Quantcast

Are recessions like traffic?

Published: Monday, July 15 2013 4:58 p.m. MDT

Break lights along I-15 between Lehi and University Parkway in Orem.

Stuart Johnson, Deseret News

Enlarge photo»

I have been in Korea the past few weeks, and part of that time I was attending the Society for Economic Dynamics (SED) meetings at Yonsei University in Seoul. The Society for Economic Dynamics is a relatively new organization. It was founded in the 1990s, and its official journal, The Review of Economic Dynamics, has quickly become one of the leading journals in macroeconomics (along with the Journal of Monetary Economics, and the Journal of Economic Dynamics and Control). The SED meetings are arguably the best forum today for state-of-the-art macroeconomic research.

One of the more intriguing papers I ran across at the Seoul meetings was by Jennifer La'O from the Booth School of Business at the University of Chicago titled, “A Traffic Jam Theory of Recessions.”

La’O started by noting that there is a fairly extensive literature on traffic flow in engineering. The basic models there describe the behavior of drivers as attempting to drive at a target speed under normal conditions. However, when the distance between the driver’s car and the car ahead (the “headroom”) becomes small, drivers will slow down. As long as the traffic density stays sufficiently low, cars will accelerate to the target speed, and headroom becomes irrelevant. This turns out to be a stable result; that is, regardless of the initial bunching of cars on the highway, they will spread out in the long run and drive at the target speed.

As traffic density increases, however, headroom begins to bind, and cars will tend to slow down. As long as the density remains below a key threshold level, however, the long-run equilibrium is one where cars are equally spaced and traffic velocity is constant, though perhaps below the target level.

When density crosses the critical threshold, however, traffic begins to behave quite differently. Cars show stop-and-go behavior even in the long-run steady state. This behavior is characterized by “lumpiness” of traffic. Knots of slow-moving cars emerge even though there is no obvious reason for a bottleneck, such as an accident or lane merge. Cars at the front of the knot are free to accelerate because headroom there is relatively wide. Cars entering the slowdown from the back must brake because headroom is smaller.

La’O’s insight is to apply the basic logic of the traffic flow model metaphor to aggregate economic activity. She thinks of velocity not as vehicle speed, but as flows of money over a period of time. Headroom is equivalent to cash on hand. When cash on hand shrinks, households and firms in the economy slow their spending. Just as drivers of cars try to avoid collisions where there is little headroom, economic agents strive to avoid insolvency, where available cash spending balances go to zero.

Just like the simple traffic flow models, a simple cash flow model exhibits two kinds of steady states. One is where spending is constant across agents and cash flows are also constant. This state corresponds to an economic boom or recovery. The other state is characterized by stop-and-go spending where some agents face cash-on-hand constraints and are stuck in a slow-moving “knot” and others are relatively unconstrained. This state corresponds to a downturn or recession.

One of the challenges facing economists trying to model economic fluctuations is that many times the reason for a downturn is not obvious. Sometimes the cause is traceable to a particular event, such as the subprime mortgage crisis in 2007 or the oil price shocks of the 1970s. But other times there is no obvious cause. Traffic flow models offer an explanation for this phenomenon by positing a coordination problem that may occur under certain circumstances. In the traffic interpretation, it is related to the density of cars per mile on the highway. In the macroeconomic interpretation, it is related to the number of economic agents per unit of available money.

La’O’s model is only a first step, but the parallels so far are very intriguing. As more realism is added to these models they will begin to diverge from the traffic flow models. Perhaps they will turn out to be a good way to describe how recessions occur and are resolved.

Kerk Phillips is an associate professor of economics at Brigham Young University. His views do not necessarily represent those of BYU.

Get The Deseret News Everywhere

Subscribe

Mobile

RSS