Top Banner
PowerPoint Slides © Michael R. Ward, UTA 2014
27

PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Jan 11, 2016

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

PowerPoint Slides © Michael R. Ward, UTA 2014

Page 2: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Cost CurvesEcon 5313

• Want to compare MR to MC.• So far, we assumed that MC was constant.• This is OK for small decisions.

• Small enough that MC is not altered much• Small enough that estimates of any change are not worth

performing.

• But may not be appropriate for larger decisions. Possibilities include:• Scale economies• Scope economies• Learning-by-doing

Page 3: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Production ExperimentEcon 5313

• Need series of volunteers (one at a time). We are going to see how many “widgets” can be made with 1 worker, 2 workers, etc. Widget production entails folding a piece of paper twice and stapling the folds together.

• Q=f(K,L) where Q is the widget, K is capital (stapler) and L is labor (student volunteers).

• Widgets are fragile and break if the fall to the ground.• Neatness does not count.• Our goal is to fill values in this table.

Page 4: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Production ExperimentEcon 5313

• Our goal is to fill values in this table

K L Q1 0 01 1  1 2  1 3  1 4  1 5  1 6  

Page 5: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Production ExperimentEcon 5313

• Our real goal is to learn about returns to scale. First we want to fill in some additional columns.

• How do we define Average Product of Labor(APL)?• APL = Q/L

• How do we define Marginal Product of Labor (MPL)?• MPL = DQ/DLK L Q AP MP

1 0 01 1 1 2 1 3 1 4 1 5 1 6

Page 6: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Production ExperimentEcon 5313

• The tabular data can be graphed and usually looks like this.

• What do we call it when MP is rising?• Increasing Returns to Scale (IRS)

• What do we call it when MP is falling?• Decreasing Returns to Scale (DRS)

1 2 3 4 5 60

1

2

3

4

5

6

7

APMP

Page 7: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Production ExperimentEcon 5313

• What caused the increasing returns to scale in our experiment?• Division of Labor (also called Specialization)

• What caused the decreasing returns to scale in our experiment?• Congestion: Only one stapler to be shared by all• Coordination: Had to agree on roles for all participants and had to

hand off work incomplete product to next team member

• Just about all production processes share these properties.• Division of Labor in Smith’s Pin Factory• Congestion from “fixity” of various inputs• Coordination from communication with ever more participants

Page 8: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

ImplicationsEcon 5313

• Often, this is where a manager add the most firm value• How do you eke out ever more gains from increased division of

labor without invoking the curse of congestion and coordination?

• This is THE feature behind the industrial revolution• Ex Ford’s assembly line but also Ford’s extensive vertical

integration• Precursor of the MBA was the “efficiency expert”

• A dominant feature of the information revolution is communication• Reducing the coordination problem implies greater specialization.• Also, participants in production need not be in one firm but can

be along the supply chain

Page 9: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

ImplicationsEcon 5313

• What does a “hump” shaped marginal product curve imply for the shape of the marginal cost curve?

• Suppose a unit of K cost $10 and a unit of L cost $10• Fill in this table from the experiment:

• AC = TC/Q• MC = DTC/DQ

K L Q AP MP TC AC MC1 1 1 2 1 3 1 4 1 5 1 6

Page 10: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

ImplicationsEcon 5313

• A “hump” shaped marginal product curve implies a “U” shaped the marginal cost curve.

1 2 3 4 5 60

1

2

3

4

5

6

ACMC

Page 11: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

ImplicationsEcon 5313

• In the real world, MC are from increments from more than one input• Measuring marginal productivity a little trickier• But general principle still usually generates “U” shaped MC curve• Costs minimization is at bottom of the “U”

Page 12: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Sony Transistor RadioEcon 5313

• Akio Morita and the Sony transistor radio way back at the founding of the company

• He was offered the opportunity to sell 100,000 units but turned it down

• Because of capacity issues, he estimated his unit costs at, $20 for 5K, $15 for 10K, and $40 for 100K

• Expansion is not always desirable

Page 13: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Scale EconomiesEcon 5313

• For some production processes, MC always below AC• No “U” shape, increasing returns to scale everywhere

• This is particularly true for “information goods”• Ex Computer software, music, movies, video games, smart phone

apps, online auctions, social networks• Now, costs minimization can lead to ever bigger firms• Role in dotcom bubble of 1999-2000?

• Example: Webvan going to offer online grocery shopping• Use web interface for ordering like Amazon• Purchased small pickup trucks specially designed to be

refrigerated for deliveries• Only makes sense for huge volume• Company failure meant $1 billion loss on trucks alone

Page 14: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Scope EconomiesEcon 5313

• Suppose we redid our production experiment with two different widget designs. 1. You make equal numbers of two designs in one process2. You make one in one process and the other in a different

process

• Which organizational form leads to the least costly production?• If 1, then economies of “scope”• If 2, then diseconomies of “scope”• C(Q1, Q2) versus C(Q1) + C(Q2)

Page 15: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Scope EconomiesEcon 5313

• Where do economies of scope come from?• Usually, economies of scale in some component

• Ex: Auto assembly – Why do they make trucks, sedans, minivans, sports cars etc.?• Typically buy components but make engines (why?)• Hard to contract over due to holdup problems• But design and tooling represents a huge FC• Make just one version of a 4 cylinder (or V6, or V8) and put it

into as many models as possible

Page 16: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Breakfast Sausage ProducerEcon 5313

• Regional - Used 18 trucks in distribution to 21 southern and Midwestern states

• But demand is seasonal – peaks in winter• Trucks were idle in summer – not exploiting scale

economies• Possible Solutions:

• Outsource distribution – supplier can “smooth” capacity utilization over other products from other customers, but vulnerable to holdup

• Vertically integrate – parent uses trucks for other products during summer, but blunts incentives

Page 17: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Pet Food ProducerEcon 5313

• A pet food company has 2,500 products (SKU’s) with 200 different formulas

• They receive a lot of pressure from large customers like Wal-Mart to reduce prices

• To respond to Wal-Mart, the company shrinks it product offerings• 70 SKUs w/13 formulas • This led to a 25% savings for the company because of

reduced production costs• How?

Page 18: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Pet Food ProducerEcon 5313

• Production with each new formula requires that the production line shut down for setup

• Line workers idle during this time• The production line exhibits scale economies

Page 19: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Learning-By-DoingEcon 5313

• Learning curves or Learning-By-Doing (LBD) slightly different from scale economies

• Scale economies from producing 20 units versus 10 units in a period of time

• Learning-By-Doing from producing 10 units in a period of time after producing 10 units in the previous period• Not necessarily due to division of labor• Due to ‘tweaking’ the process based on what you

discovered the first time

Page 20: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Microchip MakersEcon 5313

• Microchip makers stamp out increasingly complicated chips onto silicon wafers

• Test each chip to check for defects• Also test to correct design or production defects• First batch has whopping 90% defect rate

• Fix most glaring mistakes

• Next batch has 80% defect rate• Just doubled the yield per unit of effort!• Fix a few more mistakes

• Continued tweaking typically ultimately yields 15% defect rate at end of product lifecycle• One reason why prices of consumer electronics falls

Page 21: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Airplane ManufacturerEcon 5313

• Every time an airplane manufacturer doubles production, marginal costs decrease by 20%• Number produced in time period may not change

Q MC TC AC1 $100.00 $100.00 $100.00 2 $80.00 $180.00 $90.00 3 $70.21 $250.21 $83.40 4 $64.00 $314.21 $78.55 5 $59.56 $373.77 $74.75 6 $56.17 $429.94 $71.66 7 $53.45 $483.39 $69.06 8 $51.20 $534.59 $66.82 9 $49.29 $583.89 $64.88

10 $47.65 $631.54 $63.15

Page 22: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Airplane ManufacturerEcon 5313

• Same information graphically

0 2 4 6 8 10$0

$20

$40

$60

$80

$100

$120

MCAC

Page 23: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Airplane ManufacturerEcon 5313

• Am Air is considering a large airframe purchase from a Boeing. Because of this order, Boeing’s unit costs will be lower in the future. Boeing will be able to charge lower prices to future customers, airlines in competition with Am Air. Am Air could be providing a competitive advantage to its competitors!

• What are possible contracting solutions? • Am Air gets kickbacks on future orders (probably violates

antitrust laws)• Am Air gets a percent of increase in Boeing’s stock market

value when the deal is announced or buy calls (probably violates security laws)

• Exclusive contract for airframe and favorable prices

Page 24: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Anticipating Learning-by-DoingEcon 5313

• Every time you double production, your costs decrease by 50%. The first unit costs you $64 to produce

• On a project for 4 units, what is your break-even price?

Page 25: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Anticipating Learning-by-DoingEcon 5313

• On a project for 4 units, what is your break-even price?

• You can win another project for 2 more units. What is your break-even price for those units?

Q MC TC AC1 $64 $64 $64 2 $32 $96 $48 3 $21 $117 $39 4 $16 $133 $33 5 $13 $146 $29 6 $11 $157 $26

Page 26: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

From the Blog• Chapter 7• Nissan-Renault Alliance• Thoreau was a polluter• Joe Fresh at JCP• Airline Operations• Auto Alliance• Rental Car Formats• Learning-by-doing is Embodied in the Manager

Econ 5313

Page 27: PowerPoint Slides © Michael R. Ward, UTA 2014. Cost Curves Econ 5313 Want to compare MR to MC. So far, we assumed that MC was constant. This is OK for.

Main Points• Increasing Returns – Usually from division of labor• Diminishing Returns – Usually from congestion and

coordination problems• Efficient firms balance the two

• Solving some congestion issues allows for more specialization

• Scale economies means that costs fall as you produce more

• Scope economies means that the costs of producing one good are lower because you produce another good• Often from scale economies in a common component

• Learning-by-Doing means that current production will lower future costs• Anticipate this when planning over the product lifecycle

Econ 5313