I’ve puzzled about this a lot (which might be just an index of my being easily puzzled), and blathered in posts here and here etc. In short, I think exchange value should be understood as a subset of use value, such that what is sometimes characterized as a conflict of use value vs exchange value should instead be characterized as a conflict of some uses vs other uses (where one or the other side is understood to be — that use value which is — exchange value).

Marx writes that a commodity producer “not only produces use-values, but use-values for others, social use values.” In a parenthetical remark, Engels adds that “[t]o become a commodity a product must be transferred to another, whom it will serve as a use-value, by means of an exchange.” In a footnote Engels explains that he inserted his parenthetical to correct or avoid the common “misunderstanding that every product that is consumed by some one other than its producer is considered in Marx a commodity.” (41 International Publishers Edition, Moore and Aveling translation.) So, a commodity is something which is exchanged, and exchanged in a specific manner. The quit-rent-corn produced by edieval peasants for feudal lords and the tithe-corn produced by medieval peasants for parsons don’t count as commodities, which is to say, the exchange made between peasants and lords and parsons don’t count as exchanges in the sense of exchange-value. (The same goes for the intra-household and other exchanges described by the Boydston article.)

Marx writes that in capitalism “commodities must be realised as values before they can be realised as use values.” I take this to mean that in capitalism access to use values is mediated or impeded by values – monetary and labor-time.
(I commented briefly on “realised” in Marx in the second post linked to in this post. At some point treat the “real” in “real subsumption” and comparing it with the idea of being “fully capitalist” or the “specifically capitalist” mode of production as referred to here. Note also, “exchange (…) puts [commodities] in relation with each other as values, and realises them as values.” p85. Also “realised” as in make a reality: “Use values become a reality only by use” is the Moore and Aveling translation of what Fowkes renders as “Use values are only realized [verwirklicht] in use or in consumption.” 36 and 126 in the respective editions.) Marx continues, “On the other hand, they must show that they are use-values before they can be realised as use-values. For the labour spent upon them counts effectively, only in so far as it is spent in a form that is useful for others. Whether that labour is useful for others, and its product consequently capable of satisfying the wants of others, can be proved by only by the act of exchange.” (85.)

In the second post linked to above I wrote that use values are retroactively attributed to items at the moment of or after the moment of their use. Likewise for exchange value, retroactively attibuted to items at or after the moment of their exchange. Given that I think exchange value is a subspecies of use value, if I think use value is retroactively posited for all use values, I’m committed to holding this for exchange value. In both cases of retroactive attribution the past or present status (has use value/doesn’t have use value) hangs on a moment later on in time. In this passage on page 85 Marx hangs use values from exchange value. “Whether that labour is useful for others, and its product consequently capable of satisfying the wants of others, can be proved by only by the act of exchange.”

In one sense, this last line can not be true. Non-exchanged use values exist, including some involving labor. Engels’ examples of peasants are one case, intra-household exchanges are another. Marx says as much: “A thing can be useful, and the product of human labour, without being a commodity.” (40.) Plus, exchangers must have at least some belief that the items they are exchanging are useful or they wouldn’t exchange them. On the other hand, understood here in terms of what counts as labor or not – and thus can claim need for remuneration in wages – this is more true. Only labor that results in surplus value is counted as labor (and not all of the labor which contributes to surplus value), and only exchanges that result in – or have the possibility to result – increased wealth count as commodities. Exchange value dominates over or mediates other use values.

In any case, use value and exchange value. Some textual evidence. When someone brings a commodity to market, that commodity for that person “possesses (…) no immediate use-value. Otherwise [that person] would not bring it to market. It has use value for others” who are the buyers or potential buyers of the commodity. Marx then writes that for the commodity bearer the commodity brought to market’s “only direct use-value is that of being a depository of exchange value, and, consequently, a means of exchange.” That is, a commodity exchanged has the direct use value of being a means of exchange. (85.) Immediacy, direct-ness, and mediation here all bear further attention, as when Marx writes on 87 that a commodity exchanged “forms a superfluous portion of some article required for [the] immediate wants” of the person who brings it to market to sell. (This too: “What makes [commodities] exchangeable is the mutual desire of their owners to alienate them.” 87.)

Marx writes on 88 of “the distinction [which] becomes firmly established between the utility of an object for the purposes of consumption, and its utility for the purposes of exchange. Its use-value becomes distinguished from its exchange-value.” Exchange value is utility for exchange. Strictly speaking, this a use value. Use values satisfy or serve in the satisfaction of “human wants of some sort or another. Th e nature of such wants, whether, for instance, they spring from the stomach or from the fancy, makes no difference.” (35.)

Marx also writes of the “use-value of the money-commodity” specific to its being money. Aside from other use values it might have – “gold, for instance, serving to stop teeth, to form the raw material of articles of luxury, etc” – “it acquires a formal use-value, originating in its specific social function.” (89.) Marx insists that money is a commodity. Since all commodities have use value and exchange value, money must have use value and exchange value. This could be understood in the sense that money is always made up of some or the other commodity which serves the money function(s). This commodity which serves as money would have its own use values (gold can be made into fillings), its own exchange value (gold can be bought), and serves the role of being money. Marx calls the money function(s) specifically a use-value, though, which means that money qua money has a use value, as opposed to money qua gold (or paper or cattle or whatever) having a use value and money qua money having no use value. (Question: does money have exchange value? Clearly currencies do, as they are traded, but in that sense are they serving as money or as a commodity other than money?)