An economy is a system of production and distribution of goods and services. It is a system where things are made (production), and then put in the hands of people who will use those things (distribution); the things are both tangible (goods) and intangible (services).

Historically and ethnographically we can categorize economies into three types, based upon the kind of exchange predominating within the economy: reciprocity, redistribution, and markets. This taxonomy is based on the work of Karl Polanyi, and has been very influential among economic anthropologists.

1 Reciprocity

The first type of economy is based upon a kind of exchange called reciprocity. The figure below illustrates reciprocity between actors A and B. A gives something to B, and B responds by giving something of equivalent value to A. The exchange occurs between actors who have a very long-term relationship with each other, involving many, many exchanges. Equivalence is defined by cultural mores, by tradition, where each actor does what is expected, and does not try to advance own self-interest.

Reciprocity: Direct exchange between actors A and B

Reciprocity is the oldest type of economy, and we believe that it was practiced by the earliest humans (about 200,000 years ago). It is characteristic of every ethnographically known foraging (hunting and gathering) society. In these societies the “formula” determining what each actor owes to other actors is mostly defined by the sexual division of labor: men do some things; women do others. For example, among the Inuit of the Arctic, men hunt seals, caribou, etc., and women do most of the rest of the work. A man who chooses not to hunt, or a woman who refuses to work, would be considered a free rider. Free riding can cause reciprocity to break down, and the Inuit, like other foragers, handle that by expelling free riders from the band.

Identifying free riders requires that the group be relatively small, so that limited human brain capacity can keep track of who is doing what. Foragers tend to have group sizes about the same, no matter in what kind of ecological zone they are found: 500 or so in the Z-tribe, i.e., the entire endogamous population speaking the same dialect; and around 25 in the band, the small group that interacts every day, traveling from one hunting site to another. 1 We speculate that the equilibrium size of these groups are a result of the free rider problem: above these sizes, reciprocity begins to break down, because free riders go undetected, and the resulting disharmony causes large groups to fission into smaller, more harmonious, groups.

Robin Dunbar is an evolutionary psychologist at the University of Oxford. Like many in his field, he thinks the human brain evolved to manage interactions among humans (intelligence evolved for gossip, not for algebra), and his research has focused on the relationship between group size and brain size. We don’t actually know the group size of early humans, but Dunbar made a clever estimate. Taking data on group size and a particular brain metric (the neocortex ratio), Dunbar plotted them as in the graphic below. He fit a curve to these data, showing the expected group size for each value of the neocortex ratio. The human neocortex ratio is 4.1, so drawing a perpendicular up from the x-axis at 4.1 until it intersects the curve, and then a perpendicular across to the y-axis, gives us an expected group size of about 150. This number has caught on in the management literature, where it is called Dunbar’s Number. It is considered to be an upper limit to the size of a stable egalitarian group where everyone knows everyone else.

How to reconcile the Dunbar Number with the ethnographic data on forager size? One possibility is that a Z-tribe containing 500 persons would probably contain about 150 adult males, and that it is the interaction of these men that provides the glue holding the Z-tribe together.

Plotting neocortex ratio against group size for 40 primate genera

There are many examples of how limited human brain capacity constrains the number of persons we can productively interact with. Here are a few of the better known:

  • Optimal study group size: Imagine that you get together with other students in the class, to study for the final. What is the group size that will maximize the average amount learned by the group members? To answer this question, social psychologists hired students, gave each student a pre-test to see how much they knew about a topic, then had them learn about the topic in a group (with groups of varying size), and finally gave each student a post-test to see how much they now knew about the topic. The amount of learning was defined as the post-test score minus the pre-test score. Surprisingly, the optimal study group size was found to be two.
  • Optimal decision making group size: Imagine that you are a manager facing an important decision. You assemble a task force to make that decision, and you wonder how many people the task force should contain: too many, and nothing will get done; too few, and not enough different points of view will be considered. What is the group size that will lead to the best decision, on average? George Miller, in 1956, reasoned from experimental evidence, and came up with the answer 7±2 (5 to 9).
  • Intimacy group size: For how many people does the average person feel a strong emotional tie? Researchers investigated this by asking subjects to list all of the names of people who, were they to die, they would feel intense grief. The mean number of names on these lists was 11. Almost all of the names were family members of the subject: either from the family of orientation (the family in which the subject was raised), or from the family of procreation (the family in which the subject served as a parent). The research was conducted in the 1970s, among students who were part of the Baby Boomer generation, and most of these students came from large families. Were the research to be repeated today, the group size would almost certainly be smaller.
  • Match name to a face: You see a photograph of a person’s face, and you know the name of that person. For how many people, would you know their name if you see their face? Early research indicates that the average person can match a name to about 2,300 faces.

We evolved for hundreds of thousands of years in groups practicing reciprocity, and it is natural. But it doesn’t work in large groups. We can’t study well with more than a few other people; we can’t make good decisions with more than maybe nine other people; we can’t learn the names of more than a few thousand people; and in fact, we don’t really care deeply about many other people. Yet we live in a country of 330 million people, where we manage to interact peacefully and productively, creating the most advanced civilization ever seen on Earth. It’s an amazing achievement, and we should be grateful to our ancestors for giving us the institutions that allow us to work together.

2 Redistribution

Redistribution appears first about 10,000 years ago, around the time humans first began practicing agriculture. It appears in larger groups, where reciprocity would not work well.

In the graphic below, actors A and B do not directly exchange with each other. Instead they deliver what is exchanged to a third actor, C, who then decides what should go to A and what to B. This type of exchange involves hierarchy and an element of coercion.

Redistribution involves more than two actors

Societies practicing agriculture usually have higher fertility rates than foragers. The fertility rate is the average number of children women have in their lifetime (this needs to be a bit above two in order to keep the population from shrinking). In foraging societies, women typically walk about three miles a day, looking for food, and must carry their tools and their young children. Since it is not feasible to carry more than one child, foragers have quite long inter-birth spacing, usually about four years. Among agriculturalists, however, women can do most of their work at a fixed, secure house site, where multiple children can be supervised. The inter-birth spacing is only about two years in these societies, resulting in much higher population growth rates.

We speculate that as the populations of early agricultural societies grew, free rider problems appeared. Though these problems were probably usually resolved by large groups fissioning into smaller groups, there were some societies that resolved the problem by turning to redistribution. These large groups had a military advantage over smaller groups, allowing redistributors to spread and replace populations continuing to rely on reciprocity.

All of the well-studied states in the ancient world practiced redistribution: Egypt, the Mesopotamian states, the Indus valley, etc. As an example, consider the Inca Empire. In the diagram above, imagine that actor A is a village on the Pacific coast of Peru, and actor B is a village high in the Andes mountains. Village A employs irrigation to produce maize and cotton, and also catches and sun-dries small fish from the sea. Village B grows potatoes and quinoa, as well as a plant called coca, used as a stimulant. But the villages do not trade directly with each other. Instead, the Inca state demands tribute from the two villages: maize, cotton and fish from A; potatoes, quinoa, and coca from B. The Inca state then decides what A and B shall receive. That decision takes into consideration the interest of the Inca state – goods are allocated in such a way as to tighten the grip of the Inca elite on the people and resources of the Empire.

3 Markets

Market exchange looks like the diagram below. Actor A gives something to actor B; B, in exchange, gives something to A. This looks like reciprocity, but differs in that A and B are exchanging in self-interest: they are not trying to exchange equivalent things, where equivalencies are defined by tradition; each is trying to give something they value less than what they receive. In reciprocity, each exchange is part of a web of continuing exchanges with the same people; in markets, each exchange is an opportunistic, one-time exchange, and can be made with people towards whom one has no social obligations. Market exchange is exchange between strangers.

Markets: Self-interested, one-time exchange between actors A and B

The fact that market exchange occurs between strangers allows it to operate within very large groups. So, like redistribution, markets resolve the free rider problems of large groups. But, unlike redistribution, market exchanges are voluntary and egalitarian, not hierarchical and coercive.

There have always been market exchanges of a sort, whenever strangers found it beneficial to trade. For example, in the Bronze Age Mediterranean, only a few locations had mines for the important metals (copper and tin). Phoenecian sailors served as intermediaries who carried metals to locations that had none. Their usual trading practice among the Greeks was to approach a beach, land some of their cargo, and then withdraw. The local Greeks, learning of this, would approach the landed cargo, examine it, and then make an adjacent pile of items they thought would be attractive to the Phoencians. When the Greeks had withdrawn, the Phoenicians would land, and make some adjustments to the two piles. After several rounds of this, agreement would be reached, the Phoenicians would take the proffered goods and depart.

Why this long and cumbersome exchange process? Because an even easier way to obtain what one wants from a stranger is to rob him. Markets have always had the problem of ensuring the fairness of transactions–the problem of avoiding force and fraud. Over the centuries, this problem has been gradually mitigated through the power of the state and changes in social norms.

3.1 Money

Another problem faced by early markets was the difficulty of barter, a problem solved by the invention of money, in Lydia, a state in Asia Minor, around 600 B.C.E. Money serves two important functions: as a measure of value; and as a medium of exchange.

Let’s look at the measure of value function. Imagine a village market in Lydia where exchange is conducted by barter. In the diagram below, the market contains seven goods. A “price” in a barter market is the number of one good it would take to buy another good. So, for a donkey there are six prices: the number of goats it takes to buy a donkey; the number of fish it takes to buy a donkey; the amount of wine…, etc. There are seven goods in the market, each with six prices, making 42 prices measuring the value of these goods.

Ancient Lydian barter market with 7 goods and 42 prices

Mathematically, the number of prices in a market with n goods is the permutation of n things taken two at a time \(=n!/(n-2)!\). So, if a market has 30 different goods, the number of prices would be \(30!/28!=30*29=870\). Human cognitive limits render it difficult for markets to function well with more than 500 or so prices. But by picking out one particular commodity, and using it as the sole measure of value, the number of prices can be dramatically reduced.

Ancient Lydian market employing money, with 7 goods and 7 prices

Imagine that a good called Gold is selected as the sole measure of value. Now, with seven goods, there are seven prices. With 30 goods, there would be 30 prices. Money’s function as a measure of value vastly simplifies the amount of information a participant needs to transact in a market.

The second function of money is as a medium of exchange; that is, in any exchange, the seller will accept money in exchange for the good, so that exchanges are not direct exchange of good for good, but of good for money.

Imagine that you arrive at the ancient Lydian market with a donkey, which you hope to exchange for some clay pots. You find the clay pot seller, but she is not interested in your donkey, she wants fish. This is the double coincidence of wants problem that bedevils barter exchange. What can you do? Well, you try to convert your donkey into fish, but the fish seller doesn’t want your donkey, she wants wine. The wine seller also turns up her nose at your donkey but would like some olive oil, but the olive oil vendor wants only a goat. And then, finally, you find a goat seller who wants your donkey, you hurriedly exchange the donkey for the goat, the goat for olive oil, the olive oil for wine, the wine for fish, and when you breathlessly arrive at the pot seller you discover – that she’s already gotten her fish! By then, dusk is falling, the market is winding down, and you wend your way sadly home, staggering under a load of less-than-fresh fish.

Ancient Lydian barter market with difficulty matching buyer and seller

How much easier it would be if everyone simply accepted Gold in exchange for their products. You arrive at the market and sell your donkey to the goat vendor in exchange for Gold, and then go the the pot seller and exchange your Gold for the clay pots. Money, as a medium of exchange, solves the double coincidence of wants problem.

3.1.1 Characteristics of Money

In the Iliad, Homer occassionally uses cattle as a measure of value, as when he speaks of the prize in a wrestling contest as worth 12 oxen. In Anglo-Saxon England, blood money paid to a murder victim’s family by the murderer took the form of cattle. And in East Africa, up until recent times, a young man wishing to take a wife would pay a bride price in cattle to the father of his intended. In all of these cultures, cattle took on at least one of the functions of money. But why then do we not use cattle as money today?

The ideal commodity to use as money should have these characteristics:

  • It should be a commodity. That is, it should be standardized, so that one unit of money is very nearly identical to another. Cattle vary in height, weight, health, etc.
  • It should be portable. Money should be easy to transport. Cattle ambulate, so they are fairly portable, and do OK on this criterion.
  • It should be durable. Something that rots quickly, like dead fish, would never do as money. Cattle can live a few years, but they must be fed and watered, and are not ideal on this criterion.
  • It should be divisible. How does one make small change in cattle?
  • It should be scarce. Sand, for example, meets all of the above criteria, but it wouldn’t work as money–anyone can go out and dig up a bagful whenever they please.

Ethnographically, there have been a number of commodities that did fairly well on these criteria. In pre-Colombian Mesoamerica, dried cacao beans worked well. They were small, light-weight, did not spoil, and grew only in the lowlands, so were scarce in the highlands. In prisoner of war camps during World War II, cigarettes filled the function of money: standardized, small and light, non-perishable, scarce, and desired by everyone.

This last criterion (that the commodity serving as money is desired by everyone) turns out to be especially important for the emergence of money where it previously had not existed. Prisoners of war would receive many items in their Red Cross packages, from hand-knitted sweaters to books to cigarettes. Individual taste would vary on these items, but everyone wanted cigarettes. Thus, a prisoner seeking to trade his hand-knitted sweater for a book would exchange it for cigarettes, knowing that whoever had his desired book would gladly exchange the book for cigarettes. Money emerges in this way not because a single person deliberately invented it, but spontaneously, through the similar actions of many independent people. In economics we call this kind of process a spontaneous order, and we think that much of our social order emerges in this way.

Gold and silver (collectively known as “specie”) satisfy all of the requirements for a commodity suitable for money. The rulers of Lydia, in 600 B.C.E., issued lumps of specie, with the royal seal impressed upon it to certify that the metal was electrum (an alloy of gold and silver), of a standardized weight and composition.

The first coin, front and back, from Lydia 600 B.C.E.; electrum impressed with the Lydian lion

Money super-charges markets, making them a very efficient means of exchange among humans.

4 Our economy today

All three exchange modes exist in our economy. Markets are primarily responsible for integrating modern economies, and we will have a lot more to say about them as the course goes on. If you are curious about what I mean by “integrating” economies, you might like to take a look at the I, Pencil video (6.33 minutes; from YouTube):

Redistribution characterizes the activities of government: the state demands from us our taxes, and in return gives us services like law enforcement, national defense, public education, etc. But redistribution also characterizes the internal structure of a firm: the employee follows the commands of management, and then receives a paycheck.

Reciprocity exists in small groups, most conspicuously in intentional communities, such as the Hutterite communities, but the real importance of reciprocity is that it characterizes the economy of the small groups that make up the building blocks of our society: families.

In a traditional family, there is a sexual division of labor defining the contribution of its members. For example, in my own family of orientation, my father went out and brought home a paycheck. He also was responsible for yard work, and for mechanical work inside the house, such as repairing the furnace. My mother took care of the kids, cooked, cleaned, and did the shopping.

Today, many fewer families follow the traditional sexual division of labor – women, like men, often go out and bring home a paycheck. This has left a lacuna in the tasks traditionally done by women. How is the cooking, cleaning, childcare, etc. now accomplished? There are three main ways (note that none of these three involve government).

  • First, males now share some of the traditional female tasks. Economists have researched the sharing of tasks using the interesting American Time Use Survey dataset. They have found that sharing is far from equal: women still do the majority of the traditional female tasks.
  • Second, technology has evolved so as to make women’s work easier: the vacuum cleaner, the dishwasher, the microwave oven, video monitoring of children, online shopping, etc. The economic theory of induced innovation maintains that technology evolves so as to use less of the scarcest resources. In this case, women’s labor time has become scarce, and technology has evolved so that less of it is needed.
  • Third, and most importantly, market exchanges are used by families to replace tasks previously done in the household: day-care centers perform child care; maid services perform cleaning; firms sell convenience foods that already have most of the cooking work done.

For tasks within the American family, markets are replacing reciprocity. This is an enormous social change, taking place not because anyone planned it, but as a developing spontanous order.

Compiled on 2023-04-19 by E. Anthon Eff
Jones College of Business, Middle Tennessee State University

  1. Birdsell, J., Bennett, J., Bicchieri, M., Claessen, H., Gropper, R., Hart, C., . . . Thompson, L. (1973). A Basic Demographic Unit [and Comments and Reply]. Current Anthropology, 14(4), 337-356. www.jstor.org/stable/2740838↩︎