How Extremes Become More Extreme, Triggering Collapse
The signs of an increasing exodus from big cities like NYC may just be the tip of the iceberg…
“The extremes are not visible to the vast majority of participants, and so they are exposed to high levels of risk they don’t see or understand.”
How Extremes Become More Extreme, Triggering Collapse
The question “Is the weather becoming more extreme?” opens up endless debates because our perceptions may differ from actual measurements since we’re prone to recency bias, where what happened recently looms much larger than events of a decade or century ago.
In the realm of economics and markets, our perceptions of extremes are backed up with data: based on the ratio of stock valuations to GDP and corporate sales (not profits, because profits are easily gamed) to GDP, the stock market has never been as over-valued as it is today.
The rally in global stocks off the March lows is the steepest such rally ever. The unemployment rate is equally extreme, as is the Federal Reserve’s money-printing: $3 trillion has been created out of thin air since February 26 as the Fed’s balance sheet rose from $4 trillion to $7 trillion.
Financial/market extremes are becoming more extreme.
The disruptive social and political consequences of systemic unfairness and extreme wealth inequality are still unfolding, as are the global consequences of the Covid-19 pandemic.
Setting aside the specifics, can we discern systemic dynamics that could make extremes become more extreme?
Feedback loops are one such dynamic. Somewhat counter-intuitively, when feedback arises to moderate the intensity of a trend, that’s negative feedback. When feedback intensifies the trend, it’s positive feedback.
Why is this counter-intuitive? If a bad trend is moderated by negative feedback, that’s good (positive). If a bad trend gathers momentum due to positive feedback, that’s bad (negative).
When an insect population explodes higher due to ideal conditions, birds and other predators feast on the over-supply, reducing the infestation. This negative feedback moderates the damage inflicted by the infestation.
If a rapidly expanding insect horde has few predators and its range and mobility increase with every generation, allowing it to find new food sources, this positive feedback enables a vast expansion in each generation–exactly what’s we’re witnessing with locusts.
Positive feedback leads to runaway systems, i.e. run to failure where the system accelerates until it collapses.
If the system is isolated, then the damage is contained. But if the system is interconnected with others, then its failure could trigger the collapse of other systems, either as a direct (first-order) effect or as an indirect (second-order) effect.
In other words, in highly inter-connected systems, one failure can trigger a domino effect that can become non-linear once second-order effects manifest.
For example, consider the direct effects of the pandemic on small Main Street businesses. Surveys have found that around 40% of small business owners are planning to close permanently. The reasons were not surveyed, but the obvious reason is the owners don’t see a 100% return of their revenues as likely, and so it’s prudent to staunch the losses by closing now rather than risk catastrophic losses by re-opening.
The first-order effect of urban disorder is the destruction of some small businesses. This may push indecisive owners into closing for good, or considering moving to a safer locale outside the city.
The second-order effect is the re-assessment of business owners on the likelihood of further disorder in the future. If that seems probable, or even possible, the uncertainty that creates could cause customers to avoid downtown areas, even if no further disorder occurs. The uncertainty alone will diminish commerce that was already crushed by the pandemic.
There is another class of dynamics I call hidden extremes because the long-term trend appears benign even as it reaches breaking points with the potential to collapse the system.
Cost is my ongoing example. The costs of operating a small business have been rising far faster than official inflation or incomes for years. Rent, utilities, licensing fees, taxes, wages, labor overhead, insurance–virtually every category of expense has climbed inexorably for years.
These increases in fixed costs (costs that are unrelated the number of customers served) have pushed many small businesses closer to the edge of insolvency. To compensate, owners have cut employee hours and shouldered more of the day-to-day work themselves.
But there is a limit on this kind of workaround; the owner can only work so many hours a day, and every additional hour increases the odds of burnout, a complete collapse of the owner’s ability to continue over-working.
I call this the Rising Wedge Model of Breakdown: costs ratchet higher effortlessly, but reducing costs encounters extreme resistance.
In other words, a consequential percentage of small businesses were at their extreme limit in the rising wedge even before the pandemic. Now the wedge has broken as their revenues falling by even a modest percentage is enough to trigger losses they cannot sustain.
Another dynamic that can make extremes even more extreme is the Pareto Distribution, a.k.a. the 80/20 Rule: the vital 20% wields outsized influence over the 80%, and the 20% of the 20% (4%) exerts outsized influence over 80% of the 80% (64%).
Just as 80% of sales come from the top 20% of sales staff and the top 20% of households end up with 80% of the wealth, the top 4% can wield non-linear influence over the 64% if they gain the power to enforce a positive feedback loop to increase their power at the expense of the 64%.
While we hope the best 4% will gain this influence, history suggests that the worst 4% (sociopaths, etc.) are highly motivated to seek power in a vacuum or when the opportunity presents itself.
The 64% tend to hope for the best even as the 4% tighten their grip on the economy and social order. This is the totalitarian feedback loop illustrated by the rise of the Nazis in Germany and the Communists in Russia.
But the 4% need not wield direct power; it is enough that they threaten or disrupt the certainty of the 64%.
For example, if the movement to de-fund police departments triggers mass resignations of police officers, the 4% criminal element will quickly increase their predation on the 64%, who will then lose the presumption of relative safety required to conduct commerce.
Again, uncertainty becomes a self-reinforcing feedback that disrupts the economy and the social order, because people make different decisions when they lack certainty in outcomes and the future.
In other words, the actual crime rate need not increase by much to trigger a complete recalculation of risk and uncertainty that could then trigger a mass exodus from city centers by small businesses and the top 20% of households with the most to lose and the most mobility.
Once these sectors abandon the city, the economy and social order collapse to levels that no one thought possible. Again, the point here is effects everyone thinks are linear quickly become non-linear: thus a 10% increase in crime doesn’t cause a linear 10% reduction in commerce, it triggers a 50% decline in commerce which then unleashes a second wave of decline as the loss of 50% of small businesses reduces the attractiveness and safety of the hollowed-out neighborhood.
In my analysis, costs for small businesses and urban residents were already at extremes that were hidden or accepted as “normal.” What few understood was how pushing costs into the top of the rising wedge made the entire system vulnerable to non-linear breakdown. This breakdown is what I see unfolding in the economy and the social order.
Extremes will become more extreme because the positive feedback loops of the Pareto Distribution are overwhelming the moderating negative feedback loops of resilience (i.e. buffers), certainty and institutional trust/credibility.
The financial system is extremely vulnerable to disruption and collapse for the same reasons: the extremes are not visible to the vast majority of participants, and so they are exposed to high levels of risk they don’t see or understand.