An economy’s minimum wage equates a unit of the currency to an amount of labor time. For instance, in Marxist terms, a minimum wage of $15/hour sets a dollar equal to 4 minutes of simple labor power. At a macro level, this enables currency value to be defined in terms of simple labor. There are, however, at least two ways in which this connection between currency value and labor could be drawn. One way would be to adopt a labor command theory of currency value. In effect, Modern Monetary Theory (MMT) takes this approach. A second way would be to link the value of the currency to the commodity labor power. Adopting the second approach leads to a definition of currency value that is distinct from the MMT definition but closely (and simply) related to it. So far as policy implications go, especially in relation to MMT’s proposed job guarantee and prescriptions for price stability, there appear to be no important differences between the two approaches.
To be clear, the purpose of the post is not to promote one approach over the other. So far as I can tell, on the question of currency value they are equally valid and fully compatible. The purpose is simply to consider, for readers who might be more inclined toward a commodity theory of money, how some form of commodity theory (though not a metalist one) might be reconcilable with MMT’s depiction of institutional realities and the opportunities open to monetarily sovereign societies, this understanding seeming, to me at least, both unassailable and fundamental to any worthwhile macroeconomics.
Nor is there any claim to novelty. Papers by Wray and Tcherneva (an example is linked to below) cover the MMT aspects of the topic, and Marx of course wrote volumes on value and a monetary production economy, with the commodity labor power playing a key part in his analysis. The post also benefits from internet contributions, here and elsewhere, by Magpie, Hedlund, Tom Hickey and others, without implicating them in any errors.
Two conceptions of currency value
Conception 1: a labor command theory. In this view, a currency’s value is the amount of labor it commands. If the wage paid for simple labor power is $15/hour, one dollar in wage payment will command 4 minutes of simple labor power which, when expended in production, will on average result in the performance of 4 minutes of simple labor by minimum-wage workers.
According to Marx, complex labor can be treated as a multiple of simple labor, with the wage paid for an hour of complex labor tending to reflect its degree of complexity. For example, an hour of labor performed by a worker whose labor is twice as complex as simple labor will tend to be paid $30, which conceptually can be considered equivalent to a worker performing two hours of simple labor and being paid $15/hour. To the extent that this tendency is operative, at least at the level of aggregates (i.e. on average), a dollar in wage payment will command the equivalent of 4 minutes of simple labor.
In general, then, a labor command approach suggests the following definition:
Currency value = the amount of simple labor performed, on average, per monetary unit advanced on wages.
This definition is consistent with MMT. It is basically the flipside of the MMT approach, which defines a currency’s value in general terms as “what must be done to obtain it” (see, for example, this paper by Tcherneva and Wray, especially pp. 14-15, 20). As Modern Monetary Theorists point out, the difficulty of acquiring the currency can be viewed in terms of labor time. Translated into Marxist terms, a wage of $15/hour, applying to simple labor, will require workers to sell the equivalent of 4 minutes of simple labor power to obtain a dollar and, through the expenditure of this labor power in production, perform an average of 4 minutes of simple labor.
So, viewed from one side, a dollar in wages commands, on average, 4 minutes of simple labor. Viewed from the other side, workers on average must perform 4 minutes of simple labor to obtain a dollar in wages.
Government can influence the average wage paid per hour of simple labor (w) by setting a minimum wage wmin. If, in fact, all workers were paid strictly according to the complexity of their labor, the wage paid for the equivalent of an hour of simple labor would equal the minimum wage; i.e. w = wmin.
At any given moment, wages can differ from the levels commensurate with labor complexity. If the wages paid to workers performing complex labor rise or fall systematically (or at least on average) relative to the minimum wage, w will diverge from wmin. Systematic movements in the wages paid for complex labor power can occur, in particular, over the business cycle, with these wages spreading further above the minimum wage during booms and falling closer toward the minimum wage during slumps.
Currency value, on the basis of the above considerations, can be defined as the reciprocal of the hourly monetary wage w paid, on average, for the equivalent of an hour of simple labor. The wage rate w needs to be calculated as total money wages paid to productive workers W divided by total productive employment L, where the latter is converted into hours of simple labor. The same basic idea can be expressed in Keynesian terms by defining currency value as the reciprocal of the ‘wage unit’ (i.e. the wage paid for an hour of ordinary labor) and measuring employment in hours of ordinary labor (the ‘labor unit’). Modern Monetary Theorists tend to adopt Keynes’ terminology. The two approaches are basically equivalent, with simple labor and the wage paid for simple labor (Marx) corresponding to ordinary labor and the wage unit (Keynes).
Let a currency’s value be denoted z when it is defined in terms of labor commanded (or, equivalently, what must be done to obtain it). Taking the reciprocal of the average wage paid for an hour of simple labor:
where w tends to wmin. When relative wages strictly reflect complexity,
The government’s choice of minimum wage therefore determines the level to which currency value tends. Divergence between the average wage paid for the equivalent of an hour of simple labor (w) and the minimum wage (wmin) will cause currency value to deviate from the level to which it tends.
Conception 2: a commodity theory. In the view of some, the currency expresses value to the extent that it represents an amount of a commodity that has value.
For Marxists, value is socially necessary labor, measured in hours of simple labor. When, as under a gold standard, the currency is equated to a quantity of gold, the value expressed by a unit of the currency will be the amount of simple labor needed to produce that quantity of gold.
If, instead, the currency is equated to an amount of labor power, the logic of a commodity theory leads to the following definition:
Currency value = the amount of simple labor needed to reproduce the quantity of labor power represented by a unit of the currency.
Under this definition, if a unit of the currency purchases x amount of labor power, then the value expressed by one unit of the currency would be the amount of socially necessary labor going into the cultural reproduction of x amount of labor power.
Suppose, as previously, that the average wage paid for simple labor power is $15/hour, so that one dollar equals 4 minutes of simple labor power. In that case, currency value can be interpreted as the amount of simple labor needed to reproduce 4 minutes of simple labor power. This measure of currency value will depend (in addition to the wage paid for simple labor power) on the rate of exploitation. If the rate of exploitation is 100 percent, so that workers spend half their working hours reproducing themselves and the other half creating surplus value, the value of the currency, under a commodity theory, will be 2 minutes of simple labor per dollar.
Let λ denote currency value when defined in terms of a commodity theory. In general:
Here, v and s are variable capital and surplus value respectively, measured in hours of simple labor. Their sum is value added or, equivalently, total productive employment L (i.e. L = v + s), which is also measured in hours of simple labor. The fraction v/(v + s) is the workers’ share in value added.
When relative wages strictly reflect complexity,
because of the tendency for w (the wage paid, on average, for the equivalent of an hour of simple labor) to converge on wmin. So currency value, under this definition, tends to a level that is determined by distribution and the government’s choice of minimum wage.
Relationship between the two notions of currency value
Currency value has been denoted z when defined in terms of labor commanded and λ when linked to a commodity. The two definitions of currency value are reproduced below for convenience:
It has been noted that v/(v + s) is the workers’ share in value added. The reciprocal of this share is the aggregate markup k expressed in value terms:
This makes explicit that both the markup and its reciprocal, the worker’s share in value added, depend on the rate of exploitation s/v.
On the basis of (3) and (4),
where w tends to wmin.
The price level and monetary expression of labor time
Before considering the main policy implication that flows from either conception of currency value, it is helpful to introduce two more macroeconomic concepts, namely the general price level and the ‘monetary expression of labor time’ (MELT). Currency value relates closely to both these measures.
Price level. A convenient entry point for considering prices is the aggregate markup. The markup can be expressed in price terms as nominal income PY divided by total money wages W:
P is an index of the price level, a pure number. Y is ‘real’ output. For ease of correspondence with value categories, it is taken to be net of depreciation. Real output is measured in monetary terms but deflated by the price index.
By identity, real output is the product of total productive employment L, here measured in hours of simple labor, and productivity ρ. Productivity, in the present context, is real output per hour of simple labor.
Making P the subject of (6) and replacing Y with ρL and W/L with w (the hourly monetary wage paid, on average, per unit of simple labor) gives:
This shows that, for given productivity, the price level depends on the wage paid for simple labor power or its equivalent (w) and on the markup over this wage (k).
Since w is the reciprocal of z, and z is closely related to λ, the price level can also be said to depend on both currency value (under either definition) and the markup (which is already included in currency value in the case of λ):
Monetary expression of labor time (MELT). The MELT, denoted m, is a Marxist concept. It is the amount of monetary value created per hour of simple labor. For given productivity:
Noting that s = L – v and solving for m gives:
The MELT is therefore closely related to currency value, under either definition.
The MELT is also closely related to the price level. On the basis of (9), z = k/m. For given productivity, (8) can be modified accordingly:
It should be noted that the definition of the MELT needs to be modified once productivity is permitted to vary. Variable productivity somewhat complicates the connections between currency value, the price level and the MELT. Some of these connections are considered in an earlier post.
Core policy implication
The connections just considered suggest that policymakers, by exerting influence over the value of the currency, can moderate variations in the price level and MELT.
Price stability. The key relationships pertaining to the price level are summarized in (8):
For given productivity, policy can act upon currency value (z or λ) and the markup (subsumed in currency value in the case of λ).
In this regard, MMT starts from a recognition of a currency-issuing government’s capacity to act as price setter. Such a government can exercise this capacity to a greater or lesser degree, at its discretion. At minimum, it can set one price, the economy’s minimum wage, and allow other wages and prices to vary in relation to this fixed price.
The government can also decide to exercise tighter or looser direct control over other wages and prices through its decisions concerning what prices to pay for goods and services supplied to it by non-government as well as through any price and wage controls or other regulations it decides to impose. The government’s decisions over the prices it is willing to pay for the output of the private sector directly affect the markup (k), while its wage setting and any wage controls it imposes affect the average wage paid for the equivalent of an hour of simple labor (w).
The government also exerts indirect influence over wages, currency value and prices through the use of fiscal and monetary policies designed to moderate demand.
MMT suggests that price stability would be fostered by implementing a job guarantee that anchored the nominal price structure to the program’s policy-administered wage. Since this wage would, in effect, become the economy’s minimum wage (wmin), it would tend to establish the wage received for an hour of simple labor, or its equivalent. So long as wage relativities tend to reflect the complexity of different categories of labor, the job-guarantee wage would anchor the value of the currency in the sense of z. Employers in the ‘productive’ sector (in Marx’s sense) would have to match, or better, the job-guarantee wage, in accordance with the complexity of labor required in the positions they offered.
The effectiveness of the nominal price anchor would depend, in part, on the state of the economy. To a significant degree, demand fluctuations would be moderated automatically by the job-guarantee mechanism, with government spending on the program rising and falling countercyclically in response to take-up of positions. Even so, conditions of either boom or bust would be liable to cause wages for complex labor power (outside the job-guarantee program) to deviate, on average, either above or below the levels commensurate with labor complexity. For a given markup, this spread or compression of wages (and consequent change in currency value) would cause movements in the general price level. In effect, the wage paid, on average, for the equivalent of an hour of simple labor would diverge from the job-guarantee wage (i.e. w would diverge from wmin) and, as a consequence, currency value z would also diverge from 1/wmin, with implications for prices.
The policy implications are really no different when currency value is instead interpreted as λ. Recall that in this conception the markup k is included in the definition of currency value. For given productivity, price stability continues to depend on the markup and the effects of wage spread or compression on the average wage paid for simple labor power.
Stability of the MELT. Since, under present assumptions, the MELT is simply the price level multiplied by productivity (i.e. m = Pρ) as shown in (8′), factors relevant to price stability are equally relevant to the stability of the MELT. Or, expressed another way, since currency value in the sense of λ is just the reciprocal of the MELT, policies that affect λ inversely affect the MELT.
The job guarantee as a commodity standard
Viewed through the lens of a commodity theory, the job guarantee as prescribed by MMT would essentially operate as a ‘labor-power standard’ whereby government fixed the rate at which simple labor power is convertible into currency, on demand.
Of course, trend improvements in productivity would enable sustainable, non-inflationary wage increases and so justify a policy-administered increase in the job-guarantee wage. Periodically it would be appropriate to revise the fixed rate at which government stood ready to convert simple labor power into a unit of the currency, in line with productivity.
Unlike a metal standard, a labor-power standard would leave the fiscal capacity of government unimpeded, making it more conducive to the maintenance of full employment alongside the pursuit of other worthwhile social and economic goals.