Table of Contents
Teaching Introductory Microeconomics
Using System Dynamics:
Reflections on an Experiment at WPI
James M. Lyneis
Professor of Practice
Department of Social Science and Policy Studies
Worcester Polytechnic Institute
100 Institute Road
Worcester, MA 01609-2280
jmlyneis@ wpi.edu
Abstract
Traditional microeconomics makes extensive use of static equilibrium tools to understand the
behavior of consumers and producers in markets. Yet we all know that product, labor, and
capital markets are all highly dynamic and rarely, if ever, in equilibrium. Would a system
dynamics approach offer greater insight?
Last fall this author had the opportunity to teach an introductory microeconomics class to
undergraduates at WPI. Because this course was microeconomics, and not system dynamics,
the topics covered were those that one would expect to see in a typical microeconomics course,
for example: (1) comparative advantage, specialization, and trade; (2) markets as a means of
allocating scarce resources; and, (3) the failures of markets and how governments deal with
these failures. The author used a mix of traditional microeconomics tools and system dynamics
to present the topics.
This paper describes the topics covered in this course, the use of traditional economic methods,
and how (and what) system dynamics was introduced. It concludes with some reflections on
how the course went, and where | suggest we might go from here (including the sharing of
experiences, ideas, and materials with others teaching economics). The primary audience for
this paper is system dynamicists looking to teach the concepts of microeconomics. | therefore
spend some time discussing the basics of microeconomics as traditionally taught, but do not
discuss the basics of system dynamics. For those readers whose training is in economics, |
apologize in advance and hope that some of the ideas here might prompt you to seek additional
information on system dynamics elsewhere (including the course materials and references in
this paper).
Introduction
Economics is the study of how a society deals with the problem of scarcity. Broadly speaking,
our desires almost always exceed our resources - we need and/or want more than we can
produce. Economics addresses how we deal with this problem - i.e., how a society grows,
allocates and manages its scare resources.
Economics at the college level is usually introduced via two courses: macroeconomics and
microeconomics. In fact, this dichotomy continues in advanced courses and to some extent in
practice as well. According to Mankiw [2001], macroeconomics is “the study of economy-wide
phenomena, including inflation, unemployment, and economic growth;” microeconomics is “the
study of how households and firms make decision and how they interact in markets.” While this
separation seems artificial, and even incorrect, to a system dynamicist, it is the traditional way in
which economics is studied.
While the exact content will obviously depend on the instructor and particulars of the course (its
duration, whether taught before or after macro), introductory micro usually covers the following
five topics:
1. Comparative advantage, specialization and trade;
2. Markets as a means of executing trades and allocating resources, i.e., of balancing
supply and demand;
3. “Failures” of markets (e.g., monopolies and oligopolies, leaving some behind, and
dealing with common goods);
4. Product, labor, and capital markets; and
5. Externalities, common goods, and environmental economics
Underlying the teaching of economics are several important assumptions regarding behavior.
First, that people act rationally (for example, weighing the costs and benefits of each possibility
and acting to maximize the net benefit, usually with the further assumption of perfect
information). Second, that people consider opportunity costs rather than just direct or out-of-
pocket costs. And third, people act at the margin, i.e., evaluating decisions on the marginal cost
or benefit of the action. As will be discussed further below, these concepts are taught in
traditional micro economics using the ideas and tools of marginal utility, average and marginal
cost, production functions, elasticity, and comparative statics.
Such is the teaching of microeconomics at WPI, where | had the opportunity to teach
introductory microeconomics in the Fall 2002. In that course, | introduced the use of system
dynamics ideas and models to address micro-economic problems. When one thinks of system
dynamics applied to economics, one usually thinks of its application to macroeconomics,
although with an emphasis on macro behavior from micro structure (Forrester [1979], Mass
[1975], Forrester [1982], Radzicki [2003]). However, system dynamics has also been applied to
what economists would traditionally consider microeconomic problems - understanding the
behavior of firms and industries, and regulatory policy (Meadows [1970], Ford [1997]; Paich and
Sterman [1993]; Lyneis [2000]).
This paper describes the topics covered in this course, the use of traditional economic methods,
and how system dynamics was introduced. It concludes with some reflections on how the
course went, and where | suggest we might go from here.
Teaching Approach and Materials Used
This was a course in microeconomics, not in system dynamics. Therefore, it was felt that the
five topics and ideas/analysis tools noted above must be covered. As a result, the approach
taken was to cover the topics using both traditional economics methods and system dynamics.
There were several reasons for this. First, some of the topics are best taught using the
traditional approaches. Second, certain of the traditional approaches provide structure and
parameters for the system dynamics models. And third, because some of the students in the
course may go on to study economics elsewhere, it was felt that some background in the
traditional methods should be given. For each of the topics, the merged teaching approach is
discussed and illustrated below. As a traditional textbook for this course, | used Mankiw [2001],
although Baumol and Blinder [2003], Parkin [2001], and Stiglitz and Walsh [2002] are also
common texts. For the system dynamics, Sterman [2000] was put on reserve but the primary
source of information was lectures and lecture notes posted to the course website.
1. Comparative advantage, specialization and trade
Almost all micro textbooks devote a chapter to “trade.” This is usually done first thing as it sets
the stage for the need for markets to facilitate the trade and allocation of resources. Trade here
includes both trade between individuals as well as countries. These chapters discuss the
concepts of absolute advantage, comparative advantage, and the resultant gains to be made
from specialization and trade. These ideas were introduced using traditional economics
concepts, including production possibilities frontiers and opportunity cost. As these have no
direct connection to the system dynamics topics, | will not discuss these ideas further in this
paper.
2. Markets as a means of allocating resources, i.e., of balancing supply and demand
A significant portion of microeconomics textbooks and courses covers the topic of allocating
resources via markets. By allocating resources, economists mean balancing supply and
demand. Traditional economics teaches this topic via supply and demand curves and the
process of comparative statics. | used this same approach to introduce the concepts of
markets, supply, and demand. A brief summary follows as it is used as the starting point for the
system dynamics model structure.
Figure 1 shows a typical demand curve for the product ice cream cones. As price increases,
quantity demanded decreases either because users cannot afford to buy more, or because of
“diminishing marginal utility” (in some courses and textbooks, the theory of consumer choice is
discussed in some detail before introducing the demand curve, or after introducing it to justify
the shape of the demand curve; | skipped the details of this theory and relied on common sense
summary arguments to justify the shape of the demand curve). Economists distinguish between
changes in “quantity demanded” and changes in “demand.” Changes in quantity demanded
reflect changes as a direct result of changes in price. A system dynamicist might view this as
the short-term feedback response. Changes in demand reflect changes in the entire demand
curve as a result of non-price factors, such as the weather, price of substitutes, etc. Toa
system dynamicist, some changes in demand are exogenous and others reflect other or longer-
term feedback responses. These differences are illustrated in Figure 2.
Figure 1. Demand Curve
Price of
Ice-Cream
Cone
$3.00
2.50 |—--—» Law of Demand: Demand
decreases with increasing
2.00 price because: (a) unable to
. afford more; (b) “diminishing
1.50 marginal utility” (having more
‘ is not worth extra cost)
1.00
0.50
Quantity of
01234567 8 91011 12 Ice-Cream
Cones
Figure 2. Quantity demanded versus shifts in the demand curve.
a. Change in quantity demanded as a function of changes in price
Price of
Cigarettes
per Pack
A tax that raises the
C price of cigarettes
$ALOO forsesssicsnsnsisessnnin’ results in a movement
f H along the demand
curve.
2.00
D,
® 12<————-20
Number of Cigarettes
Smoked per Day
b. Shift in demand curve of...
Price of
Ice-Cream
Cone Caused by changes in
anything other than
price, e.g. income,
price of substitutes,
weather, ...
D 2
D Dy
3 Quantity of
(?)
Figure 3 shows a typical supply curve. As price increases, quantity supplied increases. There
are two reasons why supply might increase with higher prices: first, because additional, higher-
cost producers become willing to supply, and second, because of additional production from
current producers, given the assumption that marginal costs increase as production increases
(again, most texts and courses spend some time on the cost structure of the firm, and how total,
average, and marginal costs change as production volume increases; | covered this later in the
term before introducing the difference between a perfectly competitive market and a
monopolistic market). Again, economists distinguish between “quantity supplied” and “supply.”
Changes in quantity supplied result from the direct feedback effect of changes in price; changes
in “supply” reflect shifts in the supply curve right or left caused by such factors as changes in
technology (embodied in capital stock), shifts in input prices, etc. Again, some of these factors
may be exogenous, and other the result of longer-term feedback effects.
Figure 3. Supply Curve
Price of
Ice-Cream
Cone
$3.00
2.50 Law of Supply: Supply
increases with
2.00 increasing price (based
on increasing marginal
1.50 | costs as production
increases)
1.00
0.50
iii Quantity of
0123456 7 8 91011 12 Ice-Cream
Cones
Markets are in balance, or cleared, when supply equals demand. This occurs at the intersection
of the supply and demand curves as illustrated in Figure 4. If there is an initial imbalance,
suppliers either increase or lower their prices until the market is cleared.
Figure 4. Equilibrium of Supply and Demand
Price of
Ice-Cream
Cone
$3.00
2.50
1.50
1.00
0.50 i Demand
Ax Quantity of
0 123 4 56(7)8 9 1011 12 Ice-Cream
Cones
Economists use these supply and demand curves in the process of “comparative statics” to
determine how an event or policy change will affect market equilibrium. This process consists
of three steps:
1. Decide whether the event shifts the supply or demand curve (or both);
2. Decide whether the curve(s) shift(s) to the left or to the right; and
3. Examine how the shift affects equilibrium price and quantity.
These steps are illustrated in Figure 5.
Figure 5. Comparative Statics
a. How an increase in demand affects the equilibrium
Price of 1. Hot weather increases
toe:cipam the demand for ice cream...
$259
2.00
2. «resulting
in a higher
price...
H D,
io Quantity of
. juan
3. ..and a hi her —— Ice-Cream Cones
quantity sold.
b. How a decrease in supply affects the equilibrium
1. An strike at one supplier
Price of f
Ice-Cream reduces the supply of ice cream...
Cone Ss
2
Ss
oe peyrescce .
2.00;
2. ...resulting
in a higher
Price...
< Initial equilibrium
Demand
1 | 1 | 4 I
0123 47 8 9 10111213 Quantity of
~, 3. ..and a lower Ice-Cream Cones
Missing from the discussion thus far is any concept of time. This is partially introduced in
traditional economics during the discussion of elasticity. Elasticity is the change in quantity
(supplied or demanded) in response to a change in some input factor (usually price, but also
income, price of complements, price of substitutes, etc.). In order to determine elasticity, one
needs to consider time. For example, supply may be inelastic (i.e., not change much with
changes in price) in the short-term because of delays and constraints in increasing or
decreasing production, but elastic in the long-term. Similarly for demand. Therefore, in
performing a comparative statics analysis to estimate the impact of event “X”, | require my
students to conduct four rather than three steps:
1. Determine the time frame of the analysis; estimate shapes of supply and demand curves
based on the expected elasticities in that time frame (and they might be asked to
perform this analysis for two time frames, e.g., short-term and long-term);
Decide whether the event shifts the supply or demand curve (or both);
Decide whether the curve(s) shift(s) to the left or to the right; and
4. Examine how the shift affects equilibrium price and quantity.
Wn
Comparative statics provides some useful insights to students, specifically that:
* People acting in their self-interest tend to drive the market toward equilibrium.
+ Market systems are efficient in allocating resources (further analysis of the supply and
demand curves, done in class and texts, shows that the equilibrium produced by the
market maximizes what economists call consumer and producer “surplus”)
* Market systems enhance productivity growth by encouraging self-interested behavior.
+ Attempts to interfere with the market reduce the benefits (analysis of price caps, floors,
and taxes are shown to reduce the total surplus to consumers and producers).
However, comparative statics has several serious shortcomings. First, the path to the new
equilibrium is generally undefined. Economists are hazy about how the balancing occurs, often
stating that it occurs either as a result of “negotiation and bargaining between buyer and seller,”
or of “competition.” Moreover, how quickly and/or smoothly this process occurs is not
discussed. Some economists gloss over the problem, or assume that the adjustment process is
not important as long as a new equilibrium is reached:
“How quickly equilibrium is reached varies from market to market, depending on how
quickly prices adjust. In most free markets, however, surpluses and shortages are only
temporary because prices eventually move toward their equilibrium levels. Indeed, this
phenomenon is so pervasive it is sometimes called the law of supply and demand: the
price of any good adjusts to bring the supply and demand for that good into balance.”
[Mankiw 2001, p. 81]
If you read the news, you would get the impression that the market never does a good
job. It makes credit card interest rates are too high. It makes the wages of fast-food
workers are too low. It causes the price of coffee to go through the ceiling every time
Brazil has a serious frost. It increases the world price of oil whenever political instability
threatens the Middle East. These examples are not cases of market failure. They are
examples of the market doing its job of helping us to allocate our scarce resources and
ensure that they are used in the activities in which they are most highly valued. [Parkin
2000, p.9]
However, not everyone is happy about economists’ lack of concern about disequilibria. In the
article “Heretics of the Market” by Ted C. Fishman, Worth, April 1997, the author makes several
points:
In the mainstream [economic] view, financial markets are populated by rational, perfectly
informed traders, who buy and sell in an orderly fashion in reaction to news. Yet, in fact,
markets move unpredictably, often seeping to extremes of euphoria or despair that are
unjustified by the news.... And it has even less to offer in explaining the causes of
economic growth at all. According to mainstream economics, markets tend toward
equilibrium, a relatively static state in which prices and goods match up. Butin real life,
the economy rarely rests, and when it grows it does so in fits and starts. Mainstream
economics has virtually nothing convincing to say about what causes growth to speed up
or slow down....And finally, critics of mainstream economics, inside and outside the
profession, ask, How can economics make any claim to being a science when its
predictive powers are so limited? Physicists can time the arrival of a comet to the
minute, but economists can’t predict the next day’s interest rates.
{J ohn] Reed [Chairman of Citicorp] felt that as academic economics grew rich in theory
and method it increasingly disengaged from real-world problems. He found particularly
rankling that his own bank’s economists hadn't offered any warning of the Latin-America
debt crisis in the early 1980s. Blindsided, Citibank dropped billions. Reed complained
that the bank’s models took as givens the very things - interest rates, stock volatility,
and currency-exchange rates - that bankers need to predict. More fundamentally, the
models assumed a world in which prices and goods were largely in balance, leading to a
picture of a global economy that was, for the most part, static. Even the most casual
observer knew that just the opposite was true: The world economy is in constant flux
and frequently in upheaval. [Fishman, p. 102-103]
As system dynamicists, we would argue that the behavior of these markets in total affects, and
is affected by, the behavior of the economy as a whole. The path toward equilibrium matters.
A second problem with comparative statics is that it makes too many assumptions about “other
things remaining equal,” when in fact other things rarely remain equal. The analysis is therefore
incomplete and often ignored in policy debates. Economists know, for example, that price caps
(e.g. rent control) are inefficient in that they reduce total consumer and producer surplus. But
how then do we deal with people who cannot afford housing? Taxes and direct payments to
tenants? This option also has its shortcomings. Which is better, rent control or subsidies?
Similarly, how should we deal with the financial problems of farmers? In the comparative statics
analysis, side effects not considered explicitly and the long-term consequences often ignored.
As a result, comparative statics does not help us understand why markets are unstable, or when
and how to interfere with supply-demand laws to improve overall performance in the long-run
(but, as a famous economist once pointed out, “In the long-run we're all dead”). As a policy tool,
comparative statics is often ignored.
Students readily recognize these problems. Questions often arise in class that have to be
answered with “We are assuming other things remain equal.” Or, ‘We are considering only the
short-term in this analysis.” But! also used system dynamics to bring home these short-
comings. For example, | gave the students several articles to read about rent control, and ask
them what they would do and why. We then have a discussion using comparative statics,
followed by a discussion using system dynamics (reference modes and causal-loop diagram).
In summary, supply-demand curves and comparative statics provide a useful initial framing of
the problem of balancing supply and demand. They are relatively easy to explain and provide
some useful insights. However, their shortcomings are also apparent and lead naturally into the
system dynamics methodology.
3. “Failures” of markets
As hinted at in the quote by Parkin above, economists recognize that markets are not always
perfect in dealing with problems of scarcity and resource allocation. In particular, economic
texts explicitly discuss three “failures” of markets that must be addressed by other means.
1. People try to push the limits to their advantage (e.g., by reducing the number of
competitors; stretching and/or breaking the laws);
2. Some are left behind and unable to buy the “necessities of life”,
3. Forsome goods - common goods - the market mechanism does not automatically
work efficiently on its own;
To this list above, | add a fourth:
4. Markets can be unstable to the detriment of players and the economy as a whole
Pushing the limits. Some people will always try to cheat the system and make money in ways
that do not create value to society as a whole. We have therefore developed laws and
regulations to guide acceptable behavior. These laws and regulations evolve in response to the
inventiveness of businessmen and capitalists. For example, in response to the excesses of the
1920's (and before), various antitrust laws were passed (and also regulations on how much
stock can be bought on margin, etc.).
But even behaving legally, some industries move from being “perfectly competitive” to becoming
oligopolies and even monopolies. Traditional microeconomics discusses the differences
between monopoly, oligopoly, monopolistic competition, and perfect competition. The
discussion of monopolies centers on how they might arise, their impact on the efficiency of
markets (consumer and producer surplus), natural monopolies and their regulation, and antitrust
laws. Oligopoly discussion involves their impact on efficiency of markets and discussions of
strategy using game theory. Finally, monopolistic competition introduces the concepts of non-
price competition. In teaching these topics, | summarize the main ideas presented by traditional
economics, and then supplement them with a discussion about how a company builds
dominance via exploiting positive feedback loops using some of the materials from
Sterman[2000, Chapter 10]). The growth of Microsoft and the recent antitrust case against it
provide a good discussion of the issues, especially in an engineering school. | then discuss
how different market structures affect stability using system dynamics (see section below).
Some people are left behind. Other than pointing out this problem and showing how market
interventions (price caps and/or floors) are ineffective ways of dealing with the problem,
traditional economics does not deal with this in introductory microeconomics courses (and
neither did I).
Public goods. Some products are what might be called “public goods” - for example, national
defense, basic research, education, fighting poverty. Investment in these goods benefits society
as a whole, often more than the investment would benefit an individual provider. Hence these
goods are properly seen as being provided by governments outside of the market system.
Other than a short discussion of this, traditional economics does not deal with this in
introductory courses (and neither did |).
Market Instability. As noted above, markets are rarely in equilibrium. Why should we care?
There are several reasons. First, understanding dynamics should allow us to forecast better
(avoid investing at the peak!) and manage our companies better (e.g., counter-cyclical
investments, increase price or production? )(see Lyneis [2000]). Second, understanding
dynamics should allow us to manage our economies better. At an industry level, how might we
improve stability - buffer stocks? price supports? minimum wage laws? Atan economy level,
how does the behavior of the parts affect the behavior of the whole? And finally, understanding
dynamics will better allow us to avoid the “law of unintended consequences.” We must consider
how what we do today will affect the opportunities and problems of the future.
Focusing on micro-economic dynamics - the behavior of industries - we observe four basic
dynamic patterns:
1. Supply-chain (as illustrated by the “beer game”)
2. Production and Investment Cycles in Mature Industries (from commodities to
monopolies)
3. Boom and Bust in New Industries and Products
4. Bubbles, panics, and manias
These behaviors are illustrated in Figure 6. | have borrowed these from Sterman [2000] as
noted. In class, | supplement these with numerous transparencies of behaviors taken from
newspaper and magazine articles. For bubbles, | show the behavior of stock prices.
Figure 6. Examples of market dynamics
a. Supply-chain dynamics in the oil and gas industry
Oil ‘and Gas
Well Drilling E
ow
20
Fractional Growth Rate (%/yeai
°
d ? vee E
i
K Z if uw" ¥ | 1 Oiland Gas
-20 + Petroletim i i 4 ! \ if Production F
Consumption it i 1 i
‘ ‘ y 7
& HO 4
1970 1975 1980 1985 1990 1995 2000
Sterman [2000] Figure 17-3a Amplification in supply chains
Oil and gas drilling fluctuates far more than production or consumption.
The graph shows 12-month centered moving averages of the annualized fractional
Growth rate calculated from the monthly data.
b. Supply-chain dynamics in the commercial jet aircraft industry
80 7 7 i i i
Aircraft Orders
ao | ie
i |
0 a L
sag! il
1970 1975 1980 1985 1990 1995 2000
Sterman [2000] Figure 20-4 Worldwide aircraft orders. Annual growth rate in orders for
commercial aircraft (jets with > 50 seats), commercial air travel (revenue passenger km/year),
and GDP weighted by each region’s share of world air travel demand. Demand for aircraft
exhibits a large amplitude cycle of roughly 10 years.
c. Cycles in a mature commodity industry
140 L L L L L L L L L L L L L
fe Stocks
120 | I
Trend
Cattle Stocks
(million head)
e
So
6
2
36
60 roe
1930 1940
1950 1960. 1970 1980 | 1990 2000
d. Cycles in an oligopolistic industry
1.40 L L L L L L L L L L
1.20 5 f=
3
Inventory Coverage (months)
US Pulp and Paperboard Industry
fl
1
1975
Sterman [2000] Figure 20-18b Pulp and paper, inventory coverage
e. Cycles in a monopolistic industry
30 4 fe
25 4 b
‘lity
(%)
20 4 E
1950 1960 1980 1990 2000
Sterman [2000] Figure 20-5 US electric utility capability margin
Capability margin is the margin by which generation capacity exceeds peak
summer load.
f. Boom and bust in a new industry
3000: 1 1 n n 1 1 n
Sales Rate
2000,
Units/Year
3
1981 | 1983 | 1984 1986 | 1988
Sterman [2000] Figure 9-14 Sales of the Digital Equipment Corporation VAX 11/750 in Europe
After this introduction, | discuss system dynamics basics (feedback loops, stocks and flows),
and then move into a discussion of how the basic supply-demand curves can be viewed from a
system dynamics perspective. As shown in Figure 7, we first discuss how the supply-demand
curves in reality reflect two negative feedback loops - an increase in price causes an increase
in supply (and/or decrease in demand), which in tum improve the supply-demand balance and
counteract the increase in price.
Figure 7. Relating Comparative Statics to Dynamics
oO Q Quantity
e
be
supply-Demand
Balance )
Demand
This diagram initiates a discussion of how equilibrium is reached. Where are the delays - do
supply and demand change instantaneously? Further, given that buyers and sellers do not
know the supply and demand curves, how are price changes determined? This leads to the
stock-flow diagram in Figure 8a, in which response delays are noted, and in which price is
represented as a stock. Change in price responds to the supply-demand balance. When
supply exceeds demand, prices decrease and keep decreasing until supply and demand are
brought back into balance. And conversely when demand exceeds supply, prices decrease. As
shown in Figure 8b, this structure produces damped oscillations to the new equilibrium
depending on the length and order of delays.
igure 8b.
Table for
Change in Price
Supply-Demand
Balance
Gain on Price
Change
Change in Price
produces oscillations movement to equi
length and order of delays
igure 8a. Supply-demand model with delays ...
Table for Supply
Schedule
Supply
Time to React to
Price Changes
Delayed Price
for Supply
Initial Price
Time for Demand to
Delayed Price for React to Price
Demand —_— Changes
Demand Step Shift
Shift in Demand
Table for Demand
Schedule
Random
Variation
Price
325
‘Time (Month)
SD? Longer Delays & Higher Gi
D2 Longer Delay
$2 Step Gai
SD2 Step
rium price depending on
While the structure in Figure 8 captures some of the causes of instability, it also raises
additional questions. Specifically, what is “supply?” Answering this question leads to the
concepts of inventory and production delays, and the model of one-stage of a supply chain
shown in Figure 9. We build the model in class, show its behavior, and discuss amplification
down the supply chain. We conclude with a discussion of how to increase stability. (While
logistics precluded it in this course, the Beer Game would make an excellent lead-in to the
supply chain discussion.)
Figure 9. Structure of one component of a supply chain
Production Time
Ice . pe Ice Cream
| in
Orders Production Production Inventory Shipments
Completions
Inherent
Demand.
+— Demand
Inventory
Correctiong__
Shift in Demand
Time to Adjust
Inventory Desired
Inventory
Random
Variation
Desired Days
Supply
Next, | introduce a more complex model that captures one supply chain with price feedback,
and with capacity as a resource as shown in Figure 10 (based on Sterman’s [2000] commodity
model, Chapter 20). | use this model to show how perfect competition produces cycles of two
periodicities. Then, interweaved with the traditional microeconomics discussion of market
structures noted above, | show how monopolies and oligopolies change the structure of the
basic commodity model, and thus its behavior (Figures 11, 12, and 13). For example,
monopolists substitute a more explicit desired production decision rule for the capacity utilization
and expansion loops, and prices are based on markup on unit costs. Monopolies thereby tend
to stabilize behavior somewhat. Oligopolists, however, reintroduce instability via the process of
forecasting and market share competition.
Figure 10. Commodity/perfect competition model
Production
Delay
2
5 Inventory}
Producton Rate L,
a
Consumption
Rate
Capacity Life - oe
& /
{ Social and
! Technical F actors
{- \ Inventory
Capacity
Capacity Utilization Coverage
|
Delay | | Purchasing
Capacity XN Power
Capacity Incquistion oy oie
‘Acquisition Delay Expected Profitability
} of Current Operations Relative Value of
Supply Un 4 Product
lof Capacity}
on Order
J |
fapacioy Variable Costs pict
\ Substitutes:
Expected
Profitability of Nevuga—~
oe Capacity
4
¥
Fixed or
Capacity Costs
Figure 11. Monopolistic actions
Production
Delay
Prodiicton Production Rate Constimption
ado Rate
Capacity Lite
Capacity Loss Social and
‘aed Technical Factors
Inventory
Capac i
Capacity Uieatong Desired Coverage Demand
T Production
D Purchasing
Capacity
Capacity Acquistion Weaker ie Power
Acquisiion Belay Expected Proftabilty
‘of Current Operations Relative Value of
Product
‘upply Line] Delay
ot Capac
on Side Fi 7
avonae Samer ose: Price of
Substiutes
Expected
bal Profitability of Ne
Capacity
Fixed or
Capacity Cos
Figure 12. Oligopolist’s actions
Production
Delay
Production
Starts
apo (
Production Rate Consumption
Rate
ome
Inventory
Capacit
Capacity Ualation Desired Coverage Demand
Production
Di
Capacity
Capacity lacquistion Weaker ity
Acquisition Delay Expected Profitability
of Current Operations Relative aren
apply Ug] Product
fof Capacity
on Order’
Capacity Variable Costs
Initiation Price of
Substitutes
Expected
Del Profitability of New.
Market Share Capacity
Competition
Multiplier Fixed ‘or
Capacity Costs
Social and
Technical Factors,
Purchasing
Power
Figure 13. Different behavior produced by different market structures
Production Capacity
200
170
140
Commodity Oligopoly
Monopolist
80
0 10 20 30 40 50 60 70 80
Time (Year)
Production Capacity : Noise
90 100
Units/Y ear
Production Capacity : Noise Monopoly6
Units/Y ear
Production Capacity : Noise Oligopoly Forecast +MarketPtsh—— Units/Year
We conclude with a discussion of boom and bust using a diffusion model illustrated in Figure 14
(similar to Sterman [2000], Chapter 9), and extending that to consider capacity, stock prices,
and bubbles. All along, | give examples, quotes from newspaper articles, and data to make the
discussion more real. At the end, the students have seen a wide range of industry behavior and
the structure that causes it. They have related the basic ideas of system dynamics to industry
structure and behavior.
Figure 14. Boom and bust structure
Average Product Life
yo!
&
Discard B
Rate er
Replacement
ae
+
Potential
Adopters == pe] aaeetes
P —_s
i se f Ne )
Sauration + + Mouth
Total
Adoption seen Sei + Population
Advatising ofMouth «-——— N
ay Ss, A an
Advatising ction
Effectiveness f
a Contact
Rate
Product, labor, and capital markets. In most economics texts, the discussion of supply and
demand is illustrated using a product market (e.g., ice cream cones), with the occasional labor
market introduced to discuss the impact of minimum wages. With this background, later in the
term the texts quickly show how the basic supply-demand curves apply to other markets such
as labor, financial capital, and land. With regard to labor markets, the texts then discuss
earnings differentials and public policy related to wages (minimum wages, discrimination, equal
pay for equal work). | summarize this discussion and augment it with newspaper articles on the
supply and demand for engineers. We then use a mixed stock-flow-causal diagram shown in
Figure 15 to understand the observed behavior, and introduce the idea of wage inflation.
Figure 15. Labor market structure
Time to Adjust
Labor
a ae lye |
Other Workers in €. p>! Employed
Areas Net Migration Market Hires/Fires Workers
EffectofWage ig. Fraction /
Rate on ae Migrating
Supply of
Workers
Sensitivity of
Fraction to a
Reference ———_——»_
Wage Rate oe oe
Workers Needed
Table for ce of (Demand) = —*—__shiftin Demand
\
Supply-Demand Balance Curve
on Wage Rate ime to Adjust
Wages
Effect of Wage
Rate on Demand
Sensitivity of
Lagged Relative Demand to Wages
Wage Rate
Time for Demand to
Adjust to Wages
Externalities, common goods, and environmental economics. Another failure of markets is in
dealing with externalities and common resources. We follow the traditional economics
discussion of externalities, but then use system dynamics to discuss the tragedy of the
commons. The “Fishbanks” game and model are excellent for this. Finally, | devote one class
to the structure and issue raised by World Dynamics [Forrester, 1971; Meadows, et. al. 1974;
Meadows, et. al., 1992].
Reflections and Next Steps
All students at WPI are required to take two social science courses in order to graduate. These
courses are selected from economics, psychology, system dynamics, and public policy.
Therefore, introductory classes can be large (88 in my case), and most students are there
because they have to be. It should also be noted that each course is 7 weeks long and meets 4
times a week.
How did the students like it, and did they learn “more?” This is difficult to answer. Certainly
class attendance was well above the norm for other social science classes, but this could have
been because there was no textbook and most of the material was presented in class. Course
evaluations were positive (in spite of two exceptionally long exams and a research essay!), with
many finding the subject matter “interesting.” But unfortunately we did not do a controlled
experiment (e.g., same instructor using traditional methods). A number of students appreciated
the insights the course gave to our current economic situation (except for the outlook for
computer science engineers!), and mentioned how they were better able to appreciate the news
and discussions they have had with business people. However, the course has not yet
attracted students to the system dynamics courses.
What worked and what did not? This was my first experience teaching undergraduates (rather
than engineering and business graduate students). In retrospect, too much material was
covered, and it was therefore done too quickly. In particular, the system dynamics needed to be
presented more slowly and in less detail. The complete commodity/monopoly/oligopoly model
was probably too advanced for an introductory course, but would fit well in an intermediate
micro course. In addition, the students did not get any hands-on experience with the models. In
other courses, | have made much greater use of games (Fishbanks, Beer Game, B&B) followed
by models of the games that students (in part) developed and could experiment with. | believe
that this approach could work well in economics, although at WPI the class size and course
schedule presents implementation difficulties in the introductory courses.
Next Steps. | am scheduled to teach this course again in the spring of 2004. That will give me
a chance to revise and improve the course materials. After that experience, | will have a much
better set of lecture notes, models, and exercises for teaching microeconomics using system
dynamics. In the interim, | would like to share experiences, ideas, and materials with others
teaching economics.
References
Baumol, William). and Alan S. Blinder, Economics: Principles and Policy (9" Edition), Mason,
Ohio: South-Western, 2003.
Ford, Andrew “System Dynamics and the Electric Power Industry,” System Dynamics Review
Vol. 13, No. 1, Fall 1997.
Forrester, Jay W. World Dynamics, Cambridge, MA: Wright-Allen Press, 1971.
Forrester, J ay, W. “An Alternative Approach to Economic Policy: Macrobehavior from
Microstructure,” in Kamrany, N. and R, Day (eds.), Economic Issues of the Eighties,
Baltimore, MD: J ohns Hopkins University Press, 80-108, 1979.
Forrester, Nathan B. A Dynamic Synthesis of Basic Macroeconomic Theory: Implications for
Stabilization Policy Analysis. PhD thesis, MIT Sloan School of Management, Cambridge,
MA (unpublished), 1982.
Lyneis, James M. “"System Dynamics for Market Forecasting and Structural Analysis," System
Dynamics Review, Vol. 16. No. 1, Spring 2000.
Mankiw, N. Gregory. Principles of Microeconomics (2"? Edition), Harcourt College Publishers,
2001.
Meadows, Dennis L. Dynamics of Commodity Production Cycles. Cambridge, MA: The MIT
Press, 1970.
Meadows, Donella H., Meadows, D.L., Randers, J., and Behrens, W. W. The Limits to Growth,
New York: Universe Books, 1973.
Meadows, Donella H., Meadows, D. L., and Randers, J. Beyond the Limits. Post Mills, VT:
Chelsea Green Publishing Company, 1992.
Mass, Nathaniel). Economic Cycles: An Analysis of Underlying Causes. Cambridge, MA:
MIT Press, 1975.
Paich, Mark and Sterman, J ohn D. “Boom, Bust, and Failures to learn in Experimental
markets,” Management Science, Vol. 39. No. 12, 1993.
Parkin, Michael. Microeconomics (5" Edition), Reading, MA: Addison-Wesley, 2000.
Radzicki, Michael J . “Mr. Hamilton, Mr. Forrester, and a Foundation for Evolutionary
Economics,” J ournal of Economic Issues (Vol 37, No. 1), March 2003.
Sterman, J ohn D. Business Dynamics: Systems Thinking and Modeling for a Complex World,
McGraw-Hill, 2000.
Stiglitz, J oseph E. and CarlE. Walsh, Principles of Microeconomics (3% Edition), New York:
W.W.Norton & Company, 2002.
Back to the Top
CC BY-NC-SA 4.0