Supply and Demand: The Problem With “Them Curves”

“Supply and Demand” has ruined “economics” with bad concepts. We must leave aside the math until we fix our philosophy. Then we can use accounting and other math effectively.

The supply and demand model works by determining a market price level from two functions of prices:

• y = supply(price)
• y = demand(price)

These curves quantitatively answer the following questions:

• Supply: “Given a price level, how much product will be offered?”
• Demand: “Given a price level, how much product will be consumed?”

It is important to note, that economists plot these curves backwards of how mathematical functions are normally plotted. In their convention, the function’s input(price level) is the vertical axis, while the function’s output(supply or demand) is the horizontal axis. If you ever see me plot these curves, I will ignore the economist’s convention, and simply plot price along the horizontal axis, because that’s how I am used to plotting mathematical functions.

One possible reason for this unusual convention, is that the inputs and outputs of the functions are different from the information required by the model and the answer it provides. The model requires these mathematical functions as prerequisite information, and determines the equilibrium market price, whereas the functions describe the behavior of buyers and sellers as output responding to a price input.

These two curves determine how marginal behaviors lead to an equilibrium price at their intersection. For example, sellers might follow this algorithm:

• if(demand > supply) then raise prices;
• if(demand < supply) then lower prices;

These marginal descriptions assume the following:

• Both functions are monotonic.
• Supply and price are positively correlated.
• Demand and price are negatively correlated.

These are reasonable statements. The problem is not their accuracy, but rather that they are the wrong starting point for describing our social market behaviors.

The “Supply and Demand” model depends on complex emergent social concepts like “price”, “product”, and even though it is unstated: “money”. Supply and Demand uses these complex abstractions to try describe real world relationships and behaviors, instead of using our relationships and behaviors to accurately describe these abstractions.

This makes the model inflexible and misleading. If we don’t understand where our concepts come from, any unusual relationships can invalidate them, and we won’t know how to assess outcomes in these unusual conditions.

The supply and demand model has biased the economics profession. It is put forth as the norm, and deviations are described as “special cases”. Terms like “sticky prices”, “menu costs”, “luxury goods”, and more, all reveal this bias.

The Incredibly Complex Assumptions of Supply and Demand

If you owned a lemonade stand, you might try lowering prices, and seeing how customers respond. Then you raise prices, and see what happens.

From the perspective of an individual business owner, it is easy to see where this model came from. Business owners can use experience and feedback to model a “Demand Curve” of customer responses to their prices.

But I propose to you that supply and demand curves are determined by deviations from market price, not vice verse.

It is surprising the specific behavioral information the supply and demand model requires before you can start, and only gives us the market price equilibrium as output.

Before you can tell me the market price, you must tell me the response of buyers to every possible price level, as well as the response of sellers to every possible price level. That is the first clue that something is strange here. (In the real world, the market price can simply be observed directly, while supply and demand curves require complex empirical research that may not be valid in future scenarios.)

The breakthrough comes when you look at the units of price, and ask what it is and what it actually represents.

Money is the unit of price. Money is no simple physical thing or measurable physical quantity, but rather a symbol created by social peers.

These social peers use this symbol to make decisions together and communicate with one another.

This is where chartalism comes in. Chartalism describes money as a political and legal construct for organizing the actions of social peers.

I am not of the persuasion that money has to work a certain way or represent a certain thing. It is only an informational product we use to make decisions. I am open to arguments that certain designs of money are more effective and more sustainable, and thus peer groups who use such organization will be more effective at survival. The disfunction of the modern financial world clearly shows us that what money represents is arbitrary, even though a social context needs to exist to create these representations.

This is why I think we should use money to conduct accounting. Accounting allows us to accurately assess the state of the world as well as history and make decisions as tiers of social entities interacting. A nation can using accounting to decide how to allocate resources and so can an individual. A corporation can do this or a city.

One of the most harmful fictions of the modern world is that we are just individuals. This is about as useful as saying we are all protons, neutrons and electrons. We needs lots of different abstractions of physical reality at different resolutions to assess the what happens in the physical world most accurately.

I don’t think accounting makes any intrinsic statements about who owns what or reciprocity. Again, I am open to arguments that reciprocity is a positive social convention, but the important part of accounting is not justice, but rather measurement. Only by effectively making accurate measurements can we effect our notion of justice accurately.

“Supply and Demand” assumes “The Market”, “Costs and Priorities” describes the relationships that give rise to markets and other social organization.

“Supply and Demand” requires that social peers follow a preset narrative for interacting. “Costs and Priorities” recognizes these peers can conform their behaviors to any narratives they choose.

I’m not offering new mathematical tools here. Our accounting should be accurate and representative, our decision making should be deliberate, and our relationships as entities should be explicit. Once you describe these things, the mathematics to use in each scenario becomes clear.

Accounting is simply addition and subtraction. If you want to design more elaborate resource programs you can use Linear Programming or other mathematical optimization tools. If you want to describe outcomes of our interactions as social peers, you can create agent models where the agents follow whatever behavioral narrative you are trying to promote. If you want to track quantities at a systems level without specifying the details of the underlying relationships, you can use tools from feedback and control theory and dynamical systems. There’s a rich ecosystem of mathematics that can be applied if we can philosophically recognize what we are trying to represent.

Just because this a highly conceptual and philosophical foundation, doesn’t mean we can’t get hard and firm with the mathematics if we need to. The tools are there, and by describing our social existence clearly, we can find the right tools to use.

Quantitative models have their place, and agent models as well. Quantitative models depend on abstracting discrete entities and interactions into continuous measurements. Agent models depend on effectively abstracting and computationally representing complex interacting entities.