Leads on Predicting Institutional Collapse

James K. Hazy, July 3, 2012

A recent study uses catastrophe models, an approach that has been proven across multiple disciplines from physics to ecology, to describe tipping-point events in dynamical systems. These are points, or values of a parameter--describing environmental conditions, the "container," wherein the system functions--that once crossed, can lead to sudden collapse in population density. If similar models can be applied to human organizing, then lessons from these systems may also apply to potential catastrophic collapse of social, political, and economic systems.

The study (Dai et al., 2012) was conducted using yeast cultures, and therefore is not directly applicable in the human case. But it does provide possible leads for human systems researchers. For example, as is described below, the study identifies two leading indicators that can signalthe approach of a “tipping point” toward instability. If analogous leading indicators can be discovered within human activities, and the relevant parameters can be identified and measured, then there is potential that early warning signals could be defined with regards disruptions to political and economic institutions. Such analysis would add more robust analytical support behind naysayers’ warning cries during times of crisis. Had such signals been known to exist in 2006 and 2007, it is possible that the financial crisis could have been averted, or at least, its impact might have been mitigated through policy interventions.

Possible Leading Indicators of Institutional Collapse

The first indicator identified in the study is an observed lessening in resilience when the system is shocked by an external event. This implies that the probability declines that the system will recover to its prior stable state. The second seemingly related indicator is called critical slowing down. This means that the system begins to take longer to dampen fluctuations or perturbations that impact it, dynamics that can observed as increasing observable variance in populations. Each of these is further described below.

Loss of resilience

The first indicator of an impending catastrophe identified in the Dai et al. (2012) study was loss of resilience when the population under study was subjected to an external shock. In the study, the yeast culture was subjected to a one-time salt bath as an external shock. (This would be the metaphoricalequivalent to the affects of a natural disaster in human systems.) When this occurs, a lower value for the parameter that described the systems dynamics translated into a greater probability that the yeast culture would survive after the shock. A higher value for the parameter implied a greater change the system would collapse.

Thus, when a stable condition, like the financial markets, begins to take longer to recover from shocks, one might expect that a tipping point toward catastrophic collapse might be approaching as the value of the identified parameter approaches the tipping point. The presence of this condition would signal a reason for caution.

A critical slowing down

The second indicator involves what is called a critical slowing down in the dampening of "fluctuations,"experiments that arise within the system’s functioning. When a variant process arises to challenge normal activity, it tends to remain organized for a longer period during critical slowing down, and thus the orbits of many observable fluctuations remain predictable longer. 

As Dai et al. point out, a critical slowing down can be identified as an increase in variance across samples in the distribution of outcomes as well as through an increase in autocorrelations within the samples. This would indicate that organized deviations—or “experiments,” as described by Goldstein, Hazy and Lichtenstein (2012)—are remaining as active alternatives for a longer period within the system.

Both of these indicators reflect the fact that informational differences with respect to observed events are being retained as “options” within the system (at the fine-grained level), and that this information has not yet been incorporated into the system as (Coarse-grained) structural “innovations,” ones that would make the organization's behavior more predictable as a system.

Implications for research in HID

The dynamics described here as they apply to a yeast culture, although simple, might offer a useful metaphor for certain aspects of human interaction dynamics.  The same fold catastrophe (or the cusp with two parameters) may be a useful model for describing how externalities impact the stability of relevant aspects of social system functioning.

The read more click here.


Dai, L., Vorselen, D., Korolev, K. S. and Gore, J.(2012). Generic Indicators of Loss of Resilience Before a Tipping Point Leading to Population Collapse.  Science, 336, 1175-1177.

Goldstein, J. Hazy, J. K., Lichtenstein, B. (2010). Complexity and the Nexus of Leadership: Leveraging Nonlinear Science to Create Ecologies of Innovation.  Palgrave-Macmillan, Englewood Cliffs, NJ.