Documente Academic
Documente Profesional
Documente Cultură
CHAPTER OUTLINE
7-1 7-2 7-3 7-4 7-5 7-6 The Nature of Costs Short-Run Cost Functions Short-Run Total and Per-Unit Cost Functions Short-Run Total
and Per-Unit Cost Curves
Long-Run Cost Curves Long-Run Total Cost Curves Long-Run Average and Marginal
Cost Curves Case Study 7-1: The Long-Run Average Cost Curve in Electricity
Generation
Trade in Inputs Case Study 7-4: The IBM PC and the Boeing 777 and 787 Are All but Economies of Scale Immigration of Skilled Labor
Logistics
at National
Cost-Volume-Profit
Analysis
Nano Car
Summary Problems
262
CHAPTER7
263
KEY TERMS
ExpLicit costs Implicit costs ALternative or opportunity costs Economic costs Accounting costs ReLevant cost IncrementaL cost Sunk costs TotaLfixed costs (TFC) TotaLvariabLe costs (TVC) TotaLcosts (TC) Average fixed cost (AFC)
Average variabLe cost (AVC) Average totaL cost (ATC) MarginaL cost (MC) Long-run totaL cost (LTC) Long-run average cost (LAC) Long-run marginaL cost (LMC) PLanning horizon Economies of scope Learning curve Foreign sourcing of inputs New internationaL economies of scaLe
Brain drain Logistics Cost-voLume-profit or breakeven anaLysis Contribution margin per unit Japanese cost-management system Operating Leverage Degree of operating Leverage (DOL) Engineering technique SurvivaL technique
e saw in Section 1-2 that the aim of a firm is generally to maximize profits. Total profits equal the positive difference between total revenue and total costs. The total revenue of the firm was examined in Part Two of the text, which dealt with demand analysis. In this chapter we examine costs and their importance in decision making. The firm's cost functions are derived from the optimal input combinations examined in the preceding chapter and show the minimum cost of producing various levels of output. Clearly, cost is an important consideration in managerial decision making, and cost analysis is an essential and major aspect of managerial economics. The chapter begins by examining the nature of costs of production. These include explicit and implicit costs, relevant or opportunity costs, and incremental costs. We then derive the firm's short-run and long-run total, average, and marginal cost curves. After that, we examine plant size and economies of scale, economies of scope, and the learning curve. Subsequently, we discuss international trade in inputs and the immigration of skilled labor, as well as logistics or supply chain management. Finally, we discuss breakeven analysis and examine the empirical estimation of cost functions.
7-1
264
(say, in managing another firm), and the highest return that the firm could receive from investing its capital in the most rewarding alternative use or renting its land and buildings to the highest bidder (rather than using them itself). In economics, both explicit and implicit costs must be considered. That is, in measuring production costs, the firm must include the alternative or opportunity costs of all inputs, whether purchased or owned by the firm. The reason is that the firm could not retain a hired input if it paid a lower price for the input than another firm. Similarly, it would not pay for a firm to use an owned input if the value (productivity) of the input is greater to another firm. These economic costs must be distinguished from accounting costs, which refer only to the firm's actual expenditures or explicit costs incurred for purchased or rented inputs. Accounting or historical costs are important for financial reporting by the firm and for tax purposes. For managerial decision-making purposes (with which we are primarily interested here), however, economic or opportunity costs are the relevant cost concept that must be used. Two examples will clarify this distinction and will highlight its importance in arriving at correct managerial decisions. One example is from inventory valuation. Suppose that a firm purchased a raw material for $100, but its price subsequently fell to $60. The accountant would continue to report the cost of the raw material at its historical price of $100. The economist, however, would value the raw material at its current or replacement value. Failure to do so might lead to the wrong managerial decision. This would occur if the firm decided not to produce a commodity that would lead to a loss if the raw material were valued at its historical cost of $100 but to a profit if the raw material were valued at its current or replacement value of $60. The fact that the firm paid $100 for the input is irrelevant to its current production decision since the firm could only obtain $60 if it sold the input now. The $40 reduction in the price of the raw material is a sunk cost that the firm should not consider in its current managerial decisions. Another example is given by the measurement of depreciation cost for a long-lived asset. Suppose that a firm purchased a machine for $1,000. If the estimated life of the machine is 10 years and the accountant uses a straight-line depreciation method (that is, $100 per year), the accounting value of the machine is zero at the end of the tenth year. Suppose, however, that the machine can still be used for (i.e., it would last) another year and that the firm could sell the machine for $120 at the end of the tenth year or use it for another year. The cost of using the machine is zero as far as the accountant is concerned (since the machine has already been fully' depreciated), but it is $120 for the economist. Again, incorrectly assigning a zero cost to the use of the machine would be wrong from an economics point of view and could lead to wrong managerial decisions. In discussing production costs, we must also distinguish between marginal cost and incremental cost. Marginal cost refers to the change in total cost for a l-unit change in output. For example, if total cost is $140 to produce 10 units of output and $150 to produce 11 units of output, the marginal cost of the eleventh unit is $10. Incremental cost, on the other hand, is a broader concept and refers to the change in total costs from implementing a particular management decision, such as the introduction of a new product line, the undertaking of a new advertising campaign, or the production of a previously purchased component. The costs that are not affected by the decision are irrelevant and are called sunk costs.
CHAPTER7
265
[ 7-2
SHORT-RUN
COST FUNCTIONS
In this section we distinguish between fixed and variable costs and derive the firm's total and per-unit cost functions. These cost functions are derived from input prices and the optimal input combinations used to produce various levels of outputs (as explained in the previous chapter).
Within the limits imposed by the given plant and equipment, the firm can vary its output in the short run by varying the quantity used of the variable inputs. This gives rise to the TFC, TVC, and TC functions of the firm. These show, respectively, the minimum fixed, variable, and total costs of the firm to produce various levels of output in the short run. Cost functions show the minimum costs of producing various levels of output on the assumption that the firm uses the optimal or least-cost input combinations to produce each level of output. Thus, the total cost of producing a particular level of output is obtained by multiplying the optimal quantity of each input used times the input price and then adding all these costs. In defining cost functions, all inputs are valued at their opportunity cost, which includes both explicit and implicit costs. Input prices are assumed to remain constant regardless of the quantity demanded of each input by the firm. From the total fixed, total variable, and total cost functions, we can derive the corresponding per-unit (average fixed, average variable, average total, and marginal) cost functions of the firm. Average fixed cost (AFC) equals total fixed costs (TFC) divided by the
I In incremental-cost analysis, semivariable costs are often encountered. These are cost changes that arise if output falls outside some speci fied range. For example, by contract the firm may be able to reduce the salary of top management if output falls sharply or must pay bonuses for large increases in output.
266
PART THREE
level of output (Q). Average variable cost (AVC) equals total variable costs (IVC) divided by output. Average total cost (ATC) equals total costs (TC) divided by output. Average total cost also equals average fixed cost plus average variable cost. Finally, marginal cost (MC) is the change in total costs or the change in total variable costs (IVC) per unit change in output.i That is, AFC = TFC [7 -2] [7-3]
Q
AVC= IVC
Q
ATC MC
==
TC
AFC+ AVC
[7-4] [7-5]
llTC llQ
llIVC llQ
Cost Schedules (4) Total Costs $ 60 (5) Average Fixed Cost (6) Average Variable Cost (7) Average Total Cost (8) Marginal Cost
0
1
$60 60 60 60 60 60
2 3 4 5
$ 0 20 30 45 80 135
$60 30 20 15 12
$20 15 15 20 27
$80 45 35 35 39
$20 10 15 35 55
2 Since the difference between TC and TVC is TFC, which are fixed, the change in TC and the change in TVC per unit change in output (MC) are identical. In terms of calculus,
MC
d(TC) dQ
d(TVC) dQ
since
d(TFC) dQ
=0
CHAPTER 7
267
Costs ($)
TC
200 180 160 140 120 100 80 60 Total fixed costs 40 20 Total variable costs Total fixed costs
o
Per unit costs ($) 80 70 60 50 40
Output
(Q)
ATC
Ave
30 20 10
o o
1 1.5 2 2.5 3 3.5 4 5
Output
(Q)
FIGURE 7-1 Short-Run Total and Per-Unit Cost Curves The top panel shows that TVC is zero when output is zero and rises as output rises. At point G' the law of diminishing returns begins to operate. The TC curve has the same shape as the TVC curve and is above it by $60 (the TFC). The bottom panel shows U-shaped AVC, ATC, and MC curves. AFC = ATC - AVC and declines continuously as output rises. The MC curve reaches a minimum before the AVC and ATC curves and intercepts them from below at their lowest points.
268
returns is not operating. Thus, the TVC curve faces downward or rises at a decreasing rate. Past point G' (i.e., for output levels greater than 1.5 units in the top panel of Figure 7-1), the law of diminishing returns operates, and the TVC curve faces upward or rises at an increasing rate. Since TC = TFC + TVC, the TC curve has the same shape as the TVC curve but is $60 (the amount ofthe TFC) above it at each output level. These TVC and TC schedules are plotted in the top panel of Figure 7-1. The AFC values given in column 5 are obtained by dividing the TFC values in column 2 by the quantity of output in column 1. AVC (column 6) equals TVC (column 3) divided by output (column 1). ATC (column 7) equals TC (column 4) divided by output (column 1). ATC also equals AFC plus AVe. MC (column 8) is given by the change in TVC (column 3) or in TC (column 4) per unit change in output (column 1). Thus, MC does not depend on TFe. These per-unit cost schedules are plotted in the bottom panel of Figure 7-1. Note that MC is plotted halfway between the various levels of output. From Table 7-1 and the bottom panel of Figure 7-1 we see that the AVC, ATC, and MC curves first fall and then rise (i.e., they are U'-shaped). Since the vertical distance between the ATC and the AVC curves equals AFC, a separate AFC curve is not drawn. Note that AFC declines continuously as output expands as the given total fixed costs are spread over more and more units of output. Graphically, AVC is the slope of a ray from the origin to the TVC curve, ATC is equal to the slope of a ray from the origin to the TC curve, while the MC is the slope of the TC or TVC curves. Note that the MC curve reaches its minimum before (i.e., at a lower level of output) and intercepts from below the AVC and ATC curves at their lowest points. We can explain the V shape of the AVC curve as follows. With labor as the only variable input, TVC for any output level (Q) equals the wage rate (w, which is assumed to be fixed) times the quantity of labor (L) used. Thus,
[7-6]
Since the average physical product of labor (APL or QIL) usually rises first, reaches a maximum, and then falls (see Section 6-2), it follows that the AVC curve first falls, reaches a minimum, and then rises. Since the AVC curve is U-shaped, the ATC curve is also U'-shaped. The ATC curve continues to fall after the AVC curve begins to rise as long as the decline in AFC exceeds the rise in AVe. The V shape of the MC curve can similarly be explained as follows:
w MPL
[7-7]
Since the marginal product of labor (MPL or f..Qlfo.L)first rises, reaches a maximum, and then falls, it follows that the MC curve first falls, reaches a minimum, and then rises. Thus, the rising portion of the MC curve reflects the operation of the law of diminishing returns.
CHAPTER 7
269
7-3
+ (4K)($1O)
= $80
This is shown as point A' in the middle panel, where the vertical axis measures total costs and the horizontal axis measures output. From point C on the expansion path in the top panel, we get point C ($100) on the LTC curve in the middle panel for 2Q. Other points on the LTC curve are similarly obtained.' Note that the LTC curve starts at the origin because there are no fixed costs in the long run. From the LTC curve we can derive the firm's long-run average cost (LAC) curve. LAC is equal to LTC divided by Q. That is, LAC = LTC [7-8]
For example. the LAC to produce I Q is obtained by dividing the LTC of $80 (point A' on the LTC curve in the middle panel of Figure 7-2) by 1. This is the slope of a ray from the origin to point A' on the LTC curve and is plotted as point A" in the bottom panel of
3 Point E' on the LTC curve in the middle panel of Figure 7-2 is based on the assumption that 3Q is produced with 5.4L and 5.4K (not shown on the expansion path in the top panel in order not to clutter the figure), so that LTC = $108. The shape of the LTC curve will be explained in terms of the LAC curve that is derived from it.
270
o
LTC ($)
456
12
15
18
24 LTC
30
Labor
300
R'
180 120 80
C' E'
108
Wo
o ------.----,~--_.----_.----_r----,_----------~
o
LAC,
Output (Q)
20
o o
1 2 3 4 5 6
Output
FIGURE 7-2 Derivation of the Long-Run Total, Average, and Marginal Cost Curves From point A on the expansion path in the top panel, and W= $10 and r= $10, we get point A' on the long-run total cost (LTC) curve in the middle panel. Other points on the LTC curve are similarly obtained. The long-run average cost (LAC) curve in the bottom panel is given by the slope of a ray from the origin to the LTC curve. The LAC curve falls up to point G" (4Q) because of increasing returns to scale and rises thereafter because of decreasing returns to scale. The long-run marginal cost (LMe) curve is given by the slope of the LTC curve and intersects the LAC curve from below at the lowest point on the LAC curve.
CHAPTER 7
271
Figure 7-2. Other points on the LAC curve are similarly obtained. Note that the slope of a ray from the origin to the LTC curve declines up to point G' (in the middle panel of Figure 7-2) and then rises. Thus, the LAC curve in the bottom panel declines up to point G" (4Q) and rises thereafter. It is important to keep in mind, however, that while the U shape of the short-run average cost (SAC) curve is based on the operation of the law of diminishing returns (resulting from the existence of fixed inputs in the short run), the U shape of the LAC curve depends on increasing, constant, and decreasing returns to scale, respectively, as will be explained in Section 7-4. From the LTC curve we can also derive the long-run marginal cost (LMC) curve. This measures the change in LTC per unit change in output and is given by the slope of the LTC curve. That is, LMC = D.LTC
~Q
[7-9]
For example, increasing output from OQ to lQ increases LTC from $0 to $80. Therefore, LMC is $80 and is plotted at 0.5 (i.e., halfway between OQ and 1Q) in the bottom panel of Figure 7-2. Increasing output from IQ to 2Q leads to an increase in LTC from $80 to $100, or $20 (plotted at 1.5 in the bottom panel), etc. Note that the relationship between LMC and LAC is the same as that between the short-run MC and ATC or AVe. That is, the LMC curve reaches its lowest point at a smaller level of output than the LAC curve and intersects the LAC curve from below at the lowest point on the LAC curve.
272
100 80 60 40 20 0 0 1 2
3
G"
Output
$ 100 80
LAC
60 40 20
F" G" H"
o o
1
Output
FIGURE 7-3 Relationship Between the Long-Run and Short-Run Average Cost Curves In the top panel, the LAC curve is given by A "B*C"E*G"j*R " on the assumption that the firm can build only four scales of plant (SAC1, SAC2, SAC), and SAC4). In the bottom panel, the LAC curve is the smooth curve A"B"C"D"E"F"G"H"J"N"R" on the assumption that the firm can build a very large or infinite number of
plants in the long run.
Thus, if the firm could build only the four scales of plant shown in the top panel of Figure 7-3, the long-run average cost curve of the firm would beA"B*C'E*G"J*R". If the firm could build many more scales of plant, the kinks at points B*, E*, and J* would become less pronounced, as shown in the bottom panel of Figure 7-3. In the limit, as the number of scales of plants that the firm can build in the long run increases, the LAC curve approaches the smooth curve indicated by the LAC curves in the bottom panels of Figures 7-2 and 7-3. Thus, the LAC curve is the tangent or "envelope" to the SAC curves and shows the minimum average cost of producing various levels of output in the long run, when the firm can build any scale of plant. Note that only at point G" (the lowest point on the LAC curve) does the firm utilize the optimal scale of plant at its lowest point. To the left of point G", the firm
CHAPTER 7
273
in Electricity
Generation
power companies have been buying more and more power from independent power producers. But all of this is changing very rapidly as the indu try braces for deregulation and the end of their monopoly power (see Chapter 12). Furthermore, recent technological advances have greatly reduced the average co t of producing electricity with micro-turbine generators, and tills may soon provide even small businesses with the choice of generating their own electricity efficiently.
Figure 7-4 shows the estimated LAC curve for a sample of 114 firms generating electricity in the United States in 1970. The figure show that LAC is lowest at the output level of about 32 billion kilowatt-hours. The LAC curve, however, is nearly L-shaped (the reason for and significance of thi are explained in Section 7-4). In order to avoid the increasing costs that they would incur in producing more power themselves to satisfy increasing consumer demand, electric
LAC ($)
e
.s::
5.99 -
:= 5.83 0
5.67 5.51 5.36 5.20 5.04 4.88 4.73 LAC
i: '" ~
:><
0 0 0
... ..
0-
~ .!!!
0 c
?
0
10
20
30
40
50
60
70
Billions of kilowatt-hours
FIGURE 7-4 The Long-Run Average Cost Curve in Electricity Generation The figure shows the estimated LAC curve in the generation of electricity in the United States for a sample of 114 firms in 1970. The lowest LAC occurs at the output level of 32 billion kilowatt-hours, but the LAC curve is nearly t-shaped,
Source: L. Christensen and H. Green, "Economies of Scale in U.S. Electric Power Generation," Journal of Political Economy (August 1976), p. 674; "Electric Utilities Brace for an End to Monopolies," The New York TImes (August 18, 1994), p. I; "Energy: Power Unbound," The Wall Street Journal (September 14, 1998), pp. R4, RIO; The Royal Academy of Engineering, The Cost of Generating Electricity (London: The Royal Academy of Engineering, 2004); and OECD, Projected Costs of Generating Electricity (paris: OECD, 2005).
274
operates on the declining portion of the relevant SAC curve, while to the right of point G the firm operates on the rising portion of the appropriate SAC curve (see the top panel of Figure 7-3). The long run is often referred to as the planning horizon because the firm can build the plant that minimizes the cost of producing any anticipated level of output. Once the plant has been built, the firm operates in the short run. Thus, the firm plans for the long run and operates in the short run (see Case Study 7-1).4
If
7-4
If the firm is uncertain about the level of demand and production in the future, it may want to build a more flexible plant for the range of anticipated outputs, rather than the optimal plant for producing a particular level of output at an even lower cost (see Problem 8). 5 The technological forces for economies of scale are sometimes referred to as "plant economies" because they operate at the plant level. On the other hand, the financial reasons for economies of scale are often referred to as "firm economies" because they arise at the firm (as opposed to the plant) level.
CHAPTER 7
275
Besides the technological reasons for increasing returns to scale or decreasing costs, there are financial reasons that arise as the size of the firm increases. Because of bulk purchases, larger firms are more likely to receive quantity discounts in purchasing raw materials and other intermediate (i.e., semi processed) inputs than smaller firms. Large firms can usually sell bonds and stocks more favorably and receive bank loans at lower interest rates than smaller firms. Large firms can also achieve economies of scale or decreasing costs in advertising and other promotional efforts. For all these technological and financial reasons, the LAC curve of a firm is likely to decline as the firm expands and becomes larger. Decreasing returns to scale, on the other hand, arise primarily because as the scale of operation increases, it becomes ever more difficult to manage the firm effectively and coordinate the various operations and divisions of the firm. The number of meetings, the paperwork, and telephone bills increase more than proportionately to the increase in the scale of operation, and it becomes increasingly difficult for top management to ensure that their directives and guidelines are properly carried out by their subordinates. Thus, efficiency decreases and costs per unit tend to rise. In the real world, the forces for increasing and decreasing returns to scale often operate side by side, with the former prevailing at small levels of output (so that the LAC curve declines) and the latter tending to prevail at much larger levels of output (so that the LAC curve rises). The lowest point on the LAC curve occurs when the forces for increasing and decreasing returns to scale just balance each other. In the real world, however, the LAC curve is often found to have a nearly flat bottom and to be L-shaped rather than U-shaped. This implies that economies of scale are rather quickly exhausted and constant or nearconstant returns to scale prevail over a considerable range of outputs in many industries. In these industries, small firms coexist with much larger firms." There are some industries, however, in which the LAC curve declines continuously as the firm expands output, to the point where a single firm could satisfy the total market for the product or service more efficiently than two or more firms. These cases are usually referred to as "natural monopolies" and often arise in the provision of such utilities as electricity and public transportation. In such cases the local government often allows a single firm to supply the service to the entire market but subjects the firm to regulation (i.e., regulates the price or rate charged for the service). Three possible shapes of the LAC curve (U-shaped, L-shaped, and constantly declining) are shown in Figure 7-5 and examined in various U.S. industries in Case Study 7-2. Economies of scale have to be distinguished from economies of scope. The latter refer to the lowering of costs that a firm often experiences when it produces two or more products together rather than each alone. A smaller commuter airline, for example, can profitably extend into providing cargo services, thereby lowering the cost of each operation alone. Another example is provided by a firm that produces a second product in order to use the by-products (which otherwise the firm had to dispose of at a cost) arising from the production of the first product. Management must be alert to the possibility of profitably extending its product line to exploit such economies of scope.
The inability to observe rising LAC in the real world may be due to the fact that firms avoid expanding output when LAC begins to rise rapidly.
276
LAC of Small Firms as a Percentage of LAC of Large Firms Industry Hospitals Commercial banking Demand deposits Installment loans Electric power Airlines (local service) Railroads Trucking 116 102 112 100 100 95 Percentage 129
Source: H. E. Frech and R. L. R. Mobley, "Resolving the Impasse on Hospital Scale Economies: A New Approach," Applied Economics (March 1995); F. Bell and N. Murphy, Costs in Commercial Banking, Research Report No. 41 (Boston: Federal Reserve Bank of Boston, 1968); L. Christensen and W. Greene, "Economies of Scale in U.S. Electric Power Generation," Journal of Political Economy (August 1976); G.!Eads, M. Nerlove, and W. Raduchel, "A Long-Run Cost Function for the Local Service Airline Industry," The Review of Economics and Statistics (August 1969); Z. Griliches, "Cost Allocation in Railroad Regulation," The Bell Journal of Economics and Management Science (Spring 1972); R. Koenker, "Optimal Scale and the Size Distribution of American Trucking Firms," Journal of Transport Economics and Policy (January 1977); "Automobiles: GM Decides Smaller Is Better," The Margin (November-December 1988), p. 28; and "GM Outstrips Ford on Vehicle Productivity, Study Finds," Financial Times (June 14,2002), p. 18.
CHAPTER7
277
LAC
LAC
LAC
($)
($)
<.:
0 0 0 0 0
L
0
($)
0 0
FIGURE 7-5 Possible Shapes of the LAC Curve The left panel shows a U-shaped LAC curve, which indicates first increasing and then decreasing returns to scale. The middle panel shows a nearly L-shaped LAC curve, which shows that economies of scale quickly give way to constant returns to scale or gently rising LAC. The right panel shows an LAC curve that declines continuously, as in the case of natural monopolies.
7-5
LEARNING CURVES
As firms gain experience in the production of a commodity or service, their average cost of production usually declines. That is,for a given level of output per time period, the increasing cumulative total output over many time periods often provides the manufacturing experience that enables firms to lower their average cost of production. The learning curve shows the decline in the average input cost of production with rising cumulative total outputs over time. For example, it might take 1,000 hours to assemble the 100th aircraft, but only 700 hours to assemble the 200th aircraft because managers and workers become more efficient as they gain production experience. Contrast this to economies of scale, which refer instead to declining average cost as the firm's output per time period increases. Figure 7-6 shows a learning curve which indicates that the average cost declines from about $250 for producing the lOOth unit of the product (point F), to about $200 for producing the 200th unit (point G), and to about $165 for the 400th unit (point H). Note that the average cost declines at a decreasing rate so that the learning curve is convex to the
Average cost ($)
250
Learning curve
200
H
150
o o
100 200 300 400
Cumulative total output (0) FIGURE7-6 Learning Curve Learning curve FGH shows that the average cost is about $250 for producing the 100th unit (point F), about $200 for the 200th unit (point G), and about $165 forthe 400th unit (point H).
278
PARTTHREE
origin. This is the usual shape of learning curves; that is, firms usually achieve the largest decline in average input costs when the production process is relatively new and less decline as the firm matures. The learning curve can be expressed algebraically as follows:
C=aQb [7-10]
where C is the average input cost of the Qth unit of output, a is the average cost of the first unit of output, and b will be negative because the average input cost declines with increases in cumulative total output. The greater the absolute value of b, the faster the average input cost declines. Taking the logarithm of both sides of Equation 7-10 gives log C = log a + b log Q [7 -11]
In the above logarithmic form, b is the slope of the learning curve. The parameter of the learning curve in the double-log form of Equation 7-11 (i.e., log a and b) can be estimated by regression analysis with historical data on average cost and cumulative output. Suppose that doing this gives the following result: log C = 3 - 0.3 log Q [7-12]
In Equation 7-12, C is expressed in dollars, log a = 3 and b = -0.3. Thus, the average input cost of the 100th unit is log C = 3 - 0.3 log 100 Since the log of 100 is 2 (obtained by simply entering the number 100 in your hand calculator and pressing the "log" key), we have log C = 3 - 0.3(2)
= 3 - 0.6
=2.4 Since the antilog of 2.4 is 251.19, the average input cost (C) of the 100th unit of output is $251.19. The average input cost for the 200th unit is log C = 3 - 0.3 log 200
= 3 - 0.3 (2.30103) = 3 - 0.690309 = 2.309691
Therefore, C = $204.03. The student can determine in an analogous way that for the 400th unit, C = $165.72. These are, in fact, the values shown by the learning curve in Figure 7-6. Learning curves have been documented in many manufacturing and service sectors, ranging from the manufacturing of airplanes, appliances, shipbuilding, refIned petroleum products, to the operation of power plants. They have also been used to forecast the needs for personnel, machinery, and raw materials, and for scheduling production, determining the price at which to sell output, and even for evaluating suppliers' price quotations. For example, in its early days as a computer-chip producer, Texas Instruments adopted an aggressive price strategy based on the learning curve. Believing that the learning curve in chip production was steep, it kept unit prices low in order to increase its cumulative total
CHAPTER 7
279
output rapidly and thereby benefit from learning by doing. The strategy was successful, and Texas Instruments became one of the world's major players in this market. How rapidly the learning curve (i.e., average input costs) declines can differ widely among firms and is greater the smaller the rate of employee turnover, the fewer the production interruptions (which would lead to "forgetting"), and the greater the ability of the firm to transfer knowledge from the production of other similar products. The average cost typically declines by 20 to 30 percent for each doubling of cumulative output for many firms. Firms, however, do not rely only on their production experience to lower costs and are looking farther and farther afield from their indu try to gain insights on how to increase productivity (see Case Study 7-3).
Source: "To Compete Better, Look Far Afield," The New York Times (September IS, 1994), sec. 3, p. I I; "GM Takes Advice from Disease Sleuths to Debug Cars," The Wall Street Journal (ApnlS, 1999), p. B I; "Motorola's ew Research Efforts Look Far Afield," The Wall Street Journal (June 17, 1999), p. 86; and 'The World of Ideas," Fortune (JlIly 25. 2(05), pp. 90-96.
280
7-6
7 See "The Outing of Outsourcing," The Economist (November 25, 1995), pp. 57-58; "Has Outsourcing Gone Too Far?" Business Week (April I, 1996), pp. 26-28; "The Hidden Costs of Outsourcing," Sloan Management Review (Spring 2001), pp. 6()"{)9; and "Out of the Backdoor,' The Economist (December 1, 2(01), pp. 55-56. 8 See W. H. Davidson and J. de la Torre, Managing the Global Corporation (New York: McGraw-Hill, 1989); and "Reengineering 101," Forbes (May 13,2002), pp. 82-88.
CHAPTER 7
281
The IBM PC and the Boeing 777 and 787 Are All but American!
Table 7-3 shows that of the total manufacturing cost of $860 for the mM PC in 1985, $625 was for parts and components made abroad (of which $230 was from U.S.-owned plants). Although all the parts made overseas could be manufactured domestically, they would have cost more and would have led to higher PC prices in the United States (and reduced competitiveness of mM PCs in international markets). Today, even a larger proportion of parts and components going into the mM PC are made abroad and, in 2004, mM sold its laptop business to China's Lenovo for $l.75 billion. Only 13 of the 33 major components of the new Boeing 777-330ER jetliner are made in the United States; 7 are made in Japan and another 13 in other countries (England, Italy, South Korea, and Spain). Even less of the brand new Boeing 787 Dreamliner jet is made in the United States.
Distribution and Abroad Total manufacturing cost: Portion made abroad: In U.S.-owned plants In foreign-owned plants
of Manufacturing
Distribution of manufacturing costs: Monochrome monitor (Korea) Semiconductors (Japan) Semiconductors (U.S.) Power supply (Japan) Graphics printer (Japan) Floppy disk drives (Singapore) Assembly of disk drives (U.S.) Keyboard (Japan) Case and final assembly (U.S.)
Source: "America's High-Tech Crisis," Business Week (March 11, 1985), pp. 56-67; "Outsourcing to the U.S.," The New York Times (December 25, 2(05), p. C 1; Boeing ews Release (June 11, 2(02); and "A Plastic Dream Machine," Business Week (June 20,2(05), pp. 32-35.
282
So widespread and growing is international trade in inputs and the opening of production facilities abroad that we are rapidly moving toward truly multinational firms with roots in many nations rather than in only one country, as in the past. And this affects more than multinationals. Indeed, firms that until a few years ago operated exclusively in the domestic market are now purchasing increasing quantities of inputs and components and shifting some of their production to foreign nations. For example, Malachi Mixon, an American medical-equipment company, now buys parts and components in half a dozen countries, from China to Colombia; 10 years ago it did all of its shopping at home. The popular Mazda Miata automobile, which is manufactured in Japan, was conceived in Mazda's California design lab by an American engineer at the same time that Mazda opened production facilities for other models in the United States.
Economies of Scale
Firms must constantly explore sources of cheaper inputs and overseas production in order to remain competitive in our rapidly shrinking world. Indeed, this process can be regarded as manufacturing's new international economies of scale in today's global economy. Just as companies were forced to rationalize operations within each country in the 1980s, they now face the challenge of integrating their operations for their entire system of manufacturing around the world in order to take advantage of these new international economies of scale." What is important is for the firm to focus on those components that are indispensable to the company's competitive position over subsequent product generations and outsource other components in which outside suppliers have a distinct production advantage. 10 These new international economies of scale can be achieved in five basic areas: product development, purchasing, production, demand management, and order fulfillment. In product development, the firm can design a core product for the entire world economy, building into the product the possibility of variations and derivatives for local markets. Firms can also achieve new economies of scale by purchasing raw materials, parts, and components globally rather than locally, no matter where their operations are located. Firms can coordinate production in low-cost manufacturing centers with final assembly in high-cost locations near markets. They can also forecast the demand for their products and undertake demand management on a world rather than national basis. Firms can achieve large economies of scale by shipping products from the plants closest to customers, allowing the firms to hold less inventory around the world. These new international economies of scale are likely to become even more important as we move closer to a truly global economy.
9 See "Manufacturing's New Economies of Scale," Harvard Business Review (May-June 1992), pp. 94-102; and "The New Dynamics of Global Manufacturing Site Location," Sloan Management Review (Summer 1994), pp.69-80. 10 See "Strategic Outsourcing," Harvard Business Review (November-December 1992), pp. 98-107; "Strategic Outsourcing," Sloan Management Review (Summer 1994), pp. 43-55; and "How to Think Strategically About Outsourcing," Harvard Management Update (May 2000), pp. 4-6.
CHAPTER 7
283
II See "Wanted: 675,000 Future Scientists and Engineers," Science (June 1989), pp. 1536-1538; and "Supply and Demand for Scientists and Engineers: A National Crisis in the Making," Science (April 1990), pp. 425-432. 12 See "Software Jobs Go Begging, Threatening Technology Boom," The New York Times (January 13, 1998), p. I; "Alien Scientists Take Over USA!" The Economist (August 21, 1999), p. 24; "Congress Approves a Big Increase in Visas for Specialized Workers," The New York Times (October 4, 2000), p. I; and Richard Freeman, Emily Jin, and Chia- Yu Shen, "Changing Demographics of U.S. Science-Engineering PhOs.," NBER Working Paper 10554 (June 2(05). 13 See "Keeping Out the Wrong People," Business Week (October 4,2004), pp. 90-94; "U.S. Slips in Attracting the World's Best Students," The New York Times (December 21,2004), p. I; D. Salvatore, International Economics, 8th ed. (Hoboken, .1.: Wiley, 2004), sec. 12.6; and "Jobs and Immigrants," The Wall Street Journal (August 26, 2005), p. A 12.
284
7-7
LOGISTICS OR SUPPLY-CHAIN
MANAGEMENT
Logistics, or supply-chain management, refers to the merging at the corporate level of the purchasing, transportation, warehousing, distribution, and customer services functions, rather than dealing with each of them separately at division levels. Increasingly, logistics or supply-chain management is seen not simply as a way to reduce transportation costs, but as a source of competitive advantage. For example, one health care company was able to substantially increase its market share by establishing overnight delivery to the retailer and next-day service to the customer. Monitoring the movement of materials and finished products from a central place can reduce the shortages and surpluses that inevitably arise when these functions are managed separately. For example, it would be difficult for a firm to determine the desirability of a sales promotion campaign without considering the cost of the inventory buildup to meet the anticipated increase in demand. Logistics can also help avoid other serious (even amusing) problems. For example, to get rid of an excessive stock of green cars in the mid-1990s, Volvo's marketing department offered attractive deals on green cars. Noting the increase in sales of green cars without knowing about the promotion, the manufacturing department began to produce even more green cars! In short, logistics can increase the efficiency and profitability of the firm. There are three reasons for the emergence and rapid growth of logistics. First is the development of new and much faster algorithms and ever-faster computers that greatly facilitate the solution of complex logistic problems. The second is the growing use of justin-time inventory management, which makes the buying of inputs and the selling of the product much more tricky and more closely integrated with all other functions of the firm. The third reason (as seen in Section 7-6) is the increasing trend toward globalization of production and distribution in today's world. With production, distribution, marketing, and financing activities of the leading world corporations scattered around the world, the need for logistic management becomes even more important-and beneficial. For example, the 3M Corporation saved more than $40 million in 1988 by linking its American logistic operations with those in Europe and the rapidly growing Pacific Rim region. Similarly, Sun Microsystems, a computer maker, saved $15 million and increased revenue by $30 million in the first quarter of 2001 alone by using logistic management that practically eliminated product shortages and hence sales losses, as well as unsold products, which lose value very quickly. By centralizing several logistic functions, companies achieve greater flexibility and savings in ordering inputs and in increased revenues in selling products. Despite its obvious merits, however, only about 10 percent of small companies now have expertise and are highly sophisticated in logistics; this is certainly likely to change during this decade. Among the companies that are already making extensive use of logistic management are Philips India, ffiJL, Future Group, Reliance Fresh, Samsung India, as well as the express courier companies (Bluedart, Express, AFL, First Flight, and Gati).14 Case Study 7-5 discusses logistics at National Semiconductors, Saturn, and Compaq.
14 "Logistics: A Trendy Management Tool," The New York Times (December 24, 1989), sec. 3, p. 12; "Chain Reaction," The Economist (February 2, 2002), pp. 13-14; and "A Moving Story," The Economist (December 7,
2002), p. 65.
CHAPTER 7
285
CASE STUDY 75
Source: "Delivering the Goods," Fortune ( ovember 28, 1994), pp. 64-78; "Logistics Aspires to Worldly Wisdom," Financial Times (June 17, 1999), p. II; "Supply Chain Logistics Moving Up the Corporate Agenda," Financial Times (December 1, 1998), p. I; M. A. Cohen et al., "Saturn Supply-Chain Innovation: High Value in After-Sales Service," Sloan Management Review (Summer 2000), pp. 93-101; "Global Goods Jugglers," The Wall Street Journal (July 5, 2005), p. All; "Shippers Warn of Supply Chain Chaos," Financial Times (September 1, 2005), p. 2; and "A Distribution System Brought to Its Knees," The New York Times (September 1, 2005), p. Cl.