Warren Buffett, MS ’51, once famously called derivatives “financial weapons of mass destruction,” in part because he regarded the debt contracts’ often-convoluted transactions and lack of collateral requirements as rendering risk assessment nearly impossible. Extend the weapons metaphor to the underlying financial models on which derivatives are based, Professor Daniel Beunza says, “and, in light of the financial crisis, models can be viewed as the nuclear physics that gave rise to the atomic bomb — powerful, bewildering and potentially lethal.”

The last few months have seen a wide-ranging controversy over financial models. At one extreme, critics have attacked models for their lack of transparency and limited accountability. Models, they charge, are black boxes that even expert users fail to understand and that become dangerously inaccurate when the world changes. And whenever a model fails, it is all too easy for traders to deflect blame with the perfect storm excuse.

At the other extreme, academics in finance and Wall Street practitioners dismiss the backlash, and point instead to bad incentives. “The argument is, ‘Models don’t kill banks — bankers kill banks,’” Beunza explains. “To proponents, the crisis underscores the need for more and bigger models.”

Critics and proponents both miss the mark, says Beunza, who believes that what the economy needs is neither more nor less models, but better models. “To see this, one needs to go beyond ideology and open the black box of financial modeling. What is a financial model, ultimately? Whether an equation, an algorithm or a fancy spreadsheet, a financial model is no more than a perspective, a point of view about the value of a security.”

As such, models also present a paradox. They are powerful instruments that can reveal hidden opportunities, yielding extraordinary profits. But when models are misused, they can inflict destructive losses on a bank. “Because stock prices are the outcome of human decisions, financial models do not always work like the iron law of Newtonian gravity,” Beunza points out. “And because a model entails a complex perspective on issues that are typically fuzzy and ambiguous, it can skew traders’ perspectives, leading, as we’ve seen, to billions in losses.”

Given the two-edged nature of financial models, Beunza and his colleague David Stark of Columbia University’s Department of Sociology posed the question, Can banks reap the benefits of models while avoiding their dangers?

For three years, Beunza and Stark conducted a sociological study of a derivatives trading room at a large bank on Wall Street. The bank reaped extraordinary profits from its models but emerged unscathed from the credit crisis.

The key, they found, lay not with the models but with the bank’s culture, organizational design and leadership. The bank introduced reflexivity in every aspect of its organization: from junior traders on up to executives, everyone was ready to question their own assumptions, listen for dissonant cues and respect diverse opinions.

The bank accomplished this in part by hiring people with a healthy dose of humility and an appreciation for the limits of their smarts. Beunza notes, “This often meant older traders rather than younger hotshots.”

But the key to the bank’s reflexivity did not lay just with individuals. “Reflexivity doesn’t mean superintelligent traders engaged in some heroic mental feat — splitting and twisting their minds back on themselves like some intellectual version of a contortionist,” he says. “Rather, reflexivity is a property of organizations.”

The architecture of the bank, for instance, was crucial. The open-plan trading room grouped traders with different strategies in the same shared space. Each desk focused on a single model, developing a specialized expertise in certain aspects of the stocks.

“Think of a stock as a pie. Investors often eat the pie whole, with predictably dire consequences. The professionals that we saw, by contrast, sliced stocks into different properties.” Each desk was in charge of a different property, and the different desks then shared their insights with one another. This could happen in a one-minute chat between senior traders across desks or in an overheard conversation from the desk nearby. “This communication,” Beunza says, “allowed traders to understand those aspects of the stock that lay outside their own models — the unexpected ‘black swans’ that can derail a trade.”

To make communication easier, the bank fostered a culture that prized collaboration. For instance, it used objective bonuses rather than subjective ones to ensure that envy did not poison teamwork. It also moved teams around the room to build the trust that physical proximity engenders.

Most important, Beunza and Stark reported, the leadership of the trading room had the courage to punish uncooperative behavior. “The manager of the room made it abundantly clear that he would not tolerate the view, prevalent among some, that ‘if you’re great at Excel, it’s OK to be an a--hole,’” Beunza says. “He conveyed the message with decisive clarity by firing antisocial traders on the spot, including some top producers.” In other words, the culture at the bank was nothing like the consecration of greed that outsiders often attribute to Wall Street.

The bank went so far as to use its own models to be reflexive about modeling. The traders translated stock prices into the estimates that their competitors put in their models, effectively reverse engineering their competitors’ models from the freely available prices of the securities. This information often planted healthy doubts about the traders’ own estimates, prompting them to start over when necessary. The researchers termed this organized dissonance.

Perhaps not surprisingly, Beunza and Stark came to view the “more models or fewer models?” choice as a misguided one. The researchers developed a set of managerial procedures, which they call reflexive modeling, that they say can allow banks to reap the benefits of models while avoiding their accompanying dangers.

Ultimately, the researchers’ study suggests that a lack of reflexivity may be behind the current credit crisis. Beunza and Stark liken the reflexive stance to infantry officers of old who instructed their drummers to disrupt the cadence while crossing bridges. “The disruption prevents the uniformity of marching feet from producing resonance that might bring down the bridge,” Beunza says. “As we see it, the troubles of today’s banks may well be a consequence of resonant structures that banished doubt, thereby engendering disaster.”

Daniel Beunza is assistant professor of management at Columbia Business School.