Thursday, 21 January 2010

No unemployment in barter economies?

There is a widespread view that unemployment cannot occur in barter economies. E.g. see the quote from Prof Paul Davidson at the top of this site. Plus see here. And see here (12th para – starting “MMT tells us....”.)

Barter economies certainly cannot suffer from the “paradox of thrift” cause of unemployment. But they could still experience unemployment if the relative price of different products was not flexible enough. Here is whey.

Consider what Robinson Crusoe does on arriving on a desert island where there are about five families and a barter economy. And let’s assume just three products: fish to eat, grass skirts to wear and mud huts to live in.

Crusoe is good at fishing, so he goes fishing and offers fish in exchange for a hut and a skirt. The additional supply of fish provided by Crusoe won’t be a problem as long as the price of fish falls in terms of huts and skirts. However if there is NOT price flexibility, Crusoe will not be able to sell his fish. He will be unemployed (or someone else who specialises in fishing will be unemployed).

Exactly the same problem occurs in money using economies: wages are “sticky downwards” as Keynes pointed out. That is, given an excess supply of a particular type of labour, the price of that type of labour will not fall, or at least will fall far too slowly to avoid unemployment for some members of the type of labour concerned. (Arguably Keynes was not referring in his “sticky downwards” point to SPECIFIC types of labour, but rather to labour in the aggregate, but never mind.)

The relative price of different types of labour is probably MORE flexible in barter economies than in money economies, but it is unlikely that prices are PERFECTLY flexible in barter economies.

To summarise, on purely theoretical grounds, barter economies seem to have significant advantages over money economies when it comes to unemployment. But unemployment is not impossible in barter economies.

Afterthought (25th Jan):

Thanks to an idea put by Tim Hickey (see below), I would now like to state the “Musgrave/Hickey General Theory of Unemployment In Barter Economies”, which is thus.

1. In ultra simple economies, that is where there is no specialisation or “division of labour” as economists call it, unemployment is almost impossible. The only possible cause would be a severe shortage of natural resources. E.g. in a drought people might not be able to grow food.

2. Given division of labour, a possible cause of unemployment arises, namely insufficient flexibility in the relative price of different types of labour and products.

3. When an economy gets even more “sophisticated” and introduces money, another possible cause of unemployment arises: the well known “paradox of thrift”.


  1. I believe that Say's law is not stated correctly at the Bizcovering link. It should read, "whatever is produced [for which there is adequate demand] in the barter economy is sold out.” Say recognized that not everything is salable at a price that makes it worthwhile to produce, and that which is not salable will cease to be produced.

    Entrepreneuring works in barter economies, too, as anyone who participates in a barter club knows. Some products and services just aren't worth trying to trade for what one would take because no one wants them. You either have take less than you want, or give up and figure something else out, or else you are are not gainfully employed in that environment. Entrepreneurs reduce this risk by responding to felt needs and innovation instead of entirely new products unless they have deep pockets to absorb risk, in a barter economy, a lot of leisure.

    However, I think that the real issue is overlooked when economies are modeled on barter economies with money added as a medium of exchange that doesn't really do anything in the economy but facilitate transactions. This is to equate macro with micro aggregated, which it is not because of the role of sovereign governments as monopoly providers of currency. Then, this becomes the foundation of the economy.

    Moreover, barter economies were necessarily simple and individuals could fend for themselves. This is no longer true in a modern economies, where most employment involves taking a role in a complex system in which individuals cannot support themselves and their families independently of the system. This is the basis of modern macro, where the government is the employer of last resort to maintain the stability of the society. On a south seas island with a barter economy, what is the difference between being "unemployment" and subsisting at leisure on nature's produce available in the commons. That's not an option in a modern society.

  2. Ralph, I was not able to find precisely what Bill Mitchell means at the link above. He simply says, "MMT tells us that state money introduces the possibility of unemployment. There is no unemployment in non-monetary economies." I'm sure he has explained this more fully somewhere else since unemployment is this thing, but I don't know where specifically.

    I suspect that what he means is that there is no involuntary unemployment in a barter society unlike in a monetary one. According to neoliberal principles there is no involuntary employment in either barter systems or monetary systems, since they are comparable.

    A good question to ask Bill.

  3. Tom: There seems to be some argument as to what Say’s law consists of. The conventional definition is “supply creates its own demand”. But some claim that Say’s law is essentially the idea that products have value because of the value placed on them by consumers, and not (as used to be thought) because of the cost of producing products.

    Your second last sentence poses a nice question. My answer is that in an ultra simple economy where anyone can produce anything (food, clothing, shelter, etc) unemployment is impossible. Or the word becomes meaningless – which is what you are saying, I think.

    But once people start specialising, the problems start. In particular, in my Robinson Crusoe example, if the price of different products in terms of each other is not flexible enough, unemployment can arise.

    Re Bill Mitchell, I assume he is just trotting out the conventional idea, which I quite agree with, namely that introducing money brings yet further problems (the big one being, as I mention above, the paradox of thrift). But Bill is wrong to suggest that this idea is one that the world owes to modern monetary theory. The paradox of thrift idea was around before MMT. I’ve come across several other instances of Bill making this mistake: attributing ideas to MMT which are not unique to MMT.

    I agree with much of MMT, but definitely disagree with some of it. The main thorn in the flesh of what you might call the “MMT brigade” (Wray, Mosler and Mitchell) is a British Prof of economics, Malcolm Sawyer. See Sawyer’s paper at this JSTOR site: I find JSTOR an irritating system, but if you fiddle around with it long enough, you might be able to down load the paper. Alternatively try this Questia site:;jsessionid=LZpGmRXswn2nJvj1V8pktTtGQYkh9GhLBQT3MBjRh0ZpZJdypGZj!610724718!-2002178534?docId=5002080292

  4. Thanks for the links to Sawyer's article, Ralph. They didn't work for me, but I was able to download a copy gratis at

    I'll take a look at it.

    Yes, I was saying that voluntary unemployment is largely meaningless. A person who is "voluntarily unemployed" is either at leisure or is otherwise occupied, e.g., as a student.

    "Voluntary unemployment" does have some meaning in a modern complex economy, where it is called frictional unemployment. But I would not regard this as actual unemployment, and frictional unemployment really should not be figured along with cyclical and structural unemployment in the same way. An economy is properly said to be at full employment when frictional unemployment is not considered because it is pretty much a constant.

    Regarding Say, my understanding is not that he claimed, "build it and they will come," but that what is for sale is economically determinative rather than demand, in that supply of a good creates demand for it (although this doesn't imply that everything produced will find a demand) and not vice versa. There has to be a market before there are shoppers. Thus, the macro debate is between followers of Say and followers of Keynes. I would say the point is moot practically speaking as innovation shows. A hot new product creates a new market, it is true, but no one invests time, energy and money in bringing a new product to market without determining that there is a demand for it. However, it is often difficult to determine demand and many good new products fail at first because of the timing. For example, Apple runs on innovation but it miscalculated demand for a couple of rollouts, like the Palm and Cube, the concepts of which were premature and similar products prospered later on.

  5. Tom: I don’t agree with your second para. “Voluntary unemployment” is not “called frictional unemployment”. The former phrase is normally used, as the phrase suggests, to describe “doing nothing through choice”, or “inactivity caused by demanding an unrealistically high wage”.

    The phrase frictional unemployment is normally used to describe unemployment caused by short term mismatches between supply and demand for particular skills, in particular areas. The phrase structural unemployment normally refers to longer term mismatches, e.g. persistent high unemployment in particular areas. Sawyer is one of the many people to use the word structural in this way, if memory serves.

  6. Ralph: "“Voluntary unemployment” is not “called frictional unemployment”. The former phrase is normally used, as the phrase suggests, to describe “doing nothing through choice”, or “inactivity caused by demanding an unrealistically high wage”."

    I am suggesting this is normative not descriptive.

  7. Most if not all references to Say's Law are quoting James Mill rather than J.B Say himself.

    Were it not for James Mill I doubt we would have ever heard of it.

    Most people are confused moving from barter to a money economy for one reason - Uncertainty.

    That is why our good friend Mr Walras introduced the "auctioneer to the tattonement process. The Auctioneer by making all trades occur also made all uncertainty disappear.

    Pretty simply really.


Post a comment.