Saturday, November 23, 2013

Marginalism versus Post Keynesianism on Mark-up Pricing

One important response to the theory of full cost and mark-up pricing and the threat it poses to neoclassical economics is to argue that the actual result of mark-up pricing – on the basis of average unit costs and profit mark-up – is the same as if a firm deliberately and consciously were to equate marginal cost and marginal revenue (Keen 2011: 168, citing Langlois 1989).

This type of argument is made, for example, by Langlois:
Langlois, Catherine. 1989. “Markup Pricing versus Marginalism: A Controversy Revisited,” Journal of Post Keynesian Economics 12.1: 127–151
This article provoked a symposium and a number of replies in the Journal of Post Keynesian Economics Winter 1990–1991 issue:
Symposium: The Marginalist Controversy and Post Keynesian Price Theory

Frederic S. Lee. 1990–1991. “Introduction,” Journal of Post Keynesian Economics 13.2: 233–235.

Mongin, Philippe. 1990–1991. “The Early Full-Cost Debate and the Problem of Empirically Testing Profit Maximization,” Journal of Post Keynesian Economics 13.2: 236–251.

Lee, Frederic S. 1990–1991. “Marginalist Controversy and Post Keynesian Price Theory,” Journal of Post Keynesian Economics 13.2: 252–263.

Earl, Peter E. 1990–1991. “Normal Cost versus Marginalist Models of Pricing: A Behavioral Perspective,” Journal of Post Keynesian Economics 13.2: 264–281.

Langlois, Catherine C. 1990–1991. “The Empirical Testing of Pricing Rules: Response to a Symposium,” Journal of Post Keynesian Economics 13.2: 282–292.
I have yet to read all these papers, but I hope to review and comment on this debate when time permits.

The argument put to me recently is that, if marginal cost and average cost are the same or tend to coincide (given the shape of a marginal cost curve), then a neoclassical profit-maximizing firm will use average cost as a proxy for a profit-maximizing mark-up.

But do marginal cost and average cost really tend to coincide given that mark-up pricing obviously uses unit fixed costs or overhead costs,* as well as variable costs? Marginalism assumes that fixed costs (or overhead costs) are not used in setting prices, but it would appear that certain discretionary fixed costs (such as marketing and advertising, maintenance, and R&D) can be rather expensive.

Even given that average costs probably decline for many firms, if that firm has a constant marginal cost below average cost, then average cost and marginal cost will not be equal, nor will they tend to coincide if average cost stabilises at some point above marginal cost.

Thoughts on this (and the relationship between marginal cost and average variable cost) from people who know the specialist literature and empirical data better than I do are welcome.

Note
* Some notes are below on different types of costs for those who may be confused by the whole topic:

Average cost
This is total production costs per unit of output produced by a business. This equals (1) total fixed (overhead) costs plus (2) total variable costs divided by the number of units of output produced. Given that many businesses can use economies of scale and increase their output over time, average costs may fall too, because average fixed (overhead) costs fall, since they are divided by more units of output. Standard theory says that the average total cost curve is U-shaped, because eventually firms face diseconomies of scale. However, empirical evidence suggests that most firms’ long-run average cost curves within realistic output ranges do not hit diseconomies of scale, so that long-run average cost curves are L-shaped.

Fixed costs or overhead costs
Fixed costs (or overhead costs) are short-run costs that do not vary with the changing volumes of output produced, including rents, depreciation of fixed assets, marketing, etc. Average fixed costs will fall as output increases. Also called indirect costs.

Variable costs
Costs that vary with the rate of output, usually labour and raw materials costs. These are sometimes called operating costs, prime costs, on costs, or direct costs.

Marginal cost
The cost accruing from an additional unit of output.

Sunk costs
These are defined as those costs that a business cannot recover if it ceases operation, such as plant or machinery that cannot be re-used or resold. These constitute barriers to entry.


BIBLIOGRAPHY
Earl, Peter E. 1990–1991. “Normal Cost versus Marginalist Models of Pricing: A Behavioral Perspective,” Journal of Post Keynesian Economics 13.2: 264–281.

Frederic S. Lee. 1990–1991. “Introduction,” Journal of Post Keynesian Economics 13.2: 233–235.

Heflebower, R. F. 1955. “Full Costs, Cost Changes, and Prices,” Business Concentration and Price Policy. Princeton University Press, Princeton. 361–392.

Keen, Steve. 2011. Debunking Economics: The Naked Emperor Dethroned? (rev. and expanded edn.). Zed Books, London and New York.

Langlois, Catherine C. 1989. “Markup Pricing versus Marginalism: A Controversy Revisited,” Journal of Post Keynesian Economics 12.1: 127–151.

Langlois, Catherine C. 1990–1991. “The Empirical Testing of Pricing Rules: Response to a Symposium,” Journal of Post Keynesian Economics 13.2: 282–292.

Lee, Frederic S. 1990–1991. “Marginalist Controversy and Post Keynesian Price Theory,” Journal of Post Keynesian Economics 13.2: 252–263.

Mongin, Philippe. 1990–1991. “The Early Full-Cost Debate and the Problem of Empirically Testing Profit Maximization,” Journal of Post Keynesian Economics 13.2: 236–251.

11 comments:

  1. A main issue here is dynamics. It is quite plausible that the mark-up over average variable costs used in a particular industry is equivalent to the profit maximizing mark-up over marginal costs at a typical quantity of production ... however, the gap between variable cost and marginal cost may be different at different production quantities and the profit maximizing mark-up over marginal cost will also vary in different demand conditions as the elasticity of demand shifts. So the marginalist mark-up is responsive to changes in cost and demand conditions as they occur, while the mark-up over average variable costs is a strategic decision which is changed infrequently, with the timing of changes coinciding with the strategic decision making calendar.

    ReplyDelete
  2. Lord Keyns! Very important information!

    http://www.ebha.org/ebha2011/files/Papers/EBHA-Paper submitted by Tamara Ehs.pdf

    This article about austrian-interventionist.

    ReplyDelete
  3. Take a very simple model where land and raw materials are in abundant supply and all goods are produced by vertically-integrated companies who whose only cost is labor. labor is homogeneous. This labor is applied to extract raw materials , and then make the necessary capital and intermediate goods to produce the end product of consumer goods. All production processes are perfectly scalable. Full employment always holds Production techniques are fixed, and demand for output unvarying. Relative demand for all products is assumed to be the same no matter which pricing model is used, or how income is distributed.

    In a cost+markup world: Costs would be fixed at cost-of-labor+markup. All firms would adjust their output to the demand that allowed them to produce just the qty they could sel at the cost+markup price. The only things determining relative prices would be cost of production and strength of demand for those products at that price.

    In a marginal cost world: All prices are set by supply and demand. Assume imperfect competition so that as firms increase production they have to increase their input costs (wages in this simple model), and get a lower price for output. The set output and price at the point where the final piece of output generates a sufficient difference between cost and price to be worth producing. Assume that this "sufficient difference" is the same as the mark-up in the other model. Because of the assumptions I have made once the market price has been set it will true (by accounting principals) that all prices will equal to cost+markup (even though this is not the basis for price-setting used by firms)


    I think it can be shown that market forces in this simple model will cause relative prices and output to tend towards the same as under cost+markup. The reason for this can be seen by looking at the situation where this was not true for a specific good.

    In both cases prices can be described in terms of the unit of labor embedded in them plus an addition for profit. If a price (expressed in labor terms) is higher under marginal pricing than cost+markup then it should be clear that the profit margins for that good will be higher than form other goods and (based on our assumptions) production will expand and price fall until its margins are the same as other goods. This process of price and qty adjustments will continue until all relative prices and output levels are the same as under cost+markup.

    I have made a lots of assumption and showing what happens when they are relaxed is too complex to show here.

    I believe it it is very possible to build more complex models with less restrictive assumptions than this one that show similar results. As long as one assumes that the price of raw materials and labor are set by supply and demand then cost+markup models will yeild the same results as marginal-pricing models given reasonable assumptions. Where the models differ is where these original factors are priced by other than supply and demand (for example by legislation or convention).

    ReplyDelete
    Replies
    1. Rob, I quite like this challenge because its reasonably exactly stated and so the terms are not changing. The kind of stuff LK usually gets is arbitrary and changing definitions of terms so this is very reasonable, however

      Isn't your definition self contradictory? How can you have,

      "Full employment always holds" and "All firms would adjust their output to the demand that allowed them to produce just the qty they could sel at the cost+markup price".

      One of the themes of cost+ pricing is that firms adjust output levels, rather than prices. This is not consistent with guaranteed full employment is it?

      Delete
    2. The first part of Rob's model sounds like a general equilibrium state, which is simply irrelevant to the real world.

      The rest of it is unrealistic. For one thing it requires no barriers to entry or virtually no barriers to entry.

      The profit markup across sectors in the real world will always differ because competition is not efficient enough to drive profits to zero or even some uniform mark-up.

      Delete
    3. "The profit markup across sectors in the real world will always differ because competition is not efficient enough to drive profits to zero or even some uniform mark-up."

      Actually I disagree that this is necessarily the result. My line of thinking is that the Keen (and Stigler) critique of perfect competition appears still valid to me. The result of this being that even perfect competition yields the monopoly result, shall we say under reasonable assumptions.

      The main argument I have seen against is this one (from Chris Auld), "A 'competitive' firm in economic theory is one which takes prices as given, ignoring the effect of its own output on price. This is an assumption, not a result.". Such a firm also does not exist. As long as the demand curve slopes downwards the firms change in output still effects the market (right?). Despite the multiple claims of mathematical illiteracy on Keen's part, I simply don't see it (and I have a maths degree), while the argument seems to be that you can have a firm facing a horizontal and also downward sloping demand curve (which is mathematically illiterate).

      I think a similar result could hold here, though obviously the price adjusting market doesn't have to clear in this case...

      Delete
    4. Pity, it appears Rob has left the building and I don't see any obvious way to resolve the key question of if employment is allowed to adjust or not (in Rob's statement of the problem).

      Delete
    5. he model is very simple with assumptions that clearly are not met in the real world.

      As stated I believe that most of the assumptions can be dropped and drive the same results. I am working on a more complex model to demonstrate that.

      As far as I can see there is no self-equilibrium mechanism in cost+markup models. I simply had to assume full employment so I could compare the relative prices and outputs of the 2 models (In a more complex version of the model there will have be a central authority adjusting AD to maintain full employment). Assuming AD is at the level needed to maintain full employment then outputs adjust to match the strength of demand (if the output of one good increases then the output of another good must decrease accordingly).

      Delete
    6. So you are saying that there is sufficient demand to create full employment? But that as demand falls or rises then firms have the alternative to reduce output or increase output.

      e.g we should specify the level of demand of the model which creates full employment according to the other variables, and then compare that level to the flexible model, and see if the prices come out different or the same?

      Delete
  4. I am not going to go into detail but I work at a company that you have heard of. We are currently examining the business process for pricing our product. Prices are revised on the basis of desired return on investment and incurred costs approximately yearly although there have been periods of time in which the price has gone unchanged for more than a year and a half this includes the years immediately following the recession. I think we may be said to have administered prices.

    ReplyDelete
    Replies
    1. That is interesting.

      If you set prices by calculating average costs of production per unit with the addition of a profit mark-up (and it sounds like you do ), then this is certainly a mark-up price.

      Delete