Non-Adiabatic Economics

Most economic theories assume that the system under consideration is stable or is, at least, tending toward a stable state. I'm going to make an analogy with thermodynamics.

The situation I have in mind is one of an ideal gas enclosed in a vessel. The law that we are concerned with is the gas law which states that pressure time volume equals a constant for a given temperature, i.e PV = C. In the ideal situation (say a gas under pressure from a piston) as we push down on the piston the volume will decrease and the pressure will increase. In the ideal situation this is called an adiabatic process. In the real world, however there is a chance that the temperature will change as the vessel conducts heat from the gas to the surroundings through the walls of the container. This is a non-adiabatic process.

I consider classical economics as equivalent to the simplifying assumptions of the adiabatic gas law. Take a simple example: demand vs price. The classic description is that as demand increases the price will rise which will cause a slackening in demand until a new equilibrium is reached. But in the real world when demand increases, sometimes production increases and the price doesn't increase. This is the non-adiabatic case.

Another common example has to do with competition. The basic understanding is that in a competitive marketplace the price of the item will fall until the marginal cost of producing the next copy of the item will equal the price. In other words none of the firms will be making a profit. It's true that there are sometimes limited price wars of this nature, but in many cases the behavior is not as predicted by theory. Sometimes firms sell at below cost to drive weaker firms out of the market, sometimes they under produce to prevent the cost from dropping too low, sometimes they attempt to create product differentiation perception through advertising so that demand is not uniform. Most common these days is for firms to tacitly agree to divide a market between them using various signaling mechanisms (since outright collusion is supposed to be illegal). A common example is when an airline declares a fare increase in advance of the effective date so that their competitors are alerted. If the competitors refuse to raise the price then the original airline will revoke the increase.

Still another example of non-adiabatic economics has to do with under investment by a firm. Classical theory says a firm should plan to expand its manufacturing capacity when it sees a growing market, but this doesn't always happen. The US steel industry under invested for decades and reaped high profits during the period. When newer firms entered the market they could no longer compete, their infrastructure was obsolete and the profits had all been extracted from the firms by the owners and managers. The firms collapsed. Why did these firms not behave in the way classical theory would expect? The explanation is that the objectives of those running the firms differed from those expected. They were not interested in the long-term health of the firm, they just wanted to make their bundle and get out. This was a non-adiabatic condition. Classical theory assumes that the managers of the firm are interested in its long-term prospects. This turns out not to be true in many cases.

The situation has become more glaring over the past half century or so as the separation between the owners of firms and the managers has widened. Managers have only a nominal investment in a firm and know that they will have a relatively short tenure. Traditional owners (say the Ford family for Ford Motors) are becoming increasingly rare. Most "owners" are now stockholders who tend not to have any relationship with the firm, but regard the stock itself as the investment. Performance of the firm is only of concern in as much as it influences the price of the shares over the period that the investor is involved. The only stakeholders who may be interested in the sustained health of a company are the employees, and they have no control in most cases. The classical model no longer applies.

Returning to the idea that markets tend to stability, we see this violated all the time. Selling at a loss is an attempt to permanently change a market by driving out competitors. Only in shared monopolies do firms try to preserve stability, otherwise they are always trying to effect permanent changes in the market. A firm may deliberately over invest in production in order to cash in on a fad, knowing that the demand can't be sustained over the long term. This is especially true with consumer electronics, one year it is CB radios, then it is cell phones, the year after GPS units or electronic picture frames. If the owners of these firms are not complete novices they must know that the game is to make as much profit as quickly as possible, pick up your marbles and leave the game. The excess investment in equipment and staff is somebody else's problem.

Of course economic theory has explanations for all these types of disturbances, but they regard them as exceptions rather than as a series of non-adiabatic events that never correspond to the ideal case.

The gas law is physics, it deals in the ideal case. The action in an automotive cylinder is engineering, it must always deal with non-adiabatic conditions. Economics needs to move away from the physics model and towards the engineering framework. It's messy, much harder to calculate, but if you want your gadget to work in the real world you had better understand where the ideal is inadequate.

Moral: Ideal theories make for poor policy prescriptions.

Click here to see all my essays in context.

If you have any comments or for further discussions email me at
Copyright © 2008 Robert D Feinman
Feel free to use the ideas, but the words are mine.