Janet Yellen have a good speech last Friday at a Boston Fed conference on Macro Research after the crisis.
She almost got it
Many macroeconomists work with models where groups of individual actors, such as households or firms, are treated as a single “representative” agent whose behavior stands in for that of the group as a whole. …
Prior to the financial crisis, these so-called representative-agent models were the dominant paradigm for analyzing many macroeconomic questions. However, a disaggregated approach seems needed to understand some key aspects of the Great Recession. To give one example, consider the effects of negative housing equity on consumption. Although households typically reduce their spending in response to wealth declines, the many households whose equity positions in their homes were actually driven negative by the reduction in house prices may have curtailed their spending even more sharply because of a markedly reduced ability to borrow. …
In light of the housing bubble and subsequent events, policymakers clearly need to better understand what kinds of developments contribute to financial crises. What is the relationship between the buildup of excessive leverage and the value of real estate and other types of collateral, and what factors impede or facilitate the deleveraging process that follows? Does the economic fallout from a financial crisis depend on the particulars of the crisis, such as whether it involves widespread damage to household balance sheets?
I underlined the last two words myself because all of the above matter because balance sheets matter.
In DGSE representative agent based models balance sheets dont matter because with a single agent inflows and outflow balance, there are no balance sheets, there is no distinction between stocks and flows mathematically. Hence it is impossible for the model to behave in a non linear manner with feedback loops and complexity, the model can be linearised and is hence tractable within the limited simultaneous equation linear dynamics framework of DGSE.
So whilst Janet gets empirically what is wrong with modern macro research she misses the fundamental reasons this matters.
- Economics must be rebuilt around balance sheets and fundamental accounting identities
- Hence state – balance sheets- must be the basis of all models. If the model is not a state machine it cannot describe the state of anything economic.
- Hence credit and debt, and money matters
- As debt issues are fundamentally non linear and complex models which require linearisation, such as DGSE, must be discarded, they are too broken to be fixed.