The focus of Haldane’s speech is on this second point. He talks about the need for macroeconomics to learn from other disciplines in both the natural and social sciences in order to seek “a different perspective on individual behavior and systemwide dynamics.” He argues that the profession has “borrowed too little from other disciplines” and become “a methodological monoculture,” with the associated risk that everybody in the field can be wrong in the same way and at the same time. He finds compelling evidence in a survey of American professors of social science, who were asked whether “interdisciplinary” knowledge was better than knowledge “obtained by a single discipline.” Most social scientists sensibly thought that the answer was yes, by overwhelming margins. But 57 percent of economists disagreed or strongly disagreed. Economists literally think they have nothing to learn from anyone else. The field also suffers from its rigid hierarchies and its lack of gender and racial diversity. As Haldane puts it, “It is difficult to escape the conclusion that economics remains an insular, self-referential discipline.”

Haldane’s proposed solution to the problem lies in an emerging new field called agent-based modeling, in which large amounts of computer power are used to build models where individual actors have their own motives and their own agency. These agents collide and overlap and interact and generate predictions through a messy process more akin to real life than the clean models of old-fashioned macroeconomics. Agent-based modeling has been shown to have promise in other fields, especially physics — and also in modeling real-world problems like the tendency for Brazil nuts to levitate toward the top of the mixed-nut package. (That might sound kooky, but Haldane points out that the nut research “has since found real-world applications in industries such as pharmaceuticals and manufacturing.”)

Haldane isn’t the only prominent nonacademic economist to think there are profound problems in macroeconomics. Paul Romer is a highly regarded macroeconomist who recently became chief economist at the World Bank. (It was Romer who is credited with having coined, in 2004, the famous slogan that “a crisis is a terrible thing to waste.”) Last winter he, too, issued a lacerating critique of his field, titled “The Trouble With Macroeconomics.” His argument focuses on the question of “identification,” which is shorthand in the field for how economists identify the cause of an event. Identification is a huge deal for economics, because if you can’t tell if x, y or z caused something, you don’t know which of those variables to study or measure or change. The omission of finance and the role of money from economic models is the biggest weakness in the field, Romer believes. It is hard for noneconomists to get our heads around the fact that the dominant macroeconomic models (including the ones used by central banks) made no allowance for how money works — especially not for fluctuations in how much of it is around. Such an oversight seems weird, and entirely counterintuitive. But that’s what happened, and it played a big role in macroeconomists’ ability to, as Romer puts it, “dismiss mere facts.”

The Bank of England and the World Bank are two of the most important noncommercial banks in the world. It is striking, and exhilarating and scary too, that the chief economists of these two institutions each thinks that their trade is in crisis. Not so long ago, leading macroeconomists thought that the main part of their work was basically over. The Nobel winner Robert Lucas claimed as much in his presidential address to the American Economic Association in 2003. Macroeconomics, he said, “has succeeded: Its central problem of depression prevention has been solved.” That now looks wildly hubristic. Maybe the field is roughly where physics was at the end of the 19th century, when some scientists thought that their own field’s main problems had been solved, within a Newtonian framework that is now known to be inadequate to describe reality. Maybe, like physics in the 20th century, macroeconomics is about to have a heroic period. If it does, it will begin in the work of people like Romer and Haldane, who are unflinchingly willing to talk about its huge recent failures. A crisis is a terrible thing to waste.