It always happens...

All human groups, from the smallest tribes to the largest empires exhibit social stratification; larger populations tend to generate more 'layers', but the tendency to stratify seems an invariable feature of humanity. I believe this springs from two factors; people can only closely relate to a maximum of one or two hundred others, and people generally don't like others who are not like them. These factors seem to be 'hard wired' into us, so the interaction of these two forces will always result in stratification by what we can call 'social class', loosely defined here as how successful the individual becomes. The question then becomes, what makes people 'successful', meaning aggregating money, power, fame, an attractive mate and the like. My modeling suggests that only three, easily observable factors determine the 'success' of an individual and predict the level of stratification that develops from it. The first, of course, is 'smarts'.

First, you have to figure things out...

Arguably the most important factor is intelligence, whether measured by IQ or by by the individual's observed ability to think and plan ahead. As we showed before, intelligence is multifactorial with a substantial genetic (at least 50% heritable) component. Because of its complex nature, the distribution of intelligence in a population follows a bell-shaped curve, and either at birth or not long afterward, you are stuck with what you have. If this was the only factor for success, an individual would be frozen in place for life, but we all know from experience this does not always happen, so we need to add other factors.

99% Perspiration

Thomas Edison said that “Invention is 1% inspiration and 99% perspiration”. In a similar vein, an old professor of mine had a saying: “If you pee in one spot long enough, you're bound to wear a hole”. They meant, of course, that perseverance can pay off, even in simple things, so we have a second factor for success, which we can call “diligence” (or perseverance). In this case, diligence means having the drive to overcome discouragement, put off current pleasures for future rewards, and allow one's self to be goal-directed. This trait is also likely to be distributed in the population in a bell-shaped curve, running from the 'scratch your butt and roll over in bed' to the 'on the go 24/7' personalities. Behavioral genetics strongly (and controversially) suggests that at least half of an individual’s behavior is genetically determined (and thus partially heritable). Diligence, being a behavioral trait will therefore also have some genetic components (brain serotonin/dopamine ratios??). For now, we will assume that it does not co-vary with intelligence, so the two can be treated as totally separate factors. This leads us to the final success factor – which is a bit of a cheat.

When Tyche smiles...

Luck is the third factor, and it is a grab-bag of possibilities, from marrying the bosses daughter to not falling in front of a bus, looking beautiful or ugly, being weak or well-muscled, and so on. Life is filled with stochastic events which can turn out either well or badly – just like a coin toss – so this factor also deserves a nice bell-shaped curve, describing how some people will be incredibly 'lucky' and others true losers in life's lottery. Although it has been said that 'people make their own luck', this isn't really true. Random events happen randomly, but people who are prepared (smarter and/or more diligent) will tend to make the best of the result. So, now we have the three factors, each with a nice bell-shaped distribution, and we need to put them together to calculate someone's potential for success.

Mix-Master Mathematics

As we showed before, if you combine a series of random events, you end up with a symmetrical bell-shaped curve, such as we have with each of our separate success 'ingredients'. One might think that if these are blended, the combined distribution in the population would also be a similar curve, but this turns out not to be the case at all, as these factors potentiate (or degrade) each other rather than just add up. Let us assume 'success' consists of two parts intelligence and one part each of diligence and luck, so we have 'S = I*I*D*L'. A 'person' starts out with a value drawn separately from each of the three distributions, so most people will have average results from each and end up with average total 'scores'. However, a few people will be fortunate enough to have high scores on all three, and the effect is to boost them way out of the mainstream.

This curve shows the ‘success’ distribution from the simulated population after about ten generations, including an assortative mating factor which does not allow marraige to someone who varies more than 30% from the spouses 'success' factor. Looking at this distribution, we find that the people who draw poorly in all cases don't fall as far from the average as the people at the top rise, so the distribution is highly skewed to the right. The distribution was calculated so the average would be at exactly 1.0. As you can see, we have a few values less than half that of the average but many more values two times the average (the one percenters); there are even a few near 4.0, and if the population was larger, one would find even farther outliers.

Cart track and Autobahn

If you have a five hundred horsepower car, it may not win a race against a horse-cart if you are competing on an overgrown forest trail, but on the Autobahn there will be no contest. Thus, the overall cultural environment can vastly potentiate natural advantages, and a highly complex technologic society will hugely benefit those with the most horsepower, pushing the effective right edge of the 'success' curve to many multiples of the average. Assuming (as I did in my models) that the 'success quotient' will more or less translate into earning power and accumulated wealth, the distribution of money resembles what we more or less see today. Admittedly, today's income distribution is even more heavily weighted to the very top, principally due to governmental and Federal Reserve policies, but the differences between the extremes of the population become immense and tend to grow over time because of assortative mating.

Almost another species...

One definition of a species is that it is a related group (of organisms) which can not (or do not) breed outside the group. Most species truly can not interbreed and create viable (and fertile) offspring, but there are those who can, but rarely do so. The lion and tiger are separate species of the genus Panthera, but they can interbreed. The reason that we aren't awash with 'ligers' or 'tigons' is that they don't occupy the same environmental space or have the same behavior patterns, and thus stay apart. Although humans can readily cross 'ecologic and cultural niches', as pointed out by Charles Murray (Coming Apart), there is a strong tendency to marry within one's own stratum (assortative mating) and thus perpetuate, if not accentuate any genetic and/or environmental benefits. The result of this seems to be an ever-increasing consolidation of these benefits (or deficits) within particular segments of society.

Whether this is good or bad depends upon one's view of how societies should be organized (and no doubt also upon one's place therein), but it does raise the practical question of the role (or plight) of those on the left hand side of the success curve.

Hewers of wood and drawers of water...

In earlier and less developed societies, there was always the need for the very low-skilled, and to some extent this can still be true in some rural settings. However, much of the world is no longer organized as it was in the 19th Century, and there is little gainful (or economically efficient) employment for those who don't have – and can never get – the skills needed for at least minimal success in a modern technological and largely urban environment. How large a fraction of the population this represents is difficult to say, but even if it were as low as ten percent, this is a huge number of people, and as history shows, this can present a problem.

Without gainful employment and a stake in society, these populations can quickly become restive. The classic way of handling this has always been the 'bread and circuses' approach, where minimal nutritional needs were met, but more importantly, distractions were provided. Today, we find food stamps and cable television substituting for the bread dole and the Circus Maximus, but the principle is the same.

An illusion of Democracy?

There have been many ways of organizing societies over the millenia, and they all involve some form of social stratification. As populations increased, tribal groups consisting of chiefs, shamans, warriors and members morphed into kings, clergy, nobles, knights and commoners; the basic structure of feudalism. Today we have a social pyramid going from (one or more) top political figures to Bureaucrats/Academia/Media to Military/Business to artisans and workers, down to the lightly-employed (or unemployed) masses.

In many developed countries there is a popular vote to elect the leaders (and order the society), and in theory, each person in a democracy has one vote, regardless of their social class. However, where do the people they vote for come from? We can tell from computer models and real-life experience that being on the right-hand side of the curve has a substantial hereditary and environmental component, which fosters political and business dynasties which can last for generations. Some of the progeny of these families can turn out to be ‘duds’, but it highly likely that leaders will come from this group and any ‘outsiders’ will have a hard time gaining traction (or even notice). Thus, regardless of any voting, the actual results suggest that leaders tend to be chosen from their own class, and except for cosmetic changes, policies are largely determined by the (unelected) bureaucratic class.

Relax, it's only a model...

Economics is often called “The Dismal Science”, not because it always predicts bad ends but because it points out the limits of human control over complex systems. By their nature, complex systems with unpredictable interacting parts are impossible to accurately model – economies and climate are good examples. The models may give some insight as to how different components may interact, but even the people who write them don’t claim they have any predictive value. So, in the end, if we can’t predict what will happen (and we really can’t), we just have to muddle through – like we always do.