But, if we accept this, it just raises the question: how and under what conditions could commodity prices in a capitalist economy actually equal abstract socially-necessary labour time (SNLT) as Marx defines it?
The following conditions would be necessary:
(1) demonstrate how to reduce all heterogeneous human wage labour to a common, meaningful homogeneous unit of abstract labour time;But it is perfectly obvious that this is not how capitalism prices goods and even as a simplifying assumption it throws up all sorts of severe problems for Marx’s economic theories.
(2) calculate the SNLT required to produce all commodities and assign a consistent money unit to 1 unit of SNLT, e.g., 1 money unit would consistently equal one unit of homogeneous labour time;
(3) price all commodities and non-labour factor inputs in terms of socially necessary labour time required to produce them with the assigned money value of 1 unit of SNLT.
(4) under these conditions of course, we must assume also that workers are not paid the full and proper value of their labour in wage rates corresponding to SNLT, and that there is surplus labour value and money profit extracted by capitalists.
For one thing, the choice of what money unit you assign to 1 unit of SNLT is ultimately arbitrary, and at different money rates corresponding to one unit of SNLT (say, 1, 2, 3, 4, 5, 6, 7 etc.) you can potentially have any number of given price sets corresponding to SNLT, and any number of accidental instances where a real world commodity price might equal SNLT, under a given price set with a given money rate assigned to 1 unit of SNLT.
Under another standard interpretation, Marx in volume 3 of Capital sheds his simplifying assumption in volume 1 and assumes that most prices do not actually equal SNLT (and even average long-run Classical equilibrium prices do not directly do so), but only by accident do some occasional prices do so. But how on earth would you know when any individual real world market commodity price equals the SNLT required to produce it?
Marx does not explain how in real world markets a given commodity price would equal the SNLT needed to produce it, because it is never properly made clear how you determine such a price in the first place. Marx’s whole theory, even as a simplifying assumption and even in volume 1, is grossly under-determined and ambiguous.
BIBLIOGRAPHY
Mandel, Ernest. 1990. “Introduction,” in Karl Marx, Capital. A Critique of Political Economy. Volume One (trans. Ben Fowkes). Penguin Books, London. 11–86.
Marx, Karl. 1990. Capital. A Critique of Political Economy. Volume One (trans. Ben Fowkes). Penguin Books, London.
even as a simplifying assumption it throws up all sorts of severe problems for Marx’s economic theories.
ReplyDeleteCan you elaborate on this? I don't follow your point, and the paragraphs that followed didn't shed much light:
For one thing, the choice of what money unit you assign to 1 unit of SNLT is ultimately arbitrary
Perhaps I'm misreading you, but this seems incorrect prima facie. What makes an average "arbitrary"?
For clarity, see one of the explanations of the MELT shared below this post (one by Ramos-Martinez, one by Moseley, and two by Kliman).
But how on earth would you know when any individual real world market commodity price equals the SNLT required to produce it?
By analyzing the data! Marx even wrote a huge volume of political economy explaining how to do precisely this, which we've (supposedly) been discussing for months.
it is never properly made clear how you determine such a [value-equivalent] price in the first place.
Uh... what? This is literally one of the central points of Capital.
This is a bit like reading Keynes and going "you know, the General Theory is good and all, but I wish he'd have spent some time discussing Interest, Employment, and Money. Then he'd have himself a theory."
(1) "For clarity, see one of the explanations of the MELT shared below this post (one by Ramos-Martinez, one by Moseley, and two by Kliman)."
DeleteNone of them explain it in any sensible sense I can see. Kliman just repeats his aggregate identities. If you can't explain it here to me clearly, coherently and succinctly, you're just confirming my points.
(2) "By analyzing the data! Marx even wrote a huge volume of political economy explaining how to do precisely this, which we've (supposedly) been discussing for months."
Once again, explain it here to me clearly, coherently and succinctly.
Hedlund,
Delete"But how on earth would you know when any individual real world market commodity price equals the SNLT required to produce it?
By analyzing the data! Marx even wrote a huge volume of political economy explaining how to do precisely this"
That is extremely vague to the point of being meaningless... could you possibly explain or direct me to where you have previously done so?
Thanks.
(1) It's just algebra, an aggregate "conversion factor" (Kliman) -- i.e., "a ratio between the pound(£) and the substance of value (labour time), the dimension of which is £/w.d., where w.d. stands for working day" (Ramos). You can also use other standard units (such as $/hour), so long as you're ultimately relating total prices to total labor time.
DeleteMoseley's formulation differs in notation, but not in content, meaning all the documents say the same thing. So you don't even have to read all four; even just one would suffice.
The aggregate equality you mention (technically Marx's and not Kliman's) only relates to the numerator, so no, that's not "just" that. It is a ratio of two discrete, concretely measurable figures.
(2) First, take the general rate of profit and the average OCC. (For a real-world example, The Failure of Capitalist Production uses the BEA's NIPA stats to do this.) Then, apply them to the production data of a given commodity -- i.e., add the average profit rate to costs, and finally compare its OCC to the average to discover the ratio of p to s, which will give you the value. Compare with market prices as desired.
Money terms suffice much of the time, but if you need SNLT terms, just divide by the MELT.
Now, I've answered your questions, so can you please answer the one you ignored: Can you please elaborate what "even as a simplifying assumption it throws up all sorts of severe problems for Marx’s economic theories" refers to?
(1) you say divide the dollar/pound by the working day. Whose working day? What do you mean?
DeleteAnd what is that supposed to show?
(2) so you take the average unit cost of production for some good and add an average of the profit rates in different sectors? That would give you Marx's average price/long run Classical natural equilibrium price.
What does p and s stand for?
(3) you've not even defined how you obtain a MELT.
(4) "Now, I've answered your questions, so can you please answer the one you ignored: Can you please elaborate what "even as a simplifying assumption it throws up all sorts of severe problems for Marx’s economic theories" refers to?"
The ones I already mentioned in the post: how do you determine what money unit = a unit of SNLT?; how do you know when an individual price = SNLT?
"You can also use other standard units (such as $/hour), so long as you're ultimately relating total prices to total labor time. "
Delete"You can also use other standard units (such as $/hour), so long as you're ultimately assuming that Marx's theory of value is correct"
Oi, Philippe!
DeleteThat is extremely vague to the point of being meaningless... could you possibly explain or direct me to where you have previously done so?
Well, it wasn't the explanation in and of itself, but merely pointing out that such an explanation (among other things) does exist in Capital.
In case any of the terms in my last comment were unclear, organic composition of capital (c/v) is a ratio of capital advanced on means of production (c) and capital advanced on labor (v). It's basically a measure of capital intensity. Also, adding the average rate of profit to the cost of production (or "cost-price," for which Marx uses the symbol K, which is equal to c+v) gives us the "price of production," which is sort of like a "natural price," which only equals value at the average OCC. If s = surplus value and p = profit, a higher OCC means p > s; a lower OCC means the opposite.
c+v+s = K+s = a commodity's value
c+v+p = K+p = a commodity's market price
As you can see, the two have a significant portion in common straightaway, and the difference between a good's price and value is only the difference between surplus value and profit.
Whereas ∑s = ∑p, if a firm sells its goods at a price higher than value, that means one or more firms are selling below value -- perhaps liquidating inventory or the like. Accordingly, one capital expands as another contracts, effectively leeching flows of labor from one another.
The reason for the unrealistic abstractions of volume 1 (s=p for each commodity) is basically just to neutralize this ebb and flow, to determine the source of net profit (i.e., a firm's profits that don't appear elsewhere as another firm's losses).
This is basically the executive summary, but there's quite a bit more to it than that. Whereas my time is limited, my goal here is less to spell out the entire contents of Capital than to answer those niggling questions that inevitably come up when encountering such a radically different value theory than that to which most are used.
I hope that answered yours!
Whose working day? What do you mean?
DeleteIt's aggregate labor time in the denominator — the total social labor performed in the economy. Day is just one possible unit. Hours work, too. National accounts generally take some measurement of this.
And what is that supposed to show?
It's the relationship between a given monetary numeraire (i.e., total realized price in the numerator) and labor time. It's exactly the thing you were asking for -- how many dollars or pounds represent an average hour of abstract labor.
(2) so you take the average unit cost of production for some good and add an average of the profit rates in different sectors? That would give you Marx's average price/long run Classical natural equilibrium price.
Exactly. As I said above, where the OCC is average, s (surplus value) and p (profit) will be equal for a price of production. As OCC rises above that, p will outstrip s, and as it falls below it, the opposite.
Anyway, starting with total price, which gives us our aggregate s, we can calculate our average rate of surplus value (s/v). For a given firm within the system, then, multiply with their v figure to yield s.
I can draft up some numerical examples if you like, but it may have to wait a day or so.
(3) you've not even defined how you obtain a MELT.
Dude. :|
I bolded it and everything. Aggregate monetary figure divided by aggregate time figure. I even recommended you a book that spells out its method, down to naming the tables and rows in the NIPAs. I'm happy to help, but cooperation would be much appreciated.
The ones I already mentioned in the post: how do you determine what money unit = a unit of SNLT?; how do you know when an individual price = SNLT?
What money unit as in which currency? I suppose any could work, though I see dollars more than not (possibly because of its status as primary reserve currency?). I'd say it's up to the author.
In terms of quantity, see above.
As for the latter question, you'd need to run the numbers as described, using aggregate and firm-level data. Of course, which commodities exchange at their value is always much less interesting than those that do not; therein lies no small part of capitalism's dynamism.
Once more we see massive holes in your explanation:
Delete(1) it is NOT total concrete/raw labour hours worked per day needed, but homogeneous units of abstract socially necessary labour time that are required to aggregate ALL heterogeneous types of labour, as described and specified by Marx in Chapter of volume 1 of Capital.
You have provided no answer to my question and simply confirmed that your theory is grossly deficient and doesn't even deal with what Marx was talking about.
(2) it is now clear that you simply ASSUME without any evidence provided that a price of production would equal SNLT. You've not demonstrated why it would.
Mongiovi (2002) complains that the Temporal Single System Interpretation simply reduces
the concept of labour value merely to price (Mongiovi 2002: 406, n. 20), contrary to Marx. You have spectacularly confirmed this.
(3) "What money unit as in which currency? I suppose any could work, "
No, I mean how many dollars/pounds equal 1 unit of SNLT. How is this determined?
"it is now clear that you simply ASSUME without any evidence provided that a price of production would equal SNLT. You've not demonstrated why it would."
DeleteYup.
Assume the theory is correct, therefore xyz the theory is correct.
(1) I never mentioned concrete labor, so I don't even know what you're objecting to. Abstract labor is precisely what is measured in national accounts; the NIPAs don't specify whether wage hours are spent welding or weaving. They're just time spent in labor, period.
DeleteI can see the concrete/abstract split is still causing some problems, but I'm not sure how to explain it any more simply than I have. All I can tell you is, you're still getting it wrong. If you start with the idea that what you currently have in your head is wrong, then maybe you can ponder it until you can think of another way to view it that makes sense to you, and I'll be happy to let you know if what you come up with is any closer to the mark. But at this point, I've tried a lot of ways to say the same thing, and I have not seen any evidence that any of them has stuck. I'm frankly stumped.
That said, until you apprehend what it means for something to have both a concrete/individual and abstract/social character, you should take care charging that others are not "deal[ing] with what Marx was talking about."
(2) At the average OCC, you mean? It's not an assumption, it's simply a fact of the accounting (don't forget, as I worry you may have, that no value is created in exchange in this system). If there's no difference in the rate of surplus value, the organic composition of capital, or the rate of profit from the average of these, then s has no means by which to differ from p. This is just bookkeeping, and frankly I don't even see how you can make this accusation. You might as well say that we just "assume" S=I.
I'm trying to be as nice as possible, LK. I've even offered to take the time to flesh it out with tabular numerical examples, later. All I'm asking is, make fewer declarations and ask more questions. You'll even come out of it looking much more reasonable, as you had been in this thread up to now.
I've already directed you to multiple sources that shred Mongiovi's argument, also. If you're bringing more to the table than confirmation bias, I can't fathom why you wouldn't be interested in reading them.
(3) Alright, I've answered this at least twice now. Here comes three: Aggregate money figure, divided by aggregate time. For example, the BEA has figures for "full-time equivalent employees" in private industries that works for the denominator. Kliman uses a measure of net value added in the business sector for the numerator. Other theorists may choose other measures, but the underlying theory (value produced in money terms divided by time) is the same. Can we be done asking this, now?
Assume the theory is correct, therefore xyz the theory is correct.
Thanks, Philippe. Real helpful. Nothing to say about that detailed answer I gave you?
Man, you guys really know how to take the joy out of discourse.
Hedlund,
Deleteyes, thank you for that detailed response.
It would make more sense to me if the divergence of prices from values was randomly distributed across all different types of industries and sectors.
But what you seem to be saying is that prices diverge from value in a way that completely contradicts what the labour theory of value would predict.
i.e. profits are higher than they should be (relative to value) in more capital-intensive industries and lower than they should be in less capital-intensive industries. This means that people are systematically overpaying for commodities produced by more capital-intensive industries. Alternatively you could say that people are systematically 'over-valuing' commodities produced by more capital-intensive industries. If I have understood this correctly, it seems like a very serious problem for the theory. To be honest I don't understand your explanation for why this happens:
"one capital expands as another contracts, effectively leeching flows of labor from one another."
Why should the more-capital intensive industries systematically 'leech' value somehow from the less-capital intensive ones?
It sounds a bit like adding epicycles to explain phenomena that contradict basic predictions of the theory, i.e. that in equilibrium price is equal value which is equal to labour-time.
This opposite result, in which prices of production are not equal to labour values, but are instead supposedly above labour values in more capital intensive industries, would seem to fit perfectly with the assumption that labour is not the only element which 'produces value'. If capital or some other element also 'produces value' in some way, would that not explain the discrepancy more simply?
Where in any in national accounts do governments calculate SNLT as defined by Marx? Show me the evidence.
DeleteHedund,
Deletealso you seem to be assuming that these deviations of prices of production from labour values all cancel each other out. However that does not seem to be the case:
"...Marx thought that embodied labor redistributed by the process of competition meant that in the aggregate total surplus value corresponded to total profits, even if prices of production deviated from embodied labor. As a result, on the basis of the LTV it was still possible to obtain the correct rate of profit. As it turns out, there is no reason for positive and negative deviations of prices of production from the labor values to cancel out. You cannot argue with the algebra.
Marx had no way of knowing this. Only with Bortkiewicz, Dimitriev and Tugan-Baranovsky’s work, early in the 20th century, this was clearly understood." (Matias Vernengo)
http://nakedkeynesianism.blogspot.co.uk/2012/08/sraffa-and-marxism-or-labor-theory-of.html
Hi Philippe, thanks for the responses.
DeleteThis means that people are systematically overpaying for commodities produced by more capital-intensive industries.
Not really; the theory doesn't make claims about which price is more or less "just." Rather, the differences between price and value tell us a great deal about the way some capitals grow and others shrink.
If we are "overpaying," it suggests that profits "should" be in stronger evidence in more labor-intensive industries. That's more like what you might call a "labor theory of price," which runs counter both to the theory I'm discussing and observed reality.
Why should the more-capital intensive industries systematically 'leech' value somehow from the less-capital intensive ones?
Because the profit-equalizing pressure exerted by the movement of capital operates the same way regardless of a firm or sector's capital intensiveness.
It sounds a bit like adding epicycles to explain phenomena that contradict basic predictions of the theory, i.e. that in equilibrium price is equal value which is equal to labour-time.
Well, I mean, that's not actually a premise of the theory. Marx is absolutely clear that prices of production and values are distinct concepts.
This opposite result ... would seem to fit perfectly with the assumption that labour is not the only element which 'produces value'.
I can see how some might see it that way -- in fact, I did for most of my life -- but it'd require us to equivocate on "value."
The underlying propositions remain simple and uncontroversial: there is a finite amount of social labor producing a finite output of commodities, that a finite portion of the former goes into each unit of output of the latter, that something must be produced before it can be consumed, etc.
We don't have self-operating and maintaining machinery to obviate the need for human involvement (which would call for a different value theory anyway), and if any theoretical input can produce a unit of output instantly without labor, then barring the influence of supply and demand, or intellectual property rights or the like, anyone would expect that output to be arbitraged down to the price of the input, since producing that input IS effectively production of the output.
As long as humans work together to produce, there needs to be an organizing principle. In capitalism, that principle is a form of objective (capital) value, existing only through the particular relations that we form with one another and nowhere else -- an emergent social phenomenon.
also you seem to be assuming that these deviations of prices of production from labour values all cancel each other out. However that does not seem to be the case:
No one disputes the algebra; we dispute the assumptions underlying it, and thus the way it is employed. Bortkiewicz, Dmitriev et al took a particular interpretive tack (dual-system, non-temporal/simultaneous valuation), but the algebra says something quite different if one incorporates a single system, historical time, etc.
If my interpretation can reproduce Marx's analytical conclusions from his premises and definitions, while the others cannot, that says something. Doesn't prove he was "right," but nor is he still a priori excluded from the possibility of correctness. From here, the way forward is empirical.
LK: "Where in any in national accounts do governments calculate SNLT as defined by Marx? Show me the evidence."
Huh? I never claimed governments calculate SNLT. But they do have stats (such as the examples I gave) we can use to calculate it ourselves.
Statement 1:
Delete"Abstract labor is precisely what is measured in national accounts;"
Statement 2:
"Huh? I never claimed governments calculate SNLT. "
Incoherent.
" Marx is absolutely clear that prices of production and values are distinct concepts."
DeleteThen why above do imply prices of production are equal to value?
Do you understand the difference between saying that governments keep aggregate measures of labor time and saying governments calculate the unit of average socially necessary abstract labor time?
DeleteI can explain, if not; I just wanted to give you the chance to realize the distinction I'm making on your own.
Then why above do imply prices of production are equal to value?
DeleteBecause a price of production has the average organic composition of capital, then they are indeed equal. If not, then they're not.
I have a close friend who is slightly older than I am. There is a particular point in the year in which we are the same age, measured in integer years. Does that mean our ages are not separate concepts?
Just because some conditions can bring X and Y into equality does not in any way imply that the two are interchangeable.
"Because a price of production has the average organic composition of capital, then they are indeed equal. If not, then they're not."
Deleteand you've never shown how. In short, this is just another pointless discussion.
Hedlund,
Delete"We don't have self-operating and maintaining machinery to obviate the need for human involvement"
Clearly you need human labour for production to occur - but you also need machinery, land, etc. Labour on its own is not enough.
You seem to be just assuming that only labour 'produces value'. But there is no necessary reason, as far as I can see, why this assumption must be correct. At least I can't see any reason for this in the explanations you have given.
Indeed the simplest explanation for why prices of production deviate from supposed labour values is that labour is not in fact the only thing which 'produces value'.
"it'd require us to equivocate on "value.""
Why?
"If we are "overpaying," it suggests that profits "should" be in stronger evidence in more labor-intensive industries."
If prices of production are below supposed labour values in more labour-intensive industries, and as you say profit is below supposed surplus value in those industries, then this indicates that profits in those industries should be higher. The prices they are selling commodities at are supposedly below the value of those commodities - why would they do this?
"You seem to be just assuming that only labour 'produces value'. But there is no necessary reason, as far as I can see, why this assumption must be correct."
DeleteSpot on. In fact, he has already admitted that his theory would cease if virtually all work and production was done by machines.
By the logic of that admission, there would no surplus value and no source of profit in such a world. Yet it is obvious that you could have a world where production and profits continue to occur, and where business add value to factor inputs. So what would be the source of profits in that world?
Hedlund,
Delete"Because the profit-equalizing pressure exerted by the movement of capital operates the same way regardless of a firm or sector's capital intensiveness."
This might make sense if the deviations of prices of production from labour values were randomly distributed among different industries. Some prices of production might just happen to be higher than labour values whilst others were lower, with prices of production and labour values only being equal on average, overall - due to constant change in the economy, perhaps (?).
However, the fact that prices of production are systematically higher than labour values in capital-intensive industries and systematically below labour values in labour-intensive industries means that such an explanation doesn't seem to make sense.
Instead it would appear to indicate that something else is going on, the most obvious explanation being that labour is not the only thing which 'produces value' and so the supposed deviation of prices of production from labour values is in fact simply the result of a incorrect assumption.
Kids, this is why you don't debate with a resurrected dead guy. They have all the memory of a goldfish.
DeleteI'm gonna try again, saying exactly the same things I've been saying for months. This time, though, I'll use pretty symbols, in the hopes it'll stick. If this still does not classify as a "showing," then I guess I just don't know how to show people things.
c = capital advanced on means of production
v = capital advanced on labor
K = cost-price, or c+v
s = surplus value
p = profit
c/v = organic composition of capital
s/(c+v) = rate of profit (I'll call the general rate "r")
K+s = commodity's value
K+p = commodity's price
Because no value is created in exchange:
∑K+∑s = ∑K+∑p
Therefore:
∑s = ∑p
For a given commodity, p can vary from s depending on market factors (supply and demand) and conditions of production (OCC).
A price of production's p is equal to the general rate of profit. Thus:
K+K*r = price of production
Prices of production all see the same rate of profit regardless of the OCC, meaning that higher OCC will correspond with p > s, while vice versa with low OCC, though these cancel in aggregate (see ∑p = ∑s). They are only exactly equal when OCCp (the organic composition of a price of production) is equal to OCCa (the average organic composition of capital). This is an accounting construct based on the earlier equality.
Aside from the way I gave earlier in this thread (involving aggregate s/v), another way to determine an individual commodity's value is with reference to the general rate of profit (r) and the OCC:
For a given market price:
s/(K*r) = ((1+OCCa)/(1+OCCp))
s = (K*r)((1+OCCa)/(1+OCCp))
Now, can we finally stop with the "you've never shown" nonsense?
Clearly you need human labour for production to occur - but you also need machinery, land, etc. Labour on its own is not enough.
DeleteHumans don't enter into social relations with machines or land, though. We're social beings. The economy, and markets, are social constructs.
But there is no necessary reason, as far as I can see, why this assumption must be correct.
Of course there isn't. Scientific theories are meant to explain the world, not themselves; that way lies infinite regress. It's not "necessary," but it is coherent, consistent, and plausible when properly understood.
Indeed the simplest explanation for why prices of production deviate from supposed labour values is that labour is not in fact the only thing which 'produces value'.
Agreed. But then, as someone once noted, if the nature of things coincided with their appearance, we'd have no need for science at all.
Why?
Because suddenly we'd be discussing what produces output (perhaps even useful output!) rather than what produces the (objective) value that accumulates as capital.
If prices of production are below supposed labour values in more labour-intensive industries, and as you say profit is below supposed surplus value in those industries, then this indicates that profits in those industries should be higher.
Not really, no.
The prices they are selling commodities at are supposedly below the value of those commodities - why would they do this?
Because it's profitable. Otherwise they wouldn't. Just because K+s is less than K+p doesn't mean that p isn't positive.
In fact, he has already admitted that his theory would cease if virtually all work and production was done by machines.
You say "admitted" as though this were something I would ever deny. It's a fact. "Admitting" facts is kind of my life's work.
By the logic of that admission, there would no surplus value and no source of profit in such a world. Yet it is obvious that you could have a world where production and profits continue to occur, and where business add value to factor inputs. So what would be the source of profits in that world?
Is it obvious? You seem to have thought this all out, so please tell me how society would be organized if humans left all the work to machines.
It sounds a bit like the people who own the handful of major robocorps will constitute a new untouchable aristocracy, propped up by state enforcement of private ownership. Perhaps that appeals to a Lord such as yourself?
"Is it obvious? You seem ... etc. etc.
DeleteA long red herring rant that addressing nothing I said.
LK-Marx does not explain how in real world markets a given commodity price would equal the SNLT needed to produce it, because it is never properly made clear how you determine such a price in the first place. Marx’s whole theory, even as a simplifying assumption and even in volume 1, is grossly under-determined and ambiguous.
ReplyDeleteWhy is that important? Marx's value theory is not a theory of individual prices.
If no individual price is ever determined by SNLT, then how would you know if workers are robbed of surplus value?
DeleteLK, PHILIPPE
ReplyDeleteLK - "If no individual price is ever determined by SNLT, then how would you know if workers are robbed of surplus value?"
Hey LK -
Marx held that the total value = total price and the total surplus value = the total profit. It seems Hedlund went into that above but I haven't read the exchange in its entirety.
The point is that you seemed to confuse Marx's value theory as a theory of price.
Whether or not SNLT matched price in specific cases was viewed by Marx as coincidence, and the deviations were what Marx was interested in.
PHILIPPE -
P- "You seem to be just assuming that only labour 'produces value'. But there is no necessary reason, as far as I can see, why this assumption must be correct. At least I can't see any reason for this in the explanations you have given."
LK- "Spot on. In fact, he has already admitted that his theory would cease if virtually all work and production was done by machines. "
Value creation is just as much about relations to production as actual production. It is human labor under particular circumstances that produces value (hence this is a critique of CAPITALISM - not every mode of production). I think Hedlund hits on this a bit here -
H - "As long as humans work together to produce, there needs to be an organizing principle. In capitalism, that principle is a form of objective (capital) value, existing only through the particular relations that we form with one another and nowhere else -- an emergent social phenomenon."
The colloquial and short-form summary of the theory is that labor creates value. But value-creating labor is only possible under specific material conditions - this shouldn't be surprising as the entire context of the theory is within a specific mode of production. Capitalism. Ultimately, rendering this value theory antiquated and irrelevant by changing the mode of production is exactly what many marxists aim to do. Value is as much the product of the social power over labor as it is the product of labor itself.
Pointing out that "[this] theory would cease if virtually all work and production was done by machines." doesn't mean this theory lacks explanatory power under conditions where humans engage in labor under a particular property system - e.g. this mode of production. It just means you need a different value theory for different material conditions.
LK - "By the logic of that admission, there would no surplus value and no source of profit in such a world. Yet it is obvious that you could have a world where production and profits continue to occur, and where business add value to factor inputs. So what would be the source of profits in that world?"
I think there is an equivocation with the word "value" here.