A large economy should not fluctuate too much: large economies contain a large number of firms, so the shocks they endure should average out for macroeconomic variables – at least that is how the standard argument goes. However, the United States, for example, has had an average real GDP growth of 3.1% per year since 1947, but with a standard deviation of the same order of magnitude: 2.1%. This illustrates the “excess volatility puzzle”: it is a major fundamental unsolved problem in economics. Another facet of the excess volatility puzzle is that relatively small events cause unexpectedly large bubbles and crashes in prices and supply. A classic example of such an event is the singular tanker incident in the Suez Canal in March 2021, whose macroeconomic effects have been felt for over two years.
A significant amount of effort in economics and in the physics of complex systems has been devoted to understanding the excess volatility puzzle, leading to the creation of agent-based models where firms – constituting the backbone of economic production – buy inputs from each other to create their own output, which they then sell to other firms that need them in turn. However, the timeliness aspect of production systems remains underappreciated, even though timeliness has been ubiquitously and integrally adopted as a quality standard for production systems. The basic idea is simple: for a firm to be able to produce something, it not only needs to be able to order inputs from its supplier, but it also needs to receive these inputs in time. Large delays in supply chains can cause a loss of coordination in production systems that provide the backbone for economic performance, exemplified by the ongoing backlog in the automotive supply chain as a consequence of the collapse of the Baltimore Key Bridge in March 2024.
We have recently developed a novel stylised model that has brought timeliness and operational delays in the provision of goods and services into the picture. Reinforced by competitive pressures, operators often myopically optimize for cost- and time-efficiencies, running the risk of inadvertently pushing production systems towards the proverbial “edge of a cliff” in the sense of timeliness. We have shown that this cliff edge is a true critical point – identified as timeliness criticality – implying that system efficiency and robustness to perturbation are in tension with each other [1]. Specifically for production systems, we suggest that the proximity to timeliness criticality is a measure for their fragility, resulting in large swings in being available on time can indeed cascade through supply chain, and is therefore a possible mechanistic route for unravelling the excess volatility puzzle [2].
[1] J. Moran, M. Romeijnders, P. Le Doussal, F. P. Pijpers, U. Weitzel, D. Panja, J.-P. Bouchaud. arxiv.org/abs/2309.15070 (to appear in Nature Physics).
[2] J. Moran, F. P. Pijpers, U. Weitzel, J.-P. Bouchaud, D. Panja. arxiv.org/abs/2307.03546.
About the speaker
Deb Panja is associate professor of Complex Systems, and the vice-director of Centre for Complex Systems Studies at Utrecht University, Netherlands. He specialises in networks, stochastic dynamical systems, and their applications in ecology, epidemiology, materials, transport, neuroscience and complexity economics.
