Thursday, 14 April 2011

The Viable System Model, Cybersyn and the Financial Crisis

This blog was triggered by a friend’s comments about the value of models and the need to have them for planning purposes. I don’t ignore the value of models, but rather raise the need to connect them to effective action capacity. This point seeks to emphasize the need to embed modelling activities in processes capable of supporting individual and organizational learning thorough the use of models.

The risk I see in the activities of think tanks is that they often are not underpinned by learning mechanisms aligned with the purposes that they manifestly seek to serve. This view emerges from the contrasting experiences I had in the 70s, first in Chile, in the context of Stafford Beer’s inspired CyberSyn Project, and later on at the International Institute for Applied Systems Analysis (IIASA), in the context of my contribution to the Large Organizations Project that included work in the Soviet Union. For almost two years I had the opportunity to experience (albeit at a distance) GOSPLAN’s massive planning processes.

The Chilean experience was an attempt to avoid large planning models in the benefit of models clearly focused on decision groups at the appropriate structural levels, supported by performative indices monitoring the consequences of their use. At least that was the intention. Unfortunately, in addition to the 1973 military coup, which gave no time to underpin structurally these models for effective organizational learning, this project was ahead of its time and lacked the necessary technological infrastructure to be successful. In the Soviet Union planning was supported by very sophisticated models but in practice these were models that reflected the perspectives of scientists at the global level and failed to incorporate the experiences of those affected by them at the intermediate and local structural levels. They had thousands of planners who were planning from the centre the activities of millions whose problems they vaguely understood. This goes without saying that not then, not today, nor in the future they could have the technological infrastructure to make viable this centralising dream.

The complexity of social systems, because of their human fabric, is so large that efforts to manage it with mathematical representations, developed at privileged central positions, are a mirage. In free societies the generation of complexity is distributed and so needs to be its management. This is the complexity management strategy offered by Beer’s Viable System Model; this is the strategy of recursive organizations. Rather than the delusion of large sophisticated representational models to manage social complexity, the strategy is a reflective matching of global, intermediate and local models with organizational response capacity. This matching is a performative strategy that recognises the all too common and often unavoidable errors where there is cognitive (response) capacity to deal with them, and these are errors that reflect socially agreed purposes, which support individual and organizational learning over time.

Ironically, the same mistakes that made planning ineffective in the Soviet Union for decades appear to have played a role in the financial crisis of 2008. These mistakes are evidenced by the behaviour of centres of excellence for economic research and economists before its inception. They supported building complex mathematical models divorced from the complex entanglements of the multiple institutions constituting the global financial system. They failed to see the organizational system, let alone its huge unknowable complexity. They failed to see that this organizational system required the alignment of their tacit global, intermediate and local modelling efforts, on behalf of multiple institutions, to social structures with requisite action capacity. This global organizational system had to be consistent with the visions, purposes and values of political, social and economic policy-makers, beyond the insatiable thirst for profits of financial institutions. This organizational system was necessary to enable structural capacity for distributed learning. This would have given their financial models the capability to support action in the direction of desirable political, social and economic outcomes. This proposition may be seen as idealistic, however we are now hearing from people like Gordon Brown, a past UK prime minister, institutions like the IMF and others the need for global financial regulation, but no one appears to recognise that without, at least, aligned economic projects this regulation will be rudderless and therefore ineffective to avert future financial crises.