INTRODUCTION 100 In This Chapter CHAPTER 5 Management Decisions: How to Produce? Thus far we have examined the production of one commodity where only one in- put was variable and all others were fixed at some amount. The question we were addressing was “How much to produce?” Instances of only one variable input are indeed rare. In reality firms use more than a single-variable input to produce their products, which leads to different—and more complex—decision-making criteria. This chapter addresses the second of our producer questions: “How to pro- duce?” This approach reflects the existence of more than one variable input in the production activities of a firm or farm and emphasizes the question of substitution of inputs in the production process. Because most products can be produced with varying quantities of inputs (think of how many combinations are possible in pro- ducing beef), different input combinations have different costs associated with them, Hence, the manager must decide which combination of inputs to use in or der to achieve the firm’s economic objective. This decision-making process is called the factor-factor decision and is slightly more complex than the factor-product decision. For example, when cleaning the barn on a farm, the farmer can use dif- ferent combinations of labor and capital. A tractor and loader with only one driver (high capital /low labor) or a large number of people with pitchforks (low capital, high labor) are two possible alternatives. Once the price of each is known, the low- est cost alternative can be chosen. A term often associated with factor-factor deci- sions is cost minimization, or, given that a certain level of output is desired (clean ing one barn every Saturday), what combination of inputs will produce that output at the lowest cost? The objective of this chapter is to introduce the basic decision rules appropriate for al i isi js proper managerial decision making and economic analysis in situations where te firm combines multiple variable iny . u output. Since the inputs are combi puts in the physical production process of its ined, it is possible, to a certain extent, to substitute one input for another in the production process. For example, hay ‘versus con- centrate in feeding livestock, fertilizer versus amount of seed per acre in crop Pr duction, or capital versus labor in manufacturing, When multiple inputs are avail- able, a firm or producer must decide which combination of variable inputs to use with the fixed factors to achieve the firm’s production objectives. We assume the firm will act as a rational producer, therefore the production objective is to’ combine the variable inputs in the proportion that results in the highest profit. Most prod- ucts can be produced using varying combinations of inputs, and each input com- bination will have a cost associated with it. Therefore, the decision regarding which combination of inputs to use to maximize profits also involves deciding which combination of inputs will minimize cost. Assumptions Used The factor-factor decision involves finding the least-cost combination of inputs, from the different input combinations possible, that will produce a given level of ouitput. To examine this decision, we use the following assumptions: 1. The firm produces a single product. 2. The firm has two variable inputs, others are fixed. 3. The firm is a price taker. Asin previous chapters we are assuming all decisions are made under perfect cer- tainty and we are examining a single product production function. However, as as- sumption 2 states, the firm now has two variable inputs, which can be combined in different ways, to use with the fixed factors in producing this product. As in Chapter 3, the firm has no influence over the price it pays for inputs; that is, the firm is a price taker. To examine the factor-factor decision, we begin by developing a measure- ment or tool to identify how one input substitutes for another within the firm’s pro- duction process. Then we determine what combination of two variable inputs to use to maximize profit at each level of output, recognizing that for a given level of output, maximizing profit and cost minimization are the same goal. Finally, we de- termine some decision rules to identify the most profitable level of output using two variable inputs, As we will see, the decision rules developed in this chapter can be used in the same conceptual manner for more complex situations (ie., those with more than two variable inputs). PHYSICAL RELATIONSHIPS We begin our discussion of the factor-factor decision by examining the physical pro- duction aspects of the decision. The firm has two variable inputs that it can combine in different combinations along with its fixed factors of production to produce its Chap.5 Management Decisions: How to Produce? 101102 product, Notationally, we are now using a production function that has two variably inputs, X, and X2, with the other factors fixed at some ievel. Thus the firm's produg. tion function is expressed as: Y = fiXr, Xp | Xgy.- Xn) where Y is the product, or output, produced; X, and X; are the variable inputs; ang X;,...,X, are the fixed factors of production, “In determining a measurement for describing how the two inputs physically substitute for each other, we need to introduce two new concepts. The first is a pro. duction surface; the second is a theoretical tool called an isoquant. Production Surface . The concept of a production surface describes the “hill” of increased production ob- tained as we increase useof the pariable inputs, This is nothing mioré than a production function in three dimensions (ie, Y, X;, and X,). This concept is graphically de- picted in Figure 5.1, where the height of the surface is the ‘production | level. As we ‘move up the production surface the level of production is increasing. If we were to stop at any pojnt on the production surface (or hill) and walk around it at the same glevation the production level would be the same at every point and we ‘could lentify all the possible input combinations that would produce that production level. This is illustrated in Figure 5.1 by lines A, B, and C (later we will learn these are isoquants). The production level at ine B is greater than at lineA, and the pro- duction level at C is greater than the production at both lines B and A. Why? Because C is higher up on the production surface, or at a higher elevation, than lines B or A, and as we move up the production surface production levels are in- creasing. At each point along line B the output level remains constant, or the same. Thus, as we move along line B, we can identify all the combinations of the two vari- FIGURE 5.1 . Production surface with two variable inputs, Part Tools for Decision Makingable inputs, X; and X,, that will produce this output level. We can do the same for production levels A and C, or any other production level on the production sur- face. Using the production surface allows us to identify all the combinations of the two variable inputs that will produce a given level of output. This knowledge will enable us to identify the costs associated with each input combination and allow us to identify the least-cost combination to produce a given level of output. Tsoquant If we identify all the combinations of X, and X, that produce exactly the same to- tal output, we would have an isoquaiit (sometimes called an 4soproduct), An iso- quantis icating all cot of two variable i cea givery vo? constant, level of output. Iso means “the same”; hence, isoquant means that the same quantity of output is reflected everywhere along the line, which exactly de- scribes lines A, B, and C in Figure 5.1. Thus, lines A, B, and C are isoquants. Deriving an Isoquant An isoquant identifies all the combinations of two variable inputs that will pro- duce the same level of output. To illustrate how to derive an isoquant, consider the following example. A dairy farmer can combine grain (X;) and hay (X) for his herd’s feed ration. From past production records the dairy farmer obtains the com- binations of the variable inputs (hay, grain) that will produce 11,000 Ibs of milk, as. listed in Table 5.1. Each combination of grain and hay shown produces the same level of output, thus they are all on the same isoquant. For example, the first com- bination (720 Ibs grain, 600 Ibs hay) produces 11,000 Ibs of milk as does the third combination (600 lbs grain, 700 Ibs hay) or the fifth combination (530 Ibs grain, 800 Ibs hay). As indicated in Table 5.1, if we increase the use of one variable input (hay), the other variable input can be decreased (grain). That is, hay is substituting for grain in the production of milk. To derive the isoquant graphically, we simply plot the input combinations, as illustrated in Figure 5.2. The downward slope of the isoquant indicates that grain TABLE 5.1 Input Combinations Producing 41,000 Pounds of Milk ee x, X (Pounds of Grain) (Pounds of Hay) 720 "600 650 650 600 700 560 750 530 800 510 850 500 900 _ Chap.5 Management Decisions: How to Produce? 103(wei) FIGURE 5.2 900 X. An isoquant for 11,000 (hay) pounds of milk. |= 19,000 Ibs 1,= 11,000 Ibs = 9,000 tbs FIGURE 5.3 Isoquants for alternative output levels of rik, (%) can be decreased, or substituted for, by increasing the use of hay (X,). Economic convention (the way we usually do things) places the input being re- placed, grain in this example, on the vertical axis (labeled X, 1), and the input being added, hay in this example, on the horizontal axis (labeled X,). The designation of which input is to be replaced and added is arbitrary, but is important in terms of specifying the marginal rate of substitution (MRS) and the inverse price ratio (IPR), terms we discuss later. ~ ~__ The isoquant concept is commonly used in conomic analysis, The isoquant is used not only to indicate the different combinat mbinations Of two injpufS capable of pro- ducing a given level of output, but it can represent man’ ny levels of output : Since each isoquant represents a unique output level, theoretically there is an iso. quant for each level ofoutputa firm is capable of producing, as illustrated in Figure 53. This figure shows that the distance away from the origin indicates the level of output, or the height of the production surface. The closer the isoquant is to the ori- gin, |, the lower the level of output. The farther away from the origin; Ty, the higher 104 Part -Tools for Decision MakingFIGURE 5.4 | intersecting isoquants (this is not possible; see % X_ discussion in text). the output level. The use of distance from the origin as an indication of higher pro- duction arises because of the difficulty of portraying production functions in three- dimensional space. Because each isoquant represents a uniqueGaiput lev) isoquants can never intersect. Why? If isoquants intersected, it would mea that a single combination of inputs would be capable of producing more than one level of output. This is de- picted graphically in Figure 5.4. If the isoquants Ip and I, intersected, as in Figure 544, it would mean that the input combination X> and ‘X¥ could produce two dif- ferent levels of output. Given the unique relationship between an input use com- bination and the resultant output in the physical production process, this is not possible. Marginal Rate of Substitution ‘As you can see from Figure 5.2, as we move along an isoquant the output level re- mains constant but the quantities of the variable inputs are changing as one input replaces, or substitutes, for the other. As managers we are interested in how well the inputs substitute for each other. ‘The rate at which one variable input can physically substitute for another variable input inthe physical production process of the firm is mea- sured by the marginal rate of substitution (MRS); also referred to as marginal rate of technical substitution (MEPS) in some textbooks. : The marginal rate'of substitution describes the fate at which oné input can be decreased as use of the other input incréases and, by definition; is the slope of the economic convention places the input beirig replaced on, the verti-, ied MRSis calculated by 4i- viding the change in the replaced input (A rep! e change in the added in- put (A added). Thus the computational formula and the, terpretation of MRS depend on which inj iis being replaced and which is being Kafr X,y written For exainple, 1% 4s the variable inpit being réplaced’ Chap. 5 Management Decisions: How to Produce? 105 e input being added, then’the marginal tate of substitution106 (input replaced) by the AX, (inpy, MRSx,x i calculated by dividing the AX, added), or -A\_ AX, _ replaced wegl)= AX, added variable input being replaced, X; in this rahe variable input being aided, Xin this case, ison Figure 5.5 is the change in X, gj. soquant in which corresponds exactly to the definy Graphically, as shown in Figure 5.5, th case, is on the vertical the horizontal axis. The slope of the i vided by the change in Xp (AX, + 4X2), tion of MRSy,x- Convertaly, if Xp is the variable input being replaced and X: is the variable input being added, then the marginal rate of substitution of X; for Xz, written MRSy.xy is calculated by dividing the AX; (input replaced) by the AX, (input added), or Graphically, as shown in Figure 5.6, Xz, the variable input bei i P ly, 4 6, put being replaced, is o1 vertical axis and X,, the variable input being added, is on the arena ome FIGURE 5.5 Marginal rate of substitution of Xp for X, FIGURE 5.6 Marginal rate of substitution of X; for X, Part I Tools for Decision Making:slope of the isoquant in Figure 5.6 is equal to the change in X; divided by the change in X, (AX + AX;), which corresponds exactly to MRSx, x2 ‘Note that the'subscripts on the definition of MIS match the axes labels on the graphs in Figures 5.5 and 5.6. Since the order of the subscripts on ‘MRS tells us which input is being added and which input is being replaced, itis very important that the graphical presentation be consistent with stated MRS, and vice versa. Again, con- ventionally, we usually put X; on the vertical axis but we do not have to. We indicated and illustrated earlier that the designation of which input is to be replaced and which input is to be added is an arbitrary designation. This is because MRSy,x, = / MRSx,x, SN Asillustrated, MRSx,, is the inverse of MRSx,x, thus, deciding which input is re- placed and which input is added for analysis isa management decision. Although they represent the same information, it is important to be aware of which input is being replaced and added for computation and for correct graphical presentation. ‘Using our dairy example, we can now illustrate how to calculate the marginal rate of substitution numerically. As shown in Table 5.1, the variable input hay (Xz) is being added and the variable input grain (X;) is being replaced. Therefore, the correct formula to calculate the marginal rate of substitution between these inputs is MRSx,x, = AX;/AX;, as shown in Table 5.2. The formula, which states the rate at which X, (is added) substitutes for X1 (is replaced), is equal to the change in X, divided by the change in Xo, and, as ile lustrated in Figure 57, is the slope of the isoquant. Because one input can be de- creased by adding the other input, MRS as a numerical coefficient is always nega tive, as indicated in Table 5.2 and Figure 5.7. The MRS. of —1.40 (—70/50) as we go from the first input combination (720, 600) to the Second combination (650, 650) in- dicates that 1.4 units of grain (Xj) are replaced as we add one unit of hay (X_). This —1,40 also describes the slope of the isoquant at this point, since, by definition, MRS measures the slope of the isoquant, As shown in Table 5.2, initially hay (X2) substitutes very well for grain (X;) in the feed ration because’a one-unit increase in Xallows a'greater than ‘one-tnit decrease in use of X,YLater, as decreasing returns (remember our discussion from Chapter 2) affect the trade-offs, the two.inputs sub- stitute less feel). ‘As we substitute hay for grain in our dairy example, the MRS de- creasés*in magnitude (in terms of absolute alue). Thus, an isoquant exhibits di- minishing MRS. Graphically, as in Figur ) it can be stated that when the “soquant has a steep slope (MRS > 111, where 111 is the absolute value of 1) the Chap. 5 Management Decisions: How to Produce? 107EE 108 te of Substitution of Hay (%2) for Grain (X) | Rat ing the Marginal TABLE 5.2 Calculating %. x a AX + ax, a AX (added) MRS : ; ded) oe =140 J= -70/5 720, < 7 Ry ) Se 2 -1.00 = 307% ' = 40/59 a a “a S -080 = 750 50 0.60 = —30/50 560 -30 a 040 = -20/5 530 800 4 50 oe a sia = “a _ 02 = — -10/50 500 500 2 FIGURE 5.7) Isoquant and marginal x rate of substitution of (rain X%, for X;. |= 11,000 Ibs Xx, (ray) input on the horizontal axis, usually X,, is a good substitute for X;;as the slope ¢ the isoquant decreases (MRS < |1'1), Xp becomes a less efficient siabstitute for ¥ The slope of the isoquant is the MRS between the two iniputs and the subst tutability between two inputs depends on how Tesponsive production is | changes in each input, or in the productivity of the individual inputs. Possible Rates of Input Substitution The example developed for the dairy producer had inputs that were imperfect st stitutes for each other, or the isoquant exhib ited diminishing marginal rate of st stitution. Diminishing (IRS ‘occurs where ae one unit of an input can be exthang Part 1. Tels fr Decision Makingre 6 for another, but at a decreasing ate due to diminishing returns in the production function. As illustrated in Figure 5.7, initially Xp substitutes rather freely for Xila large decrease in graih (X,) for a small increase in hay (X,)]; but as decteasing re- turns affect the trade-off between X, and X2, the two inputs substitute less freely [a small decrease in grain (X;) for a large increase in hay (X2)]- Two other possibilities for substitutability between inputs exist, as discussed later. Constant, or perfect, substitutability occurs when one unit of an input can be exchanged for another input on a consistent basis of one-to-one or some other un- changing ratio. An example of nearly perfect input substitutes is nitrogen (N) fer- tilizer use on corn. A producer can get a specified amount of actual N on com Wi ing ammonia nitrate (33% N) ot anhydrous ammonia (84% N). The producer's choice is determined by personal preferences, soi types, availability, and other fac- tors but in any case, the rate of substitution is almost linear. In this constant case, diminishing marginal rate of substitution does not occu, and the isoquant is lin- ear, as illustrated in Figure 5.8(a). Fixed proportions (or perfect complementarity) occurs when inputs must be used ina fixed ratio, There is no choice regarding what proportion of each input to use. For example, every time a truck is needed, a driver is also required. Insucha case, the substitutability is zero, thus the isoquant is a'single point as illustrated in Figure 5.8(b). Theither of these cases, constant substitutability or fixed proportions, the role of the manager is not very active since litle choice is available. The most com monly seen relationship between inputs is one of diminishing MRS. Why? Because MRSis the ratio of the marginal products of the two inputs, and decreasing returns are evident in most production functions (i., the law of diminishing returns; see Chapter 2). One application for isoquants is to illustrate the options available to producers in the Corn Belt should nitrogen fertilizer use be restricted in the face of groundwater x % Perfect Fixed substitutability proportions % (a) (b) FIGURE 5.8 Constant (perfect) gyestitutes and fixed proportions isoquants. Chap.5 Management Decisions: How to Produce? 109TTT ECONOMIC RELATIONSHIPS 110 i Bushels of £53. Cropland and Nitrogen Combinations to Produce 14,000 Com ABLE 5: = Total Corn Production is Nps ee Acres Planted Be oo _ ne 14,000 110 ; 1000 14,000 M40 ° 14,000 153 ‘a 14,000 160 160 at the University of Minnesota el eal use the alternative combi. Frown in Table 5:3 to produce 14,000 bushels of com, com acreage, farmers could not be the out- sible for com sroblems. For example, that corn producers in s\ quality p suggests c nations of cropland and nitrogen! ors restricted nitrogen use On nded corn acreage May uld be one adjustment po: So, if environmental regulat maintain output by expanding come environmentalists are see! producers. acreage. Expat ing, butit wo Up to this point, we have only been comparing physical units of two different in- puts and how they substitute for each other in a technical, noneconomic relation- ship. To calculate minimum cost combinations of inputs we need factor prices to build on the technical relationships. Input prices are appropriate since we are ex- amining the factor-factor decision. Isocost Line Auuseful tool for ing the value of the two inputs is an isocost line, which is defined as g line all combinatic two variable inputs that can be purchased _fora given or same, Tevel of expenditure. It is calculated by ad thi none; spent on each input, or parce gy pet: a ee a : an teevel ly i= xB) +, 0%) . a This the total dollar expenditure on inputs (price of each input multiplied s of each input), P,, and P,., are the prices of the variable i and X,are the units of the vatfable nats purch variable inputs, and X has : i as an fe expenditure line in cca ee peo money e bes pel doa Eat that can be purchased depends on the amount of ae es ra ay) and the respective prices of the inputs. Going back carrie ain i ‘a A aie ee only has $100 to purchase inputs, bar he . and hay for $0.10 Z yut could he ay .10/Ib, how i in uld be pur Part Tools for Decision Making,TABLE 5.4 Input Combinations with Same Total Expenditure of $100 with Price of Grain (P,,) = $0.20 and Price of Hay (P,,) = $0.10 Grain Expenditure H it t lay Expenditure _ a Grain Purchased on Hay Total Expenditure MH) ux) 00) (Pap X Xo) (Pay XX +P X Xe) . a $100 0 $0 $100 aa $80 200 $ 20 $100 300 $60 400 $ 40 $100 0 $40 600 $ 60 $100 $0 1,000 $100 $100 Xx ) 1000 500 TE = $200 FIGURE 5.9 0 a 2000, _ {8000s nes for $100 and $200 total % expenditure. chased for $100 are listed in Table 5.4. Each of the combinations listed requires the same level of expenditure. For example, $100 would purchase the first combination (500 Ibs grain, 0 Ibs hay) or the third combination (300 Ibs grain, 400 Ibs hay) or even the fifth combination (0 Ibs grain, 1,000 Ibs hay). Thus, with $100 to spend on inputs, the dairy farmer could use any of the grain/hay (X1, X,) combinations given in Table 5.4. , The isocost line for, $100 is depicted graphically in Figure 5.9. If the dairy farmer spent the entire’$100, on(Xjhe could purchasq00)units of X, ($100 + $0.20 = 500). If he spent the entire $100 on X, he could purchase 1,000 units of X, ($100 + $0.10 = 1,000). These points are the end points, or intercept terms (points), of the isocost line. Connecting the two endpoints yields the isacost line (TE = $100 in Figure 5.9) for $100. Any combination of inputs on this line requires the same level of total expenditure ($100) to purchase. Thus, all of the combinations listed in Table 5.4 lie on the isocost line for $100 (TE = $100) in Figure 5.9. . If the dairy farmer has $200 and only purchases X, he can purchase 1,000 units ($200 + $0.20 = 1,000); ifhe purchases only X,, he can purchase 2,000 units ($200 = $0.10 = 2,000). The isocost line for $200 (TE = $200) is also drawn in Chap, 5 Management Decisions: How to Produce? 11.iad ili 112 1 indicates the level of ey, tal expenditure ang lower the level of tot : the farther away from the origin the higher i hased. ore inputs can be puree the ae number of isocost lines could have different expe! Figure 5.9. As trated, he distance fr m the origi Hlustrated, the ¢ om the orig) . jture. The closer t endi jnputs can be pul the Feet pal expenditure and el of total me ee you are probably aware 00 Soe amount For Figure 59 an : done for $200, ae pases we have drawn only a bers a pans eter " Svat ine drawn in Figure 5.98 ~ a. The slope of the 0605 ered from the isocost line equation: toas the inverse prise ratio. Itis BrP.) \ wwe solve the equation for %- + (Px, X Xo) ’ First we subtr Since X; is on the vertical axis, i act (Py, Xo) from both sides: (x X Xa) = TE~ Pa X2) and then we divide both sides by P,, to obtain TE Py, Jaen fl ee xX which is the familiar formula for a straight line Y = a + bX, where the intercept term (a) in this case is'TE/P,, (which identifies the X intercept on the X; axis) and the slope (b) is ~P,,/P,,- The slope of the isocost line is called an inverse price ra- tio (PR) because itis the pice ofthe imput on the horizontal axis divided by the price of the input on the vertical axis. (Later on we'll have another good way to use this term.) ‘Thus, the way the graph axes are labeled will dictate which input price is in the nu- merator and denominator of the price ratio. The slope of the isocost line tells us the relative value of the inputs, or how they substitute for one another economically; this information will allow us to identify the least-cost combination we've been looking for. Effects of a Price Change Because the slope of the isocost line depends on the relati is is E : D telative price ratio of the two inputs, if the f i i =. price of one of the inputs changes, the slope of the isocost line also ot gue i illustrates the impacts ofa change in Py . Recall our previous ex ample where TE = $100,P,, = $020, and P., = $0.10-At these prices, if the dairy Xuhe al Bae he eae purchase 1,000 units; or if he purchased only u ‘ uni ‘o in Figure 5.10). If gs number of units of X, that he can purchase with ste oe ree so INC Part I” Tools for Decision Making