Critical slowing down associated with regime shifts in the US housing market

Abstract

Complex systems are described by a large number of variables with strong and nonlinear interactions. Such systems frequently undergo regime shifts. Combining insights from bifurcation theory in nonlinear dynamics and the theory of critical transitions in statistical physics, we know that critical slowing down and critical fluctuations occur close to such regime shifts. In this paper, we show how universal precursors expected from such critical transitions can be used to forecast regime shifts in the US housing market. In the housing permit, volume of homes sold and percentage of homes sold for gain data, we detected strong early warning signals associated with a sequence of coupled regime shifts, starting from a Subprime Mortgage Loans transition in 2003–2004 and ending with the Subprime Crisis in 2007–2008. Weaker signals of critical slowing down were also detected in the US housing market data during the 1997–1998 Asian Financial Crisis and the 2000–2001 Technology Bubble Crisis. Backed by various macroeconomic data, we propose a scenario whereby hot money flowing back into the US during the Asian Financial Crisis fueled the Technology Bubble. When the Technology Bubble collapsed in 2000–2001, the hot money then flowed into the US housing market, triggering the Subprime Mortgage Loans transition in 2003–2004 and an ensuing sequence of transitions. We showed how this sequence of couple transitions unfolded in space and in time over the whole of US.