging strategy to blunt the impact of rising fuel costs, something was giving the executives discomfort about their future.
Sharp traders at Ziff Brothers could have used that information in conjunction with everything else they knew about Southwest to form a picture of a company unable to keep up its earnings streak. In fact, an executive familiar with BIA says Ziff concluded from the call that Southwest’s executives weren’t confident they’d be able to repeat their earnings success in the next quarter. The source says Ziff shorted Southwest’s stock, and in doing so, may have made a good deal of money. The positive glow from the healthy earnings report lasted only until July 18, when the stock peaked at around $14.75 per share before settling into a slump that would last for more than a month. By the end of August, the stock was trading at about $13.50 per share.
DESPITE THEIR SUCCESSES, not every client was as impressed by the BIA team’s presentation or its tactics. That same summer of 2005, BIA’s Don Carlson, who had been an attorney with Goldman Sachs, brought Phil Houston and Mike Floyd to meet with Goldman’s own internal business intelligence division. The BIA team hoped to land Goldman Sachs as a client. For BIA, working with Goldman Sachs would be a gold mine: it had thousands of employees, so a contract for BIA’s interrogation training alone could be hugely lucrative. In 2005, Goldman had more than $20 billion in revenue. The firm would have nearly unlimited bu
dgets for research and investigations. Conceivably, this one client could double BIA’s revenue as soon as it signed a contract.
A lot was on the line when Carlson, Houston, and Floyd strolled into the conference room at Goldman’s headquarters at 85 Broad Street in lower Manhattan. They were used to being greeted deferentially. Even brash investment bankers are a bit intimidated by certified spies, particularly the CIA’s legendary interrogators. But this time the BIA team met a harsh reception. Goldman’s fourteen or so employees in the room asked question after question as the BIA team struggled to plow through its standard proposal. One Goldman employee at the meeting was Jeffrey Starr, who was then on loan to Goldman’s business intelligence team from an intelligence agency at the Department of Defense. As an expert on intelligence tactics, he began to rip up the BIA argument piece by piece.
Starr pushed his chair back from the table and began to grill BIA’s presenters. Are you saying that these indicators prove someone’s a liar? Why wouldn’t actions outside the five-second window be relevant? What if somebody’s just combing his hair? How can you put your faith in this kind of stuff? It was the roughest treatment Houston and Floyd had received. When they showed a short video to illustrate their point, Starr laid into them.
In the video, an investigator quizzes five people about a stolen laptop. The audience members at the training session are supposed to look for the indicators of deception and figure out who stole the laptop. Starr pounced on a flaw. The video featured actors, not real people. If the actors had been coached on what to do, they wouldn’t be giving the indicators that real human beings would give—they’d be giving the indicators BIA told them to give. Starr was incensed. What good was this video? Why didn’t they bring in a video of a real interrogation? BIA’s team couldn’t come up with a good answer for him. They felt as though they’d failed.
