After the global financial crisis, there is greater awareness of the need to understand the interactions between the financial sector and the real economy and hence the potential for financial instability. Data from the financial flow of funds, previously relatively neglected, are now seen as crucial to the data monitoring carried out by central banks. This paper revisits earlier efforts to understand financial-real linkages, such those of Tobin and the Yale School, and proposes a modeling framework for analysing the household flow of funds jointly with consumption. The consumption function incorporates household income, portfolios of assets and debt held at the end of the previous period, credit availability, and asset prices and interest rates. In a general equilibrium setting, these all have to be endogenised and since households make consumption and housing purchase decisions jointly with portfolio decisions, there is much to be gained in modeling a household sub-system of equations. Major evolutionary structural change -namely the evolving credit architecture facing households -is handled by our 'Latent Interactive Variable Equation System'. A by-product is improved understanding of the secular decline in US saving rate, as well as of the household financial accelerator. Moreover, the models discussed in this paper offer new ways of interpreting data on credit, money and asset prices, which are crucial for central banks.JEL Codes: B22, E21, E44, E51, G11.Key Words: Finance and the real economy, financial crisis, consumption, credit constraints, household portfolios.
Non-technical summaryAfter the global financial crisis, there is general awareness, see the report on the first two years of the macro-prudential research network, ECB (2012), of how standard macroeconomics as practiced up to 2008 had failed to understand the interactions between the financial sector and the real economy and so failed to grasp the potential for financial instability. Data from the financial flow of funds, neglected previously, are now seen as crucial to the data monitoring carried out by central banks. This paper revisits earlier efforts to understand financial-real linkages, such as those of the Yale tradition, as the back-ground to new approaches to model linkages between the flow of funds and the real economy. These new approaches throw fresh light on how to interpret data on money and credit. Tobin and Brainard (1963) had anticipated the bank lending channel of monetary transmission as later highlighted by Bernanke and Blinder (1988) and Bernanke and Gertler (1995). Brainard and Tobin's 1968 stylised paper on pitfalls in financial modeling included three sectors (governments, private sector and banks), a set of seven financial assets and focused on investment (rather than consumption) as the key interaction between the financial sector and the real economy. In the 1970s, the Yale school introduced consumption with a systems approach to household flow-of-funds analysis, see Backus and Purvis (1980). However, in the ...