Sunday, 25 May 2014

The Problem with Monetary Exchange

I have seen it said that recessions are a feature of a monetary economy - that they cannot arise under barter - but I haven't really seen this explained anywhere very well.  Part of the reason is that money is a non-produced asset, so that if people try to acquire more money, it doesn't lead to more production.  But land is also a non-produced asset, so this is not the whole explanation.  No doubt someone has set out a good explanation somewhere, but as I haven't come across it myself, I thought I'd have a go. 

So, I'm going to look at a simple economy with a farmer, who grows corn, and a fisherman who catches fish.   In addition to corn and fish, the only other good is money.  The exchanges prices between each good are fixed.

The table shows the order of the marginal preferences of the farmer and the fisherman given the quantities that can be exchanged at the fixed prices.  So the farmer would prefer to have more money than anything else.  But he would also prefer more fish to corn.  The fisherman would also prefer more money to anything else.  But he would prefer corn to fish.


Both farmer and fisherman would prefer more money, but if the supply of money is fixed, there is no exchange that can achieve this.  However, they can still both make themselves better off if they do an exchange between corn and fish.  Both receive something they value more than the thing they give away.

A key feature of a monetary economy over a barter one is that all transactions take the form of goods for money, rather than goods for goods.  So here we can have trades of money for fish and money for corn, but we cannot have the trade of corn for fish.  So achieving the same end result that we had before now requires two separate transactions:

1) Farmer pays fisherman money and receives fish

2) Fisherman pays farmer money and receives corn.

If they can co-ordinate these two transactions, the fisherman and the farmer can get to where they were before.  If they are undertaken separately however there is a problem.  Transaction (1) is fine for the fisherman - he prefers money to fish, anyway.  But it is not OK for the farmer - he prefers to have the money.  If he does this transaction and then transaction (2) does not happen, he is worse off.

The same applies with transaction (2) but the other way around.  The farmer is happy with this transaction, but the fisherman is not.

So although, if both transactions happen, both parties are better off than before, if the transactions have to take place separately then in fact neither will happen.

The key things that are making this happen here are: a) that all exchanges are of the form goods for money, and b) that both parties have a preference for money over either good.  The exclusion of the goods for goods exchange is why this problem is specifically a feature of a monetary economy.


  1. I've wondered whether time is a crucial aspect of that economic glitch in the role of money. Most of the potential value "out there" is potential machine time, potential human labour, potential use of land etc. Imagine if two economies were being tried out. In one economy, everyone stayed idle for ten years and no one moved into the houses or switched the machines on or planted the crops -it would be rubbish. In the other economy full use was made of everyone and everything from the outset -that would be prosperity. And yet in an economy where some people have the money -it is the other people, who don't have the money, who suffer the consequences of the unused economic potential. Money allows value to be stored without the need to ensure that there are the real means to honor those paper claims.

    Basically money offers the individual the ability to personally store up time and transport it into the future fresh and ready to use. Obviously it can't really do that -all it really does is to shift the burden of the consequences of idleness from some people onto other people. It allows those holding money to (inadvertently) enforce unemployment onto other people and then make those unemployed people suffer the consequences of the reduced production.

    I guess land could get used as money in the sense talked about here. If people exchanged land for fish or corn, then choosing to store value rather than to ensure continued activity could likewise lead to a slump.

    I'm no fan of excessive work and consumption just for the hell of it. But the tragedy is to have stuff left undone that everyone wants to do and wants to have done simply due to a shortage of funding. As I see it the way out might be to ensure that the financial system reflects the reality that every unused hour is permanently lost and the way to do that is to replace current taxes with an asset tax.

    1. Perhaps the crucial glitch is that services can be exchanged for a durable store of value.

      Services have a time aspect to them such that if they are not used, that potential is lost forever. As such they are not really comparable to durable stores of value. And yet we exchange one for the other as if they were. The problem comes when we amass stores of value as a way to lay claim to the accumulated prosperity that has come from the deployment of services over the intervening period up until the point where we choose to spend our savings. The more claims we have stored, the less real prosperity there is to meet those claims.

      I suppose the classic idea is that by saving money, we leave resources free to be deployed in production of capital goods and research etc and so ensure that enough investment has happened to honor those monetary claims. But in reality, more investment happens when there is more demand. It is when the system is pushed to bursting that innovation happens ensuring continued capacity to meet future demand.

      The big difference is between use of natural resources and use of human resources. I don't think we take on board just how opposite they are. The EU insists that land needs to be kept ready for agricultural use BUT the best way to preserve soil is to have wilderness. The best way to preserve fishing grounds is to have exclusion zones. But the best way to preserve human potential for the future is to have a desperate shortage of man power such that everyone learns to work efficiently and design labor saving machines. Innovation comes from too much demand. Austerity leads to atrophy.

    2. Land could certainly be money here. There's nothing in what I've said here that would not apply if land were used as money.

      And I also agree that the existence of a durable store of value is important.

      I'm not sure I understand what you're saying about storing time though.

    3. Nick, I think the "storing time" aspect might be the crux of it and I wish I was less incoherent about what I was getting at.

      When we have savings, what we are really saving up is a claim to command the use of some resources for some hours. Typically it is not that a tree gets left standing so that we can cut it down in fifty years time when we need the wood. Rather we don't consume our monetary income now so that in fifty years time we can pay for 500 hours of nursing time or whatever. The problem is that "nursing hours" can't be stored up. If we don't spend now, there is no more capacity to provide "nursing hours" in the future than if we had already spent all the money. Likewise with "machine hours" or any other resource use that entails making use of a capability for a period of time.

      Net monetary saving can lead to a macroeconomic mess because it induces under-utilization of current real resources in the forlorn attempt to squirrel them away for future use. The financial claims can be effectively squirreled away BUT the "nursing hours" and other such meaningful real world means to honor those financial claims can not be squirreled away -they are instantly perishable -use them or lose them.

    4. I think you're just pointing to the desire to save and the fact that it can lead to under-utilisation of resources currently. If the only way to save was by creating new durable goods, then in order to save people would have to produce those goods in the first place. This is the usual story of barter - that saving involves producing goods then not consuming them. But having non-produced durables changes that.

      But, certainly, having people want to save more - to under-consume now - is an importat feature of what causes recession.

  2. "I have seen it said that recessions are a feature of a monetary economy - that they cannot arise under barter - but I haven't really seen this explained anywhere very well."

    This is excellent. I've had the same question in back of mind for 5 years but never explored it or asked somebody for an answer to it.

    I think maybe you've explained in your own way why recessions can happen in a monetary economy.

    But I'm not sure you've explained why they can't happen in a barter economy.

    I remain deeply suspicious of the general claim that recessions can't happen in a barter economy. And like you I haven't seen it explained to my satisfaction anywhere else.

    It has the feeling of something that should be fundamental to the logic of economics and therefore important. So it remains disturbing to me.

    1. Thank you. I don't think is a complete story, but I do think it is part of it (or maybe just a way of looking at part of it).

      I find it quite difficult to pin down what a recession might mean in a barter economy, partly because I'm not sure of what counts as a barter economy.

      I made things easy for myself here with just two agents, so avoiding any potential lack of double co-incidence of wants.

      If we have more agents, but restrict ourselves to bilateral trades, things look a little different. For example, we might have three agents: A, B, C where A wants B's product, B wants C's product and C wants A's product. They can all trade to their benefit, but it has to involve something like B exchanging his commodity for A's, then exchanging A's commodity for C's, for example. This can still be consistent with each trade being advantageous to both parties.

      But are we sure this is still a barter economy? Why do we say that A's commodity is not functioning as money here? The problem I have had here is in constructing a scenario that can clearly have a recession, without having features that might be thought of as being monetary.

    2. There is a very mundaine, if most likely far too simplistic answer. Namely that recessions are defined as monetary phenomena. So any non-monetary economy (is that an oxymoron?) can by definition not experience a recession. Same goes for unemployment. I'm not sure it's very a meangniful distinction, in fact I'm pretty sure it isn't. But it does lead one to ask, as you have done, what it is exactly that signifies a barter economy. Have they ever existed in the strictest sense of there being no commodity with monetary character? And if you believe so, define money :-).

    3. It may be right that it has to be defined as a monetary phenomena. I tie myself in knots trying to think of what this stuff means in barter terms.

    4. I personally find the case of a monetary econmy harder to model in my head because there is one variable (money) more involved.

      If we take the term recession to mean that there is less value added in real (inflation adjusted) terms in period 2 than in period 1, we have to ask:

      Is this due to some shift, say of preferences, vs. the monetary side or vs.the real (tangible stuff, including land) side of things? Or any combination thereof.

      In the barter case, the first cannot happen because there is no money thing. This is what I take the claim 'there cannot be a recession in a barter economy' to mean.

      But in the second case this would be paramount to the claim that in an economy without a money thing, there is a natural law that prohibits the volume of exchange of goods to decline from one period to the next. To me, that seems patently absurd.

    5. I think a recession here means a general excess supply, i.e. there is an excess supply of all goods, rather than there simply being an excess supply of some and an excess demand for others. So everyone wants to sell more of what they have, but can't.

    6. Did I imply a partial excess? I certainly didn't mean to. That is, if you do not take money to be a good.

      Wiki says: In economics, a recession is a business cycle contraction. It is a general slowdown in economic activity.

      I personally take this to mean a contraction of real activity. Less goods sold at the same price. But it is of course measured in monetary terms.

      The excess supply story, on the other hand, despite the colloquial understanding of the term supply, tells me nothing about the real side of things. It implies pressure on prices to fall. Which could theoretically be concomitant with a decline, no change or even a rise in real activity, no?

    7. I think I'm using the term here with a slightly differenmt meaning, because I'm wanting to focus on why exchanges might not be happening even though everyone wants them to happen. That is not the same of course as a drop in the level of activity, which could happen for example because everyone decided to take more leisure time. So I agree in that sense. I'm not sure that supply is all about prices though.

  3. Your model has two decision makers (the fisherman and farmer) and three kinds of property (fish, corn and money). Money, being property, must be made or collected by someone.

    Now to the question of recessions in a barter economy, your blog brought me to thinking that money is tool to improve the transfer of property between people. In a barter economy, we might consider that we have a permanent recession with an EXTREME lack of money!

    "The exclusion of the goods for goods exchange is why this problem is specifically a feature of a monetary economy." is an exclusion that never really occurs. Instead, a transaction occurs between farmer and fisherman that is valued in money, with actual money never part of the transaction. Both the farmer and fisherman are far too practical to let the fact that there-is-no-money-available prevent a mutually beneficial exchange.

    1. "Money, being property, must be made or collected by someone."

      I didn't go into this in the post in order to keep it simple. But we can easily come up with a story for the existence of money. We could imagine a government that occasionally made purchases (of fish or corn) using money and occasionally imposed taxes. Then, if we want, we can assume it was doing neither in the current period.

  4. Thanks, very interesting and I think important!

    It looks to me that here the exchange takes the place when both goods for money transactions can be made simultaneously. But isn't that exactly the same than not making the barter exclusion? So money cannot improve the trade in this setup but neither does it prevent it?

    1. As I only have two agents here, it is easy to see how they might get together and agree to do simultaneous money for goods transactions, thereby getting around the assumed restricton on goods for goods exchanges. If we were to add more agents so that we did not in general have double coincidence of wants, then this would be much more difficult. In a barter economy, this would involve people trading in steps towards their preferred product as described in my reply to JKH above.

  5. Nick,

    The barter economy does not need to be limited to person to person exchanges. Here is how a recession can happen with a barter economy.

    Your fisherman and farmer may be far separated. This allows for a third party who likes both fish and corn. How can this third party gain the trust of the fisherman and farmer?

    Assume that the third party (we will call him a "trader") decides he can take temporary control of the fish and corn, move it across the distance separating the fisherman and farmer, and have a little of both corn and fish to enjoy himself. The problem the trader must solve is how to gain the trust of both fisherman and farmer and convince them to give him control of the products for long enough to make the distant transfer.

    Here is how the trader solves the problem. The trader goes to both and pitches the plan. The trader will make the exchange happen and transfer the products, keeping only a little of each to cover his "expenses". This plan will work if both farmer and fisherman trust the trader, both accept his written promise of performance, or if the trader simply uses "money" to purchase the products.

    Both the farmer and fisherman will see the trader as absorbing some of the benefits of their work. Each will need to decide if the cost is worth the gain received.

    This model carries the suggestion that the use of money is every bit as much in the interest of third parties as it is for the producers of products.

    A recession will happen if the producers begin to distrust the trader, his promises, or his money.

    1. Roger,

      Yes, once we start to take into account things like spacial seperation then it starts to complicate things in a barter economy. We can incorporate various ways to deal with this but it often then becomes difficult to be clear that we are not really assuming some form of monetary exchange, in that some commodity or agreement is acting as a medium of exchange. I don't think there's any solution to this really - it's just that pure barter is a bit of a theoretical concept.