Thursday, July 9, 2020

Inefficiency Is Good!

Back in 2015 I wrote Brittle systems and Pushing back against network effects, among other things about the need for resilient systems and the importance of antitrust enforcement in getting them:
All over this blog (e.g. here) you will find references to W. Brian Arthur's Increasing Returns and Path Dependence in the Economy because it pointed out the driving forces, often called network effects, that cause technology markets to be dominated by one, or at most a few, large players. This is a problem for digital preservation, and for society in general, for both economic and technical reasons. The economic reason is that these natural but unregulated monopolies extract rents from their customers. The technical reason is that they make the systems upon which society depends brittle, subject to sudden, catastrophic and hard-to-recover-from failures.
Now, the pandemic has inspired two writers to address the bigger version of the same problem, Bruce Schneier in The Security Value of Inefficiency and Jonathan Aldred in This pandemic has exposed the uselessness of orthodox economics. Below the fold, some commentary.

Aldred, a Cambridge economist, focuses on the way economic theory constrains our view of the future by using models that assign probabilities to future events based on the past, and score their results based on efficiency:
There are clear lessons here for how to address the climate emergency: rather than focusing on the average climate impacts predicted by mathematical models that depend on probabilistic knowledge that is highly unreliable, we must think seriously about worst-case scenarios, and take steps to avoid them. Yet economic orthodoxy pushes us away from precautionary action. If mainstream economics has a single overarching objective or principle, it is efficiency.

Efficiency means getting the most “bang for your buck”, the most benefit for every pound spent. Any other course of action is wasteful, surely? But eliminating waste implies eliminating excess capacity, and we now see the consequences of that in health systems worldwide. Our obsession with efficiency, if it means failing to plan for a pandemic or a climate emergency, will cost lives.
So in practice:
this thinking means that the value of everything is measured by how much people offer to pay for it. Since the rich can always pay more than the poor, priorities get skewed towards the desires of the rich, away from the needs of the poor. So more money is spent on R&D for anti-wrinkle creams than for malaria treatments. Big Pharma has been relatively uninterested in developing vaccines, because a vaccination programme only works if the poor get vaccinated too, which limits the price manufacturers can charge.
Aldred concludes:
Our priority should be resilience, not efficiency. We need to build resilient systems and economies that are explicitly designed to withstand worst-case scenarios – and have a fighting chance of coping with unforeseen disasters too.
Well, yes, that was what I was arguing 5 years ago. But how to do that? Aldred has only this to say "we must re-think our economics from the bottom up". I would argue, as I have many times since 2011, that one thing that needs to go is Discounted Cash Flow (DCF). Using a discount rate, as Haldane and Davies pointed out always an excessive discount rate, makes the "Net Present Value" of a catastrophe a decade or so away negligible.

Schneier is a better writer than I am, has a much broader canvas, yet is much briefer. Here's a sample:
Efficient systems have limited ability to deal with system-wide economic shocks. Those shocks are coming with increased frequency. They're caused by global pandemics, yes, but also by climate change, by financial crises, by political crises. If we want to be secure against these crises and more, we need to add inefficiency back into our systems.

I don't simply mean that we need to make our food production, or healthcare system, or supply chains sloppy and wasteful. We need a certain kind of inefficiency, and it depends on the system in question. Sometimes we need redundancy. Sometimes we need diversity. Sometimes we need overcapacity.

The market isn't going to supply any of these things, least of all in a strategic capacity that will result in resilience. What's necessary to make any of this work is regulation.
Schneier is also much more specific about what needs to happen than Aldred:
  • "we need to enforce antitrust laws" Why is this?
    Our meat supply chain is brittle because there are limited numbers of massive meatpacking plants -- now disease factories -- rather than lots of smaller slaughterhouses. ... We need more local companies, more domestic corporate players, and diversity in our international suppliers. Competition provides all of that, while monopolies suck that out of the system
    Even if you believe (falsely) that the efficiencies of industry consolidation accrue to the customer rather than to the remaining executives, it is clear that the costs of future failures are not merely imposed on the customers (and thus underestimated by DCF) but also, as we see with the pandemic, on taxpayers and society at large (and thus not even included in the DCF calculation).
  • "we need ... specific regulations that require certain inefficiencies" Why is this?
    Every safety system we have is, to some extent, an inefficiency. This is true for fire escapes on buildings, lifeboats on cruise ships, and multiple ways to deploy the landing gear on aircraft. Not having any of those things would make the underlying systems more efficient, but also less safe.
    All these additional costs are imposed by "specific regulations". Boeing's 737 MAX disaster is an object lesson in inadequately "specific regulations".
Just go read Schneier's whole post. I really have nothing to add to it.

1 comment:

David. said...

David Krakaeur and Geoffrey West, who study complex systems at the Santa Fe Institute, have a fascinating post entitled The Damage We’re Not Attending To. The whole thing is worth reading but in the context of this post the key part is (my emphasis):

"A key insight from complexity science is that complex systems function by the continuous tradeoff between robustness and evolvability. Robustness describes the ability of a system to withstand a critical perturbation without a significant loss of function. ... Evolvability describes a mechanism that allows for the efficient exploration of adjacent novelties, whereby small changes to a mechanism or structure can engender new functions. ... Evolution has discovered the means of balancing the competing demands of these two principles and together they account for both the long-run stability of lineages and the diversity of life. Contrast this complex, evolved reality to our modern, social-technical world. A typical strategy of companies and corporations is to eliminate redundancies and degeneracies in the name of minimizing costs. This is the major reason why almost all companies have great difficulty adapting to change, and eventually disappear. Just as biological systems pay a cost for robustness and evolvability foregoing efficiency for long-term persistence, so too should we demand this of our institutions."