Short run time periods, for Marshallian firms, are defined by the presence of separate sets of fixed and variable production factors. As in our elaboration of the Walrasian/Paretian "pure" Neoclassical firm, we will proceed in this section with an assumption that we can, at least for purposes of presentation, homogenize the factors of production utilized by the firm, reducing entrepreneurial decisions on factor combinations to a two factor (labor and capital) case, even as we acknowledge the potential to analyze a wide array of heterogeneous factors with different temporal degrees of variability. In this sense, the critical differentiation here relative to Walrasian/Paretian firms remains the idea that for Marshallian firms certain factors must be held fixed in short run periods.
Drawing from our hypothetical firms in the previous section, for example, our interurban bus operator faces certain short run constraints on the provision of transit services. If there is a rapid and unanticipated increase in passengers on one route, the firm cannot adjust by placing additional buses on the route if these are unavailable in its fleet. If it takes a prolonged period, say four months, to order new buses and take delivery from the manufacturer, then the firm may be constrained to deal with a short term increase on one of its routes by diverting certain vehicles from planned maintenance, renting buses from other firms in the industry with unused capacity, and increasing the number of drivers in its weekly schedule to deal with increased ridership. To the extent that these measures enable the firm to deal with a temporary increase in ridership, the firm still needs to evaluate the source of the increased demand and determine whether it may be permanent. Additionally, evaluating the cost structure involved in its temporary ameliorative, the firm will likely face an increase in costs for each additional bus route scheduled, especially if it rents vehicles from other firms in place of using its own. If the firm anticipates a long term shift in ridership, then it will want to invest in new buses in order meet higher demand at a lower cost per scheduled route. This scenario demonstrates the problem of short run adjustment to changing market conditions for firms that face constraints on increasing certain factors in the short run. In the long run, the bus operator may be maximizing profits/minimizing costs like a Walrasian/Paretian firm with continuous factor substitutability and constant returns to scale. In the short run, however, it cannot adjust at least one critical factor, the scale of its fleet of buses. Consequently, as it tries to adjust to new short run market conditions, it incurs increasing marginal costs, as its variable factors exhibit diminishing marginal productivity relative to the fixed factor.
In certain respects, our bus case is more complicated than the general model that I intend to elaborate in this section. There, the entrepreneur is selecting between different means of adjustment among higher cost alternatives (renting buses from other firms) to increase services if new physical capital (new proprietary buses) cannot immediately be incorporated into production to maintain an optimal profit maximizing combination of production factors. In our basic elaboration, we will deal with a two-factor, labor and capital model, in which labor will be considered variable in the short run and capital fixed. Otherwise, we will proceed with the same sort of theoretic elaboration involved in our Walrasian/Paretian theory of the firm, developing theoretic tools to analyze physical productivity, cost structures, and market engagement from the insular frame of an individual firm in competitive conditions.
The Short Run Production Function
Our understanding of the production function in Marshallian theory will technically mirror its equivalent in Walrasian/Paretian general equilibrium economics, with the noteworthy exception that here we will have to account for the effect of holding factors fixed in the short run. Thus, in a Walrasian/Paretian economy, where agents in possession of perfect information on technology and market conditions can instantaneously adjust to changes, we represent production through the following function:
Where the quantity of commodity i is represented as a function of the quantities of labor l and capital k utilized in its production. For the Marshallian firm, in the short run, the quantity of capital available to the firm in the short run is fixed. We, therefore, represent the firm's production function as:
Where k-bar denotes a fixed quantity for capital. Thus, for the Marshallian firm, the entrepreneur is constrained, in the short run, to vary quantities of labor in response to changing market conditions, increasing labor services utilized if demand from consumers increases and decreasing labor services utilized if demand from consumers decreases.
As we increase the use of labor services against fixed quantities of capital, we presume that, to some degree, the additional productivity derived from adding each additional labor hour will decline. The extent of this decline in the marginal productivity of labor services will be reflected by the firm's short run production function. To conceptualize this, reflecting back on the process of profit maximization for Walrasian/Paretian firms, let us say that we are dealing with a firm that has a suitably constructed production function with the desirable properties from our previous theoretic approach (e.g. continuous substitutability/differentiability of the production function, linear homogeneity/homotheticity). Under relatively stable output and factor market prices, these properties enable our firm to arrive at a profit maximizing combination of production factors that will hold for all respective scales of output, absent changes in relative prices in factor and output markets. Figure 2-1 represents two output scales at two total cost levels represented by isocost curves IC1 and IC2 under a profit maximizing combination (ki*, li*), defining expansion path EP*.
Figure 1: Profit Maximizing Combinations of Labor and Capital for Firm Operating with Full Factor Substitutability
The problem, in forging a transition from the theoretic vision of Walrasian/Paretian general equilibrium to a Marshallian economy is that our firm cannot instantaneously shift from point A to point B (or vice versa) in response to changes in consumer demand in its market. In a Marshallian world, capital is not ready, in place, to be utilized to transition between production scales. It takes time to invest in new capital and to bring such investments on line for production at higher scales of operation for a firm. Our bus operator needs time to bring in new buses. Our producer of remanufactured master brake cylinders needs time to invest in new boring machines/lathes and bring them on line for production. If our feedlot needs to construct new pens to accommodate an increase in the scale of its feeder cattle stock, such an investment takes time.
Assuming, in this regard, that the firm in figure 1 is stuck, in the short run, at capital level ki1*, any change in output demand can only be addressed by increasing the quantity of labor services committed to production. Rather than bring new machinery on line, our brake cylinder remanufacturer can only add more labor hours with existing machinery to increase the quantity of master cylinders ready for sale to satisfy increasing demand. As he does so, his costs for each additional unit of output must rise. We can think of this increase in marginal costs in multiple different ways. He may simply be working his existing staff longer hours, during which time the productivity of each worker declines as they work the last few additional hours of the day. He may be hiring new workers who are less experienced at the job of remanufacturing master brake cylinders than his regular staff. Whatever causes the additional productivity of workers to decline as he raises the total number of labor hours committed to production, it is becoming more expensive to produce each brake cylinder as quantities increase beyond what the firm was capable of producing at its optimal factor combination A. Figure 2 attempts to come to terms with this decline in the marginal productivity of labor services and rise in marginal costs as we move from point A to point C.
Figure 2: Short Run Adjustment to Increased Output Quantity by Firm with Fixed Capital Constraint
Elaborating on figure 2, in relation to the conception of production functions that we have already developed in examining Walrasian/Paretian firms, the increase in output quantities represented by a shift from isoquant I1 to I2 would satisfy the firm's objective profit maximizing conditions if the firm could remain on its expansion path EP*, effecting, in this regard, a shift from point A to point B. Our point here is that, if such a change is entirely conceivable in a timeless and spaceless Walrasian/Paretian world, then it is inconceivably for the world in which Marshallian firms operate. Being stuck at capital quantity k1*, the firm in consideration can only hope to transition to output xi2 by increasing its employment of labor services to li2' to arrive at point C. Point C's location on IC2', at a total cost level/isocost curve that intersects its relevant isoquant curve at two points, indicates that the combination of production factors utilized here is suboptimal - the firm could produce the same quantity of outputs at a lower cost if it could obtain a larger quantity of capital and could employ fewer labor services. At point C, the absolute value of the slope of the isocost curve IC2' (i.e. the relative price of labor in terms of capital, Pl/Pk) exceeds that of isoquant curve li2' (the marginal rate of technical substitution (MRTS) of labor into capital, MPl/MPk), meaning that the additional productivity of labor at the margin of employment relative to that of capital is less than the relative price of labor in terms of capital. This implies that the last dollar spent renting labor earns less additional output for the firm than if it could spend the same dollar renting additional capital, which it is restricted from doing in the short run. Effectively, as the firm diverges from its long term profit maximizing expansion path to adjust to short term market conditions, the marginal productivity of its variable factor(s) diminish relative to their fixed factor(s). We can better illustrated this situation, for the firm in figures 1 and 2, by depicting the firm's short run production function in output-labor space as a two-dimensional curve, defined as a function of variable labor quantities and a single, fixed quantity of capital. We can label such an illustration of the production function in two-dimensions with fixed capital a total product curve. This is illustrated in figure 3.
Figure 3: Short Term Production Function (Total Product Curve) with Fixed Capital ki1* and Variable Labor
As drawn, this function has a number of quintessentially Marshallian features related to the average and marginal products of labor. The marginal product of labor can be represented by tangent lines to the curve, reflecting the instantaneous rate of change of output as we move along the surface of the curve. With this in mind, addition of the first units of labor demonstrate a pattern of increasing marginal productivity. Intuitively, if we have a fixed quantity of capital ki1*, represented by some mass of tools and equipment needed to undertake a production process, to which we begin adding units of labor, then the first units of labor added to this mass of capital equipment must provide a surge of output from zero to some positive quantity. As we proceed down the curve, however, the marginal productivity of labor must peak and then decline continuously. Again, if we continue to add additional units of labor services to a fixed mass of capital equipment, at some point, the additional product that we get from adding more labor must start to diminish.
The other relevant feature of the curve concerns the average product of labor and its relationship to the marginal product. The average product of labor can be represented graphically by drawing rays from the origin, the slopes of which equal the average product of labor at each point where the rays intersect with the production function. The average product of labor attains a maximum value along a ray that intersects the production function at a single point of tangency. Such a ray is drawn in figure 3, intersecting with the production function at the point A. At this point, the average product of labor equals the marginal product of labor. The reason for such an equality should be intuitive. If additional quantities of output from the rental of additional units of labor continue to exceed the average product of labor, then, mathematically, they must force the average product to increase. As soon as additional quantities of output are less than the average, the average must itself decline. Assuming the production function itself is continuously defined such that discrete additions of labor services can always generate a definite output quantity, and given the initial increasing range of marginal productivity and its subsequent continuous decline, it must be the case the average product of labor reaches its maximum where average product is equal to the marginal product. In figure 4, I have attempted to represent the average and marginal products of labor in relation to our figure 3 short run production function, adding point E to represent an inflection point in the production function where the marginal product of labor reaches its maximum value and starts to diminish.
Figure 4: Marginal and Average Products of Labor Services in Relation to Short Run Production Function with Fixed Capital
The representations of the average and marginal products of labor in figure 4 definitively attempt to convey the idea that, as we had discrete quantities of labor to a fixed capital base, we initially achieve complementarities between production factors that raise marginal productivity as we add successive discrete units of labor to production. Such complementarities must, however, be exhausted at some point, evident in the upper panel at point E. Beyond this point, marginal productivity diminishes with each additional unit of labor rented by the firm. This pattern of marginal productivity similarly mandates that the average productivity of labor must initially rise, reach a maximum level (at which average and marginal productivity are equal), and decline monotonically. Thus, graphically, the marginal and average productivities of labor are represented as inverted "U"-shaped curves.
In reference to our short run production function, it would be entirely valid to conclude that, at some point, the marginal product of labor becomes negative, indicating that progressive addition of labor hours to a production process with a fixed capital base will cause diminution of total output beyond some maximum value. Thus, it might be the case that a firm hires so many employees in the short term to deal with increased demand that they end up getting in each others way, ultimately diminishing the total production that might have been produced by a smaller number of employees. As such, we may confront a larger definition of capacity utilization, in relation to diverse fixed factors, where saturation of a work place with superfluous variable factors becomes an actual impediment to increasing outputs to meet demand. On the contrary, I am content to allow the production function to reach an output limit, beyond the range that I have illustrated, at which the marginal product of labor reaches a zero value and average productivity approaches zero asymptotically. In any case, the larger point here is that production functions in the short term reach limits, shaped by the diminishing marginal productivity of variable factors against fixed factors. The particular contour of short term production functions, moreover, shapes the contours of short term marginal and average cost curves for Marshallian firms and reflect demand functions for variable factors.
Having defined the marginal and average product schedules for labor, as the firm's variable production factor, we can further relate the firm's marginal product schedule for labor to its demand for labor in factor markets. In doing so, we will make the initial assumption that the firm operates as a price taker in a relatively competitive factor market context. If it does so, then, as a profit maximizing entity, the firm will seek to rent successive units of labor and capital to maximize the return that it receives from each successive unit of each factor subject to both the factor market price for the factor and the price of the outputs produced by the factor. This reasoning simply reiterates the general conditions for profit maximization that we introduced for the Walrasian/Paretian firm on the theme of product exhaustion. In that context, we treated both labor and capital as variable factors, against which the mutual determination of factor and output market prices enabled us to determine an objective profit maximizing/cost minimization combination of factors realizing product exhaustion. Here, however, the capital available to the firm is fixed in the short run. Thus, the firm can only choose quantities of labor in order to maximize its profits and minimize costs relative to changes in the output or factor market prices.
With these limitations on profit maximization and the necessity of adjustment to short run market conditions in mind, Marshallian firms face a somewhat different profit maximizing condition. In the short run, Marshallian firms will rent labor services from households up to the point at which the last unit of labor employed returns an additional/marginal value product equal to its compensatory rate, determined by the equilibrium wage in the factor market for labor services. We define this value product as the marginal revenue product of labor services (MRPl), calculated, as previously defined in regard to the Walrasian/Paretian theorem of product exhaustion, by multiplying the marginal product of labor by the output price of the commodities produced by labor. Thus, short run profit maximization by Marshallian firms maintains the following condition:
We can, further, qualify this condition with respect to our arguments concerning the short run marginal productivity of labor for Marshallian firms constrained by fixed quantities of capital. To the extent that Marshallian firms operate with marginal productivity schedules for labor characterized by ranges of increasing and diminishing marginal productivity, the short run labor demand schedules of Marshallian firms are represented only by the range of marginal revenue productivity values where values are less than or equal to average labor productivity and diminishing. This range is represented graphically in figure 5.
Figure 5: Marginal and Average Revenue Product Curves with Labor Demand Subset Accented
Intuitively, the reason why the firm should only demand positive quantities of labor services in this range, under the condition that it pays labor a wage equal to its marginal revenue product, is because it is only in this range that the firm can receive, from each additional worker, a return that will equal or exceed the wage rate that it pays in order to hire the last additional worker. For all preceding points on the marginal revenue product curve, the average revenue product of labor is less than the marginal revenue product - the average return from hiring labor must be less than the compensation rate paid to workers.
Before concluding this sub-section, I want to elaborate to an element of continuity between Walrasian/Paretian and Marshallian conceptions of the production function. My initial approach to the short run production function for Marshallian production functions attempts to make use of the Walrasian/Paretian architecture of isoquant and isocost curves, instruments largely superfluous to an explanation of production in the theory of Marshallian firms, at least in the short run. In arguing graphically, moreover, that the short run Marshallian production function reaches an equality of marginal and average productivity of labor at a point at which we realize an objective profit maximizing/cost minimizing combination of production factors, consistent with Walrasian/Paretian production analysis, I have sought to contextualize Marshallian theory in relation to the Walrasian/Paretian approach. For Walrasian/Paretian firms, we concluded that, for every conceivable scale of output, the firm, characterized by a linearly homogeneous production function, will operate at a combination of production factors where compensation of each production factor in accordance with its marginal productivity will precisely exhaust the total revenue product of the firm. As such, the marginal products of each factor will be proportional to their average products.
For the Marshallian firm, we concluded that short term production is apt to diverge from an objective profit maximizing combination of factors because the firm is constrained in its ability to rent certain factors that are relatively scarce for certain periods of time. Recognizing this point, it is conceivable that Marshallian firms would operate with constant returns to scale, in the same manner as Walrasian/Paretian firms, but this is a special case. The Marshallian firm constitutes a more generalized production analysis, in which certain firms will be characterized by global decreasing returns to scale, others by global increasing returns, others, like Walrasian/Paretian firms, global constant returns, and still others, characterized notably by Viner's formalization of Marshallian production analysis, by discrete regions with increasing, decreasing, and constant returns to scale. Such differentiations and their larger consequences in production analyses will have to await a long term portrait of the Marshallian firm, but, more generally, they are formative to the more complex macroeconomic portraits emerging from Marshallian theory, especially along the developmental progression of Keynesian analyses.