Old economic models couldn't predict the recession. Time for new ones
breakthroughs.csmonitor.com
breakthroughs.csmonitor.com
Meh. Plenty of people saw it coming eventually; but most of them had a stake in ignoring it, certain that they could profit handsomely from the subsequent conditions (which they did: bankers received a massive amount of money at rock-bottom prices from state actors, which they wouldn't have got otherwise, and most CxO bonuses were still paid throughout).
The failure was not an academic one, but rather a political one: nobody wanted to reign in financial speculators. We don't need new theoretical models (or rather, we don't need them more than we usually do), we "just" need better politics.
The intellectual bankruptcy on modern neoclassical economics is exposed in some of the battles Steve has had to fight over the years, such as criticism for using debt toGDP ratio [3]. If neoclassical economists cannot even describe how money is created in the modern economy, what meaningful contribution do they have to describing an economy? [4]
[1] http://www.debtdeflation.com/blogs/
[2] https://www.youtube.com/user/ProfSteveKeen/videos
[3] http://www.debtdeflation.com/blogs/2009/03/14/rory-robertson...
[4] http://www.businessspectator.com.au/article/2012/10/22/commo...
Which is not to dismiss the value of "big computing" and agent-based modelling in economics, just that the Great Recession wasn't a failure of experts to see where the course led that can be mitigated by having better models.
In any case, if you want to proceed that agent-based modelling can do better for macroeconomic modelling, than, well, we don't need long advocacy pieces, we need published research. There's sort of an aside in the article that seems to suggest that the authors think that economics research funding is inadequate, which, if it's a real issue the authors are concerned about, is probably a more fruitful subject for a long advocacy piece than "we need this particular kind of model for this domain, even though we can't point to a model of this type that does better in the domain than existing models, just some good models in related-but-distinct domains."
Hey Economists... There's plenty of money > http://cagw.org/reporting/pig-book