In many of my economic essays, especially "The Limits of "Scientific" Management" and "Greed Versus Evil", as well as "Bad Economics Part 15", "Bad Economics Part 12 ", "Bad Economics Part 10", "Liberalism, Its Origins and Consequences" (specifically "Liberalism, Its Origins and Consequences - Chapter 3 - The Truth is Out There") and "What Is Fair? or, How Game Theory Leads Us Astray", I have made the argument that government control of the economy is destined to failure because human desires are both subject to sudden and capricious change, and exist only as a shifting scale of ordinal values, the relative weight of which is constantly being determined. Or, to put it in simple terms, humans are volitional beings, whose actions might follow certain logical patterns in general, but whose preferences, desires, wants and fears are predictable neither for the individual nor the aggregate to a degree sufficient for purposes of government intervention into, or control of, the economy. It is not a new argument. It is the basis of von Mises argument in Socialism that socialist systems (including communism) are inherently doomed.

What makes this argument interesting is the reaction it receives. For some reason, people are willing to give economists and other experts almost magical powers when they develop theories of government, or else they will argue strangely that what is true of individuals somehow is no longer true in the aggregate. ("Individual and Aggregate") However, the fact remains that what is true of one is true of ten, as ten is just one plus one plus... Aggregate numbers are only predictable within certain limits if individual behavior is predictable within certain limits.

Allow me to explain. People often misunderstand statistics. They seem to think you can take unpredictable individuals, lump them together and come up with perfect, or near perfect, predictability. The problem is, that view is wrong. Predictions for groups are only possible if we can also predict for the individual. The less we know about individual choice, the less we can predict for a group of those individuals. We can predict for many only because we can predict equally well for each member. The only thing statistical analysis allows if for us to use smaller samples to extrapolate to larger ones, and to "smooth" numbers, removing the effects of aberrant outliers and other unusual data. However we can only do so because we know the relative likelihood of such outliers.

Which brings me to my main case against intervention.

If anyone believes he knows that prediction is possible, then let me ask him some simple questions. What will he eat for dinner one year from tomorrow? In a week, how many carrots would he trade for an ounce of iron? And, of all books currently written, which will be his favorite in three years? His second favorite? Third?

These sound like absurd questions, and they are. We do not know what the future holds, and certainly cannot tell with precision what we will prefer in a year or two, or even a week. In addition, when the question deals with preferences and wants, it is almost impossible to know even a short time in the future. We cannot know if tomorrow we will prefer one food to another, or will want more of one drink or another. Our personal preferences, and their relative values to one another, are impossible to predict, and that includes moment to moment, not just in the long term.

And that is the problem. If a single individual, with perfect knowledge of himself, his circumstances, his inner state, including absolute knowledge of his own relative valuations of every possible good and service, cannot tell with any certainty what he will prefer in a day, or cannot tell in an hour or a day whether he will value X more than Y or Y more than X, and in either case what the ratio of the two will be, then how can anyone claim to know that of a stranger, of whom they have less perfect knowledge, much less of the entire economy? As I said above, a single individual and an economy are the same. There is no difference between the aggregate and individual. The economy is a mass of individuals, but each actor retains his identity and acts individually. Similarly, our knowledge cannot be aggregate knowledge. To understand the aggregate we need to know the individual, and since we cannot know the individual sufficiently, we cannot know the aggregate.

Thus, so long as we agree the purpose of the economy is to satisfy the wants of those participating in it ("The Most Misleading Word", "Luxury and Necessity), we cannot promote government intervention s a possible solution, as it lacks the requisite knowledge to perform the job adequately. In fact, it not only lacks the knowledge, but cannot possibly gain that knowledge. Not only is the knowledge unavailable, it is inherently impossible to obtain, and thus, not only is an effective government run economy impossible in the present, but it is, by definition, impossible. (And that is ignoring the other issues, such as the necessity of bureaucratic management and its related problems -- "Bureaucratic Management", "The Bureaucratic Mind", "Bureaucracy Revisited", "The Inevitability of Bureaucratic Management in Government Enterprises", "Bureaucratic Management and Self-Policing" -- as well as the difficulty of defining what goals the government should set for the economy -- "The Inherent Disappointment of Authoritarianism", "The Inevitable Corruption of Protectionism".)

But some will deny that argument, theorizing that some sort of economic sampling could possibly exist which would solve these issues. It has always struck me as a weak argument, proposing theoretical possibilities, but people seem to have unwarranted faith in academic experts, especially in economics, and so they believe that somehow there will arise a model which can perform miracles. So, rather than pursue my first argument, let me make a different one.

That argument is rather simple. Physics works because the rules are absolute (assuming we are discussing Newtonian mechanics, not the less deterministic theories of the subatomic world), and because the objects with which you deal are inert, unable to express their will, as they have no will, or, if they do, it is disregarded for the purposes of the model. (Eg. Calculating the trajectory of the body when you throw a human from a catapult does not need to consider his wishes in the matter, as his landing point will not be changed by his desires.) On the other hand, social sciences, such as economics, and other sciences of human choice, deal with non-mechanistic models. Just as the economists learned when they tried to apply the "hydraulic" model of money supply inflation, humans are not deterministic in their response. We can know that increasing the money supply will cause prices to rise, but that rise is proportional to the percentage increase in money supply only by accident, thanks to human beings' volition, which can take into account expectations about future money supply growth, anticipation of future increases in prices and interest, and dozens of other factors, and as each can reach a different conclusion, the price increase caused by money supply inflation is not predictable.  In fact, in some cases, there may even be no price increase. Sometimes, in very rare circumstances, it could even fall. It all depends on the beliefs and choices of humans, and those are never predictable.

Unfortunately, our modern society with its tendency to rely too heavily on psychology, as well as turning every behavior into a "disorder", has come to imagine humans are somewhat mechanistic. For example, the belief that abused children become abusers, as if it were a predetermined response. How many people in the world have met someone who has done the opposite, however? Someone whose childhood was upset by abuse or drinking to other problem, and they swore they would never do the same to their children? And how often have we heard the supposedly inspirational stories of a child whose parents were drug or alcohol abusers, child abusers, or otherwise damaging to their family, and yet the child overcame the hardships and achieved great success? Of course there are many who go the opposite route, and end up abusing their own children, but it is worth noting that neither seems to be the "certain" course for such an individual to take, and so the simple, mechanical belief is clearly wrong.

And that is my point. Pick any experience, good or bad, small or large, and of a hundred individuals, I doubt any two will respond the same, even given the most homogeneous group imaginable. Nor is that a controversial assertion, I doubt any but the most strident Skinnerian would claim otherwise. Yet, we somehow forget this when it comes to economics, and let models and charts and numbers lull us into forgetting that we are dealing with humans, volitional beings, whose preferences are wholly unpredictable, and non-rational*, and we think instead that we are dealing with some machine called "the economy" which can be determined with absolute certainty.

It is a dangerous delusion under which to labor. Many interventionist governments have discovered to their dismay that humans are not the cogs they once imagined, nor are they as predictable or malleable as many would believe**. Relying on humans being docile followers whose wishes are easily known is the one sure way to doom an economic system.

------------------------------------------------------------------------------------------------------------

* I do not say "irrational" as many take that to mean "crazy" or "ill considered" and people's wants are not crazy, they just are not based on reason either. What I prefer is based on nothing but my individual wants, which are effectively given data. We cannot argue with others about what they do and do not find pleasing, nor about what wants they find most or least important to satisfy. I suppose we could say pursuing certain wants would be detrimental to achieving certain other goals, but even there, it is a matter of deciding which goal is more important. In any case, what I want to make clear is that the subjective valuations of each individual, the way they arrange all their wants into a hierarchy of relative importance, is not rational, it simply IS, and we must accept it as a given.

** I think in some cases this may be why the left so often claims advertising is so powerful. Their political philosophy requires them to believe people are "sheep" ("Sheep", "The Road to Violence"), and so they have to believe advertising can control the minds of consumers, well other than them and their elite peers. ("Regulated Speech")  but there is a problem even if advertising worked as well as they suggest. For advertising to work, you have to have a "hook", something the buyers want, that you can promise a good or service will provide. So, even for advertising to work you must have an idea of what their relative wants are, and so you still need the same information that advertising is supposed to obviate. Which means even with all powerful advertising you would still need to know the unknowable to make advertisements effective enough to replace the unknowable data.