Predictability and Adaptability

From No More Quick Fixes by Guy Sorman.

Economists, politicians, and pundits looking for answers to the economic crisis fall into two broad categories. Keynesians and statists argue for more aggressive interventions from governments and central banks. Distrusting the free market’s self-regulating processes, they promote public spending to create jobs and low interest rates to rekindle private investment and consumer spending. Thinkers of the classical-liberal persuasion, by contrast, argue that no quick fix can bring the economy out of its doldrums; only when the rules of capitalism appear stable and predictable again will markets revive. Put another way: Keynesians and statists believe in flexible, “discretionary” economic policies; classical liberals believe in set rules.

Economic history proves the superiority of the second approach, but democracy often makes the first more attractive to politicians. After all, in a crisis, people expect their leaders to do something; refraining from action and sticking to abstract principles play poorly to public opinion. As previous recessions demonstrate, however, public pressure for action usually leads to bad decisions that prolong or intensify a crisis. The situation is analogous to what happens on the soccer field when a goalie faces a penalty kick. Statistics show that the goalie should stay in the center of the net to increase his chances of blocking the shot. Yet in most cases, he jumps to the left or right just before his opponent kicks. Why? Because the crowd urges him to act, even though doing so reduces his likelihood of success.

It is nice to see an essay going beyond the usual labels and habit of lambasting based on labels. I think Sorman is focusing on a critical issue that often is overlooked. In any stable system, there is a very high premium on predictability as prerequisite for productivity. Everywhere and through all our recorded history, independent of the perceived fairness of the laws, all those countries with a functioning system of law and cultural habit of placing rule of law above rule of man, there is a clear correlation with increased productivity. When laws are overruled haphazardly and capriciously, you lose predictability and more critically, you lose productivity.

However - No system is inherently stable. Most growth systems follow some form of a sigmoid or S-Curve. At the bottom there is a slow take-up of an idea or opportunity or resource or technology. There are the early adopters, the gamblers. People who end up making a lot of money by being well prepared and taking big risks and being at the right place, at the right time. Then there are the companies that come in and focus on exploiting the steep growth period, focusing on either (or both) efficiency or customer service. Eventually, the idea or technology or product inflects again towards the top of the S-curve. Now is the period of decling returns where all efforts are on efficiency and brand. The market is saturated and it is hard to differentiate from the commodity market competitors.

How long each section of the S-curve lasts depends on context, circumstance and history. Here is what it has looked like for various technologies.

There are some predictable inflection points.

But what happens next? The next S-Curve. You might liken it to Thomas Kuhn's paradigm shift or to Stuart Kauffman's adjacent possible. As you reach the end of one S-curve and have fully exploited all the capabilities of the idea, resource, technology, you have set up the circumstances where it becomes feasible to consider an entirely different S-curve (usually with little capacity to predict the exact course and inclines of the new curve). Horses give way to cars; TVs give way to computer/internet; Books give way to e-readers; etc.

So how about a real world example. Here is what the double S-curve of K-12 education might look like.

So what do S-curves have to do with Sorman's article. Simply this. Increased productivity (success) depends on accurately forecasting where you are on the S-curve and the predictable rule of law is particularly critical throughout the S-curve journey but most especially so at points of transition between S-curves. At the point of transition, you have already a huge burden of risk (new technology, new business processes, new manufacturing plants, new markets, etc.) which are all detriments to making the inevitable transition. If you throw in "discretionary" policies as well or undermine the rule of law, you increase the barrier to transition even higher.

Our policy makers are much more geared towards winning their next election and therefore championing arbitrary "discretionary" policies which almost always fail on their own merits as well as generating negative unintended consequences one of which is the totally unconscious impact on reduced predictability.

The challenge is that all along the first S-curve you absolutely need predictability. However, when you are at the transition point between S-curves, you also absolutely need adaptability (or "discretionary" policies). That is just the nature of the beast.

For example, we are at the transition point between curves for any non-physical cognitive product (books, music, etc.) The copyright rules that made perfect sense for a physical world product of a book or a record no longer work well in a digital world. The law has to adapt.

So we do need both predictability and adaptability but at different times and under different circumstances. The lost productivity occurs when we misapply each response. When we willy-nilly change the rules while we sit on a single S-curve we lose predictability and therefore productivity. When we refuse to make changes at the point of transition between S-curves (often because of regulatory capture, rent seeking and entrenched interests) we also lose productivity by either failing to make the transition or incurring added costs by delaying the transition.

The ultimate challenge is that people, businesses and politicians are usually pretty bad at estimating where they are on an S-curve. There is always noise in the system with short term ups and downs of varying magnitudes. It is quite easy to confuse a short term tactical dowturn in the middle of the curve as an indication that you are actually approaching the top of the curve and beginning to flatten out. So who makes the best estimates of where you are on the S-curve and when to start making risky transition investments? Whoever is closest to the feedback loop and bearing the greatest consequence. This is a classic manifestation of Hayek's problem of knowledge.