Review of “Meltdown: Why Our Systems Fail and What We Can Do About It”

Published in March 2018, Meltdown is a book well-worth reading for executives and board directors seeking to better understand why strategic surprises and threats arise faster than ever in today’s economy — and what they can do to better anticipate and adapt to them.

For us, one of the most attractive aspects of Meltdown is that authors Chris Clearfield and Andras Tilscik explicitly confront both complexity and human nature as critical causes of system failure.

They begin with the observation that, as a wide variety of “systems have become more capable, they have also become more complex and less forgiving, creating an environment where small mistakes can turn into massive failures.”

The authors draw on Charles Perrow’s classic 1984 book “Normal Accidents” to explain why this is the case.

As we emphasize in our work at Britten Coyne Partners, Perrow noted how complex systems have many overlapping cause-effect relationships, which in many cases are characterized by non-linearity and/or time delays. In such systems, apparently small causes can have very disproportionate effects.

Perrow further noted how the presence or absence of “tight-coupling” between various parts of a complex system sets the stage for cascading non-linear effects that quickly outpace human operators’ and managers’ ability to understand what is happening and react appropriately.

As Clearwater and Tilczik observe, system “complexity and coupling create a danger zone, where small mistakes turn into meltdowns.”

One obvious approach is to mitigate this risk by reducing a system’s complexity and loosen the coupling between its component parts (i.e., by adding more slack). Unfortunately, the authors note that, “in recent decades, the world has actually been moving in the opposite direction”, with more systems now operating in the danger zone.

Indeed, as we have often noted, that is one of the key consequences of the dramatic increase in connectivity wrought by the internet, as well as increased automation of many decisions and processes.

After providing painful illustrations of various meltdowns that have occurred in recent years, across a wide range of increasingly complex and tightly coupled systems (from the failures of Knight Trading to Deepwater Horizon), the authors arrive at what for us is the book’s most important conclusion: “As our systems change, so must our ways of managing them.”

Too often, organizations focus on risks that are easy to observe, measure, and manage (e.g., “slips, trips, and falls”), rather than the complex uncertainties that pose the most dangerous threats to their survival.

Clearfield and Tilczik then review a number of steps organizations can take to reduce the risk of meltdowns. These included solutions that are familiar to Britten Coyne Partners’ clients, such as Pre-Mortem analyses and other techniques for surfacing dissenting views, and for being alert to the advance warnings of disaster that complex systems often provide in the form of anomalies, near misses, and other surprises that we ignore at our peril.

A powerful and widely overlooked point made by the authors is that, "in the age of complex systems, diversity is a great risk management tool." As they note, diverse teams are less likely to be plagued by the dangers of excessive conformity: "in diverse groups, we don’t trust each other’s judgment quite as much…Everyone is more skeptical…Diversity makes us work harder and ask tougher questions.”

We share Clearwater and Tilczik's final conclusion that putting the solutions they recommend into practice can be tough. Like them, however, we also know from experience that it is not impossible, and have seen the exceptional benefits that that make the extra effort involved well worthwhile.

blog comments powered by Disqus