Hi Julio, thanks for your latest message. My replies below. On Thu, Jan 10, 2013 at 7:11 AM, Julio Huato <[email protected]> wrote:
> Fred, > > Let's see if we can pin down the disagreements here. > > > I don’t get your distinction between a quantitative theory and a > > qualitative theory. How can marginal productivity theory (or any theory) > > be a quantitative theory without also being a qualitative theory. > > Are you saying that we cannot abstract quality from quantity? I think > it is entirely possible to say that the "substance" of profit is x > and, then, to say that the magnitude of profit is determined by y, > where x and y are two different propositions. Cf. Marx on relative > value in the very first chapter of Capital. First, he deals with its > qualitative nature and then with its "quantitative determination." > What's wrong with that? > Yes, Marx dealt with the quality (substance) of value – abstract labor – before the quantity (magnitude) of value – SNLT. But they are intrinsically related, not “different propositions”. The quantity of value is a quantity *of the quality *of value; i.e. SNLT is a quantity of abstract labor. I don’t see how the quantity of profit can be determined by the MPK, but the quality of profit be surplus labor. Please explain. > > Marginal > > productivity theory concludes that profit = MPK. The MPK is a quantity > of * > > output*, so the quality (or substance) of profit in this theory is > *output*, > > as opposed to labor or surplus labor in Marx’s theory. > > Why opposed? If you make output=labor, the opposition dissolves. > (Here I'm glossing over the distinction between use value and value, > because deep down -- as Marx shows -- they are the same thing, or -- > to be a bit more "precise" -- one is the social form of the other. > But the output that is supposed to determine profit (according the marginal productivity theory) is the marginal output of *capital*. The MPK cannot be equated with labor; it is independent of labor. According to this theory, profit is produced by capital, not by labor. > > This argument assumes that means of production can be added and output > > increased *without adding more labor*. I think there are many production > > processes for which this is not possible; e.g. another shovel without > > another worker to dig another ditch. > > I think you're making too much of this. Making too much of this? The basic concept of marginal productivity theory is marginal products, and this concept is meaningless in many production processes. This fundamental problem in marginal productivity theory has long been recognized, beginning with Hobson and Pareto and Edgeworth in the early 20th century. But neoclassical economists have continued to ignore this criticism with a wave of the hand and an unwarranted assertion that “fixed proportions can be dealt with by linear programming”. They continue ignore these problems because this theory is so useful for the ideological purpose of justifying capitalism and profit: profit is produced by capital, not by labor; i.e. workers are not exploited. > Of course you can add another > shovel without adding another worker and, thereby, alter the number of > ditches. Let the period of production be five years, L=1 worker, K=1 > shovel, and Q=100 miles of ditches. It's entirely possible that just > by increasing the number of shovels to K=2, you wind up with an > altered Q (e.g. Q=300 miles), as the worker can now safely discard an > old, worn, and blunt shovel, replace it with a new, sharper one, and > go on digging longer. Etc. > You are confusing the *stock* of shovels with the *flow* of shovels. Production functions (on which marginal productivity theory is based) are *flow** *functions that relate the flow of inputs to the flow of output in a production process. The production function for digging ditches has one shovel and one worker as inputs. If an old shovel is replaced with a new shovel, the flow ratio is still one shovel to one worker. One worker using two shovels in the same process is not possible. In order for another shovel to be useful in the same process another worker must be added. > > This argument also assumes that means of production can be added and > output > > increased *without adding more raw materials *(or intermediate goods in > > general). This is impossible in all goods-producing industries which > > require more raw materials in order to produce more output; e.g. in order > > to produce another car, one needs more wheels, brakes, glass, etc. > > I think this is another failure of abstraction. Of course you can add > means of production other than raw materials, keep the amount of raw > materials fixed, and alter (e.g. increase) output. Suppose you are > producing cars and using metal and glass as your raw material, and > also robots with sensors that monitor the amount of metal and glass > wasted in the production of your cars. Increase the number of robots, > and the amount of raw material saved may increase. (Similar results > could be attained by increasing L, human monitors, without increasing > raw materials.) As a result, you can produce more cars with the same > amount of metal or glass. QED. > When capitalists decide to increase output, they almost always increase their purchases of raw materials. Maybe not in exceptional cases, but increasing raw materials along with output is the general rule. Two implications follow from this general rule: (1) capitalists had better take into account the cost of additional raw materials in deciding whether to increase output, otherwise they would lose money; and (2) marginal products of capital or labor do not exist in the general case. > > This was my main critique of Krugman, and with respect to raw materials > > specifically. If the quantity of machines (or labor) increases in > > good-producing industries, *raw materials must also increase *in order to > > produce additional output (e.g. cars and wheels). So the marginal > product > > of capital (or labor) is *meaningless *in these production processes, > > because the concept of the MPK requires that *all other inputs be held > > constant*. You have not yet answered this key criticism, either in this > > post or in the post that you linked to on the Cambridge capital critique. > > I guess what you're trying to say is that, under certain assumptions > about the granularity of input and output units, in some processes > more than in others, the ability of producers to substitute one means > of production for another is next to zero, because production > possibilities are such that you have virtual Leontief isoquants, which > lead to undefined marginal rates of input substitution (and hence > marginal rates of input-output transformation) at given points or > segments. (I'm alluding to isoquants here as you yourself decided to > put aside the aggregation problem in your post, although I think we're > dealing with different versions of the same old argument.) Well, in > that case, your decision rule (optimal choice of input and output > levels) has discontinuities and you have to plug the holes in your > decision rule with intuition or seat-of-your-pants decision making, > but that doesn't mean that the decision rule derived under the > assumption of continuity is stupid. It just means that it's > insufficient to answer your every question. If you want a rule that > is more fine-grained, then you have to subdivide your inputs and > outputs into finer units, which may entail additional costs. Such is > life. > In the case of raw materials, the problem is not “holes in the decision rule”; the problem is that there is no decision to be made with respect to the ratio of raw materials to output. Only one ratio is possible. In order to increase output, raw materials must generally be increased proportionally. > > The fact is that the distinction between "raw materials" and "fixed" > means of production (in a firm's accounting or in the national > accounts) is much more fluid than it appears at first sight. It > hinges entirely on our conventions about the period and scale of > production. I'm not saying these conventions are entirely arbitrary > -- they make sense (are economical) for certain human purposes. But > they are conventions nonetheless. > I don’t see how the supposed “fluidity” of machines and raw materials solves the problem that I have posed. Marginal productivity theory is supposed to determine the *prices of the factors of production*, which means that it should determine both the prices of machines and the prices of raw materials, and should determine these prices *separately*, as separate variables. Hence they have to be analyzed as separate inputs, whose prices are supposedly determined by S and D of each input in separate markets. And the concept of marginal product *requires* that one of these separate inputs be increased and all the other inputs be held constant (including labor). However, frequently in the real world, “holding all other things constant” is not possible and therefore the concept of “marginal products” is meaningless. Julio, you have not yet responded to: 1. How can you make the marginal productivity theory of profit (= MPK) compatible with Marx’s theory of profit (= (MELT) (SL))? 2. How does an increase in the working day affect profit in marginal productivity theory? 3. Clark and many other neoclassical theorists have regarded marginal productivity theory to be an *alternative *to Marx’s theory and as a *refutation *of Marx’s theory (no exploitation, as opposed to exploitation). 4. Marginal productivity theory takes the rate of interest as given, and thus does not provide an explanation of the return to capital. Finally, Krugman (where all this started) is trying to explain the recent increase in the profit share of income in the US economy by “capital biased technical change”, which means (supposedly) that ↑MPK > ↑MPL. In other words, capital has become more productive, and that is why the profit share has increased; it has nothing to do with labor or surplus labor. Do you agree with Krugman? Thanks again for the discussion. Fred
_______________________________________________ pen-l mailing list [email protected] https://lists.csuchico.edu/mailman/listinfo/pen-l
