The Unconventional Legacy Of Adam Smith (Part I)

What the father of modern economics can teach us today

Daniel Issing
Sep 5, 2018 · 12 min read

I.

C.S. Lewis once wrote:

"Every age has its own outlook. It is especially good at seeing certain truths and especially liable to make certain mistakes. We all, therefore, need the books that will correct the characteristic mistakes of our own period. And that means the old books."

To what extend his advice is heeded depends on the discipline you’re looking at. At the one extreme, philosophy deals a lot with long-dead figures (some representative examples), and you probably won’t complete a degree in philosophy without reading at least some original literature all the way back to the ancient Greeks. (I would agree that this focus is by and large unjustified.) The other extreme are the natural sciences. While I was studying physics, the oldest books I ever came across were probably Landau and Lifshitz’s Course of Theoretical Physics (written in the 1930s, published in the 1950s) or Feynman’s Lectures on Physics (1961–63). However, the reason we used them was not to offer a different perspective, but because they were wildly considered the best presentations of theories that haven’t been refuted yet. You most definitely wouldn’t look for help in, say, Newton’s original work.

Where does economics fall on this one-dimensional scale? Should it focus exclusively on the latest textbooks and scholarly articles as the only relevant source of information? Or should it seek out for the time-distilled wisdom of the old masters, like Adam Smith? I try to answer this question in a two-article series, the first of which will focus on markets and emergence, while the second is centered around moral philosophy.

II.

One would be hard-pressed to find a metaphor from the domain of social sciences as persistent as the idea of an 'invisible hand'. Adam Smith, justly considered the founder of economics, popularized this notion in his world-famous An Inquiry Into The Nature And Causes Of The Wealth Of Nations, and it has been celebrated, criticized and misunderstood more than any other aspect of his work [1]. In the contemporary public image, Smith is seldom revered as an intellectual giant who laid the foundation for an entirely new branch of science, but rather as someone who, despite a few remarkable insights, is to be blamed for originating a thoroughly naive theory of the free market, according to which greedy business people somehow benefit humanity as a whole through their selfish actions. Even among economists, voices that declare the invisible hand a quasi-theological justification for dog-eat-dog capitalism, or claim it to be tied and rendered immutable by (negative) externalities, are gaining influence. Given these trends, what could possibly be found in Smith's oeuvre that would still be relevant to 21st century netizens?

It of course all depends on how your understand ‘relevance’. One take, the narrow sense, posits that we’d do well to turn our attention to the old masters when looking for solutions of today's problems. On this account, answers to complicated bioethical dilemmas would lurk in the writings of Plato or Seneca, disputes in evolutionary biology can be resolved by consulting On The Origin Of Species, and Copernicus might have a word or two to say on how to interpret recent astronomical data. The other, wider sense of the word asks whether the opus of a thinker has either directly inspired modern scientific breakthroughs, or at least been an import brick in the edifice that subsequent generations of researchers have erected. This is what we have in mind when we talk about how philosophical atomism, as developed by Leucippus and Democritus, served as a guiding light for modern chemistry.

By picking such candid examples, I should have made sufficiently clear that I believe 'relevance' in the first sense, as applied to Smith's work, to be indefensible. More than 200 years after his death, we’d be well advised not to get caught up in catechist rituals. The world is a very different place these days, and the concerted effort of numerous scholars and researchers has elevated economics to a state Smith himself could not possibly have foreseen. That being said, I do think there is a lot to be said in favor of the second, weaker sense of his continuing relevance. Some points are obvious and much talked about these days – for example that the appeal of populist movements depends to a large extent on a widespread misunderstanding of the benefits of free trade and other basic economic principles. But rather than continuing to flog this dead horse, I’d like to focus on some more unusual places where the specter of Adam Smith is still haunting us. Idiosyncratic as such a view may be, it hopefully adds a fresh perspective to the impressive literature of Smith scholarship that exists today.

III.

In Darwin's Dangerous Idea: Evolution and the Meaning of Life, philosopher Daniel Dennett describes the idea of natural selection as a "universal acid: it eats through just about every traditional concept, and leaves in its wake a revolutionized world-view, with most of the old landmarks still recognizable, but transformed in fundamental ways". Smith's invisible hand, too, displays this kind of acidic property, and in fact shares a lot of common ground with the notion of biological evolution.

To truly understand its significance, one has to picture the intellectual climate around the time of Smith's writing. In political philosophy, the impact of Hobbes' Leviathan, portraying states as the result of deliberate, rational concessions on behalf of the citizenry, dominated the debate. In economics, mercantilism had its heyday, developing variants of Colbert's dictum that "simply, and solely, the abundance of money within a state makes the difference in its grandeur and power". In the religious sphere, the watchmaker analogy served as a potent meme to transport the idea of life being intelligently designed by a creator. Unrelated as these domains seem to be, the underlying mode of thought – what Friedrich Hayek would call 'constructivistic rationalism' much later – is the same: Order and purpose can only arise from a plan, and this plan can either be discovered and mechanistically dissected, or imposed upon a country to lead it on the path to glory. That order doesn’t need a central commander, and can indeed result from the myriad of undirected interactions between individuals (or micro-level constituents, more generally), was almost inconceivable.

A French harbor during the heyday of Mercantilism.

Enter Adam Smith. Although hardly the originator of the concept of spontaneous order, he deserves credit for the hitherto most systematic exposition of institutions that are the result of human action, but not of human design. A prime example of this is how the price system in a market economy works: Despite the lack of chief planner, it captures crucial information in a way that couldn’t be calculated from scratch based on the ever-changing preferences of millions of buyers and sellers. A comparative analysis shows that it does so much more efficiently than even the most technologically advanced politburo, ensuring that you and I find apples, detergent, smartphones, musical instruments and countless other items in a nearby store whenever we desire them, without prior notice. From a God's eye view, this may well appear as if someone was pulling the strings in the background.

But the invisible hand doesn’t confine itself to the study of market prices. Being a universal acid, it has the potential to permeate almost every aspect of our understanding of the world when properly generalized. I suggest this natural generalization is emergence, described by economist Jeffrey Goldstein as the concept "refer[ing] to the arising of novel and coherent structures, patterns and properties during the process of self-organization in complex systems." Smith didn’t foresee such an extension of his work, mind you, but he did hint at a new avenue for understanding complexity that transcends the study of market economies.

Ant colonies are often cited as a prime example of emergence in nature. “colony of fire ant” by Mikhail Vasilyev on Unsplash

Admittedly, there’s something in the notion of emergence that makes it hard to accept for modern scientists, and it’s the same reason why the invisible hand is often ridiculed as a 'quasi-theological' concept. Namely, in the strong sense of the word, it implies a metaphysically questionable downward causation that cannot be reduced to processes at the micro level - think of Hegel's Weltgeist or Marx' s productive forces. To some extent, the uneasiness with which the concept has been treated in philosophy is a mirror image of the common accusation that classical liberals have blind faith in free markets: Everyone pursues their own selfish goals and poof! an invisible hand occurs to ensure they are somehow making society better off. In both cases, are we trying to get something from nothing?

Fortunately, we can do better than this. Philosopher Mark Bedau suggests a weak notion of emergence that is consistent with a naturalistic framework:

For a system S made up of micro-level components whose time evolution is governed by microdynamic D, a macro state P of S is called weakly emergent iff P can be derived from the external conditions of D and S but only by simulation.

This is a clumsy definition, so let me unravel it by means of an example that shows both why this understanding doesn’t depend on contentious metaphysical assumptions, yet still is a useful category.

The canonical illustration is John Conway's legendary Game of Life. It’s played on a two-dimensional grid with cells that are either dead (white) or alive (black). Over time, these cells either stay dead, die, stay alive or are reborn, depending on the states of the surrounding cells (the neighbors) one time step earlier. Players only interact with it at the very beginning, when they specify the initial conditions. The microdynamics of the game are governed by four simple rules:

  1. Live cells with fewer than two live neighbors die (death by solitude).
  2. Live cells with two or three live neighbors make it to the next generation (survival).
  3. Live cells with more than three live neighbors die (death by overpopulation).
  4. Dead cells with exactly three live neighbors are reborn (reincarnation).
A ‘beehive’
‘R-pentomino’

Zoom out and let things run for long enough, and you will observe certain patterns emerging. Not all of these macro states are equally interesting: There’s a whole class of configurations dubbed 'still lifes' because they don’t change at all (e.g. the ‘beehive’). Others, like the R-pentomino, show chaotic behavior, spawn 'gliders' (patterns of living cells) that 'travel' across the grid. You can actually test it yourself if you don’t believe me (plus, it’s fun!).

What determines if a given initial set-up will exhibit the propensity of creating such gliders? In the general case, the answer is that there are no simple rules to settle this question short of actually running the simulation. In fact, Life is undecidable, meaning that there are cases for which it’s impossible to know whether the initial patterns will eventually reappear. No equation or heuristic will come to our rescue; we have to let the simulation do its job in order to derive the game’s behavior.

Simulating Life.

It should be clear that Life squares well with Bedau’s definition. But how useful is this definition, really? If it were little more than a philosophical curiosity with no implications for empirical research, there wouldn’t be much of a need to bother. Fortunately, this isn’t the case. Emergent phenomena are very different from a lot of systems for which macro states can be derived without restoring to simulations - physics is ripe with them. Contrast, for example, a simple with a double compound pendulum: In the first case, the laws of classical mechanics allow us to calculate the position at some later time to almost any degree of precision; we don’t have to watch a physical pendulum swing. This is no longer true in the second case, when the equations of motion become so intricate that even the smallest changes in initial conditions result in large deviations shortly after its release [Test it yourself!]. The chaotic behavior of the double pendulum is a good example of emergence that remains firmly in the naturalistic tradition.

Once you understand the concept, it’s hard to miss its ubiquity in modern research. In many cases, emergent phenomena are so counterintuitive, given what we know about the individual constituents, that they appear inexplicable. A sonata is nothing but the combination of different acoustic vibrations, but could you grasp its beauty through an analysis of the relevant wave equations? Neural networks, a key technique in artificial intelligence, derives its strength from emergence. Life and consciousness developed from a cesspool of mindless molecules, but their chemical analysis is of no help for explaining feelings of joy, compassion, despair or excitement. The world wide web, language, cities and the stock market are other examples of emergent phenomena that surround us. Some, as the computer scientist and physicist Stephen Wolfram, even believe that simple Life-like computational mechanisms underlie almost all complexity in the universe, and we perceive them as complex because our brains (and the tools it developed) can’t outsmart brute computation.

IV.

This allows us to take the idea full circle and throw it back onto the study of market exchanges that Smith was originally concerned with. Here as in many other domains, Wolfram's observation that complex behavior can result from very simple rules, rather than from a hodgepodge of blow-by-blow instructions, bears considerable weight. But this doesn’t mean that we have to stand by helplessly and watch things unfold, without ever being able to make as much as a modest prediction about the system as a whole, to say nothing about recommending specific policies (recall Bedau's claim that deriving the macrodynamics requires simulation). After all, none of the above prevents us from observing the relative frequency of certain types of macro events, and derive probabilistic laws from it. For instance, a strong, positive correlation between a country's openness to international trade and its GDP would certainly offer evidence in favor of liberalization measures, to be judged in conjunction with similar indicators.

However, understanding markets as emergent phenomena should lead to skepticism regarding two pet projects of modern economics, namely, mathematization and, in some sense a corollary thereof, perfect competition as the benchmark against which real markets should be measured. There’s no need to suggest some sort of 'physics envy' on behalf of the economists to detect the ongoing popularity of 'mathematically rigorous' models of the economy. But if the simulation condition holds, attempts to come up with more sophisticated methods for calculating market outcomes, such as general equilibrium theory and its dynamic or stochastic modifications, are bound to fail. Quantitative predictions that pass the tests of reality are rare, and snide remarks about how "no economist has yet succeeded in making a fortune by buying or selling commodities on the basis of his scientific prediction of future prices" have aged rather well.

It also means that it’s an illusion to believe we could figure out the conditions for perfect competition, and adjust all the trucking, bartering and exchanging we actually observe to meet these criteria. Smith, to be sure, wasn’t claiming any of this: He simple observed that in most real existing markets, which are soaked in traditions, norms and unfair advantages, buyers and sellers are made better off without impoverishing third parties, division of labor leads to growing prosperity, and trading with other countries is typically superior to protectionism. Evidently, the theory of market failures and externalities has progressed a lot since his days, leading some critics to dismiss the notion of the invisible hand altogether. This move is hard to justify, though. The fact that markets don’t always deliver the best possible outcome doesn’t mean that there will be an easy regulatory fix for it, if for nothing else then because of unintended (and often unforeseen) consequences. If we understand that markets often do good without requiring deliberate action to do good, we may instead want to think about how to better design markets to avoid repugnant outcomes. Promising suggestions, ranging from auctioning schemes to the 'thickening' of markets through clever algorithms that match sellers and buyers, are already being implemented.

Next in line: Part II on Smith’s role as a moral philosopher, in particular his theory of altruism, and why it matters in the modern debate.

Footnotes

[1] To avoid getting accused of distorting Smith’s views myself, I should note that even the introductory quote is taken out of context. Smith, somewhat at odds with to the concocted version popularized by Samuelson’s and Nordhaus’ famous Economics textbook, was arguing for an invisible hand that leads entrepreneurs to invest at home rather than abroad.

Daniel Issing

Written by

Le bonheur et l'absurde sont deux fils de la même terre.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade