Systems Thinking Archetypes (Generic Structures)
Archetype Behavior Reinforcing Loop
Ÿ growth or decline of the “state of the system”
Examples Ÿ Population growth or decline Ÿ Sales growth or decline Ÿ Microphone feedback
Policy Advice Ÿ recognize that reinforcing feedback creates exponential growth that can bring on pressures to retard growth Ÿ they are two-edged swords that can work for us or against us
Reinforcing Loop
Reinforcing Loop
net increase rate S
R
(population growth)
S state of system1
S R
birth rate
population
S
Balancing Loop S
state of system2
goal O
B
S
gap S
Balancing Loop
Ÿ goal seeking Ÿ regulates system behavior Ÿ opposes system change from set target or goal
Ÿ Heating or cooling system; setting thermostat to regulate room temperature Ÿ Economic growth; Federal Reserve modifying interest rates to meet growth target
corrective action2
Ÿ recognize that balancing loops regulate the system to provide stability and, on the other hand, resist change
Balancing Loop (heating) actual temperature
thermostat setting
S O
B
S temp gap S
furnace operation Systems Thinking Archetypes © 2003 Continuous Improvement Associates
page 1
Archetype Behavior Limits to Growth (at right are two forms of this structure)
Ÿ Initial growth in the “state of the system” is eventually limited or falls off due to a resource constraint affecting or due to a "side effect" of the growing action.
Examples
Policy Advice
Ÿ Sales limited by service quality Ÿ World population growth limited by resources
Ÿ recognize that nothing grows forever Ÿ be aware of future limits and the pressures they will cause Ÿ leverage for growth is often in looking for ways to reduce or remove the limits, rather than by pushing harder on the growth loop
Limits to Growth Limits to Growth action constraint
limiting constraint
S O limiting "side effect"
O
S R
growing action O
state of system S
R
S
net increase rate3 S S
B
S limiting action
R
carrying capacity O
B fractional net increase S
R
practicing tennis S O
R
S tennis ability S
S
(rabbit population)
(tennis ability)
S O tennis elbow
resource adequacy
Limits to Growth
Limits to Growth elbow use limit
S state of system3
physical abilty limit
B
O S improvement potential
rabbit birth rate S S
R
resource effect on fertility
B
S rabbit population
O
Systems Thinking Archetypes © 2003 Continuous Improvement Associates
page 2
S
S resources per rabbit
resources available to rabbits
Archetype Behavior
Examples
Policy Advice
Ÿ look for unintended O consequences of state of actions to relieve painsystem4 goal4 S ful symptoms R O Ÿ look for root causes S unintended that are responsible B gap4 consequence for the symptoms S Ÿ when addressing S symptoms, look for corrective ways to reduce negaaction4 tive impacts Ÿ take action to both Fixes That Fail relieve immediate pain and work on long-term (maintenance budget) root causes
Fixes That Fail
Fixes That Fail
Ÿ actions produce the desired correction in the short-term, but have make the problem worse in the long-term Ÿ typically a result of addressing problem symptoms rather than root causes
Ÿ taking drugs (whether narcotics or pain-relievers) may make a person feel better immediately, but does not address root causes and makes one feel worse in the long run Ÿ corporate downsizing reduces costs immediately, but impairs the organization’s ability to perform in the future Ÿ road-building relieves traffic congestion in the near-term, but attracts more growth to again build up congestion; the larger area also makes existing public transportation inadequate
Addiction
Ÿ occurs when the short term solution has to be taken again and again because the effect wears off Ÿ the need to take recurring action acts to drive the “Fixes That Fail” dynamic Ÿ exacerbates the effects of the “Fixes That Fail” dynamic
Ÿ addictive drugs wear off and a per- Ÿ same as for son needs more; they damage body “Fixes That Fail” and mind Ÿ expediting deliveries negatively impacts other products which must also be expedited and they interfere with normal working of the production & distribution system Ÿ city growth and development without impact fees leads to infrastrucAddiction ture backlogs and the need for more growth & development (growth)
Addiction S
action based on short-term solution O S O effect of short-term solution
B
S perceived state of the system
R S
O
state of the system
B
S expenditures budget
S
R
S
equipment breakdowns
B
O overspending
O
O spending on maintenance
approve development & promote growth
O S development
B
decay of effect of short-term solution S
S taxes from initial development O perceived infrastructure backlog
S
S long-term infrastructure needs R S
actual infrastructure backlog
Systems Thinking Archetypes © 2003 Continuous Improvement Associates
page 3
B
developments completed S
Archetype Shifting the Burden
Behavior
Examples
Policy Advice
Ÿ actions taken to reduce symptoms reduce the ability to take action for the long term. Ÿ can be either shifting the burden to short-term, rather than long-term, solutions or shifting the burden to an intervenor, rather than to building system capability Ÿ not only exacerbates the effects of the “Fixes That Fail” dynamic, also it reduces the ability to take action for the long term and escape symptomatic solutions
Ÿ taking drugs makes a person feel better but reduces ability to improve life skills Ÿ giving a man a fish vs. teaching a man to fish promotes dependency Ÿ corporate downsizing reduces costs immediately, but reduces the ability to develop new products Ÿ HR deals with a manager’s problem with low employee performance, rather than assisting and training manager Ÿ government insures bank deposits and bails out banks rather than requiring sound banking practices
Ÿ same as for “Fixes That Fail”
Shifting the Burden (to an external intervention solution) external intervention O B
S effect of external intervention O
perceived S state of the system2 S
ability to pursue internal solution
R
effect of internal solution
B
S O
S
internal solution
Shifting the Burden Shifting the Burden
Shifting the Burden
(drugs to treat the short-term symptom)
(charity external intervention vs. teaching self-reliance)
O
(to the short term, symptomatic solution) action based on short-term, symptomatic solution O
man accepts a fish
take drugs
S drug high
B
man has food
B O
S perceived quality of life
B
R S
use discipline to improve life skills
O R
hunger O
actual quality of life S
improve life skills
B
effect of short-term solution1
O
O
B
S O
S
S
S
S
motivation to provide for oneself
man has long-term source of food S S
man learns to fish
Systems Thinking Archetypes © 2003 Continuous Improvement Associates
perceived S state of the S system1
B
R
effect of long-term solution S
O action based on long-term, fundamental solution page 4
S
ability to act based on long-term solution
Archetype
Behavior
Policy Advice
Examples
Eroding Goals lowering the goal
Eroding Goals
Ÿ there are two ways to close the performance gap: improve performance lower the goal Ÿ also known as the “boiled frog” syndrome
Ÿ it’s easier to lower quality targets than increase quality Ÿ it’s easier to let federal budget deficits keep rising than to increase taxes and/or decrease spending Ÿ it’s easier to relax environmental standards than reduce pollution
Eroding Goals
Eroding Goals
(on-time delivery)
(quality of life)
lowering on-time delivery goal
lowering quality of life goal S
S
O on-time delivery goal
B
S
temptation to lower on-time delivery goal S
O quality of life goal
B
S on-time delivery goal-actual
O actual on-time delivery S
Ÿ when performance is declining, examine whether it could be because goals are being relaxed Ÿ make goals clearly visible Ÿ examine the way goals are set and who sets them Ÿ goals located outside the system are less vulnerable to erosion Ÿ reward setting “stretch” goals & don’t penalize if not met, which teaches people to not set stretch goals.
B
S pressure to improve on-time delivery S
corrective action to improve delivery perofrmance
temptation to lower quality of life goal S
quality of life gap O actual quality of life0
B
S pressure to improve life skills
S
S self-improvement & learning
Systems Thinking Archetypes © 2003 Continuous Improvement Associates
page 5
S
O performance target
B
S
temptation to lower the goal S
performance gap O S actual performance S
B
pressure to improve
taking S corrective action to improve performance
Archetype Behavior Path Ÿ once one entity Dependence (person, prodor uct, organiza“Success tion, company, to the or country) gets Successful” ahead, it’s easier to get even further ahead because better performance provides more resources and a greater ability to improve performance Ÿ a “figure 8” is a reinforcing loop
Examples
Policy Advice
Ÿ the “rich get richer ...” phenomenon Ÿ monopolies increase market share (but reduce competition) Ÿ economic cluster formation Ÿ “good student” performance over “bad student” performance Ÿ home vs. work involvement Ÿ manufacturing improvement favored over engineering because it’s faster and easier
Ÿ understand that this structure requires intervention to produce and maintain a “level playing field” Ÿ examine how the system has been set up for “winner-take-all” competition. Ÿ find ways for teams to collaborate rather than compete
Path Dependence or "Success to the Successful" S
A's success
A's accomplishments
B's success S O allocation of resources to A, rather than to B
R1
competitor price
our price S
S
S
O
R1 Competitor Market Share
O our market share vs. competitor market share
R2 Our Market Share
our unit costs O
success of student A S
competitor O cumulative production
B's accomplishments S
O
S
resources to B
Path Dependence Example: self-fullfilling prophecy for student success
Path Dependence Example: the "experience curve."
competitor unit costs
R2
S
resources to A
S
S
perceived benefit from investing in student B
perceived benefit from investing in student A
S S
R1
O
R2
time spent with student A vs. student B
Student A Success
Student B Success
S our cumulative production
S investment in student A
Systems Thinking Archetypes © 2003 Continuous Improvement Associates
page 6
O investment in student B
success of student B S
Archetype Behavior Escalation
Ÿ individual action that attempts to increase security or performance at the expense of another (e.g., a competitor) results in less security or decreased performance over the long run. Ÿ this structure is brought on by unbridled competition and can only be escaped by promoting cooperation based on mutual interests Ÿ a “figure 8” is a reinforcing loop
Policy Advice
Ÿ arms race increases weaponry (but decreases security for all) Ÿ price wars increase sales and market share (but decrease profits for all) Ÿ regions compete on the basis of low taxes & less regulation for a limited number of jobs (but leads to infrastructure backlogs for all regions) Ÿ cities compete for sports teams based on expenditures to support the teams (but leads to “build us a stadium or we go elsewhere” blackmail and higher costs for all cities)
Ÿ understand that overcoming this structure requires cooperation toward a larger goal that benefits competing parties Ÿ examine how the structure reduces results in the long run (e.g., in an arms race there is less security) Ÿ examine whether perceptions of opponent’s intent is accurate (perhaps they see themselves as simply responding to your action) Ÿ examine whether perceptions of opponents ability is accurate (e.g., perhaps their arms potential is not as great as perceived)
Escalation Example:
other nations weapons stockpiles
S
U.S. weapons production S
U.S. Weapons Buildup
O S U.S weapons superiority over other nations
B1
S other nations' perception of threat
S
O
results of A relative to B
B acts to counter threat
B2
S
S threat O to A
S
threat to B
Escalation Example: price war B's sales
A's sales
S
S
S
B1
other nations' weapons production S
company A price cuts S
O B2 Company A market share Company B Competition of company A Competition vs. company B Based on S
Based on Price
Price O
U.S. O perception of threat
B's results S
A acts to counter threat
B2 Other Nations' Weapons Buildup
A's results
S
the "arms race"
U.S. weapons stockpile
B1
Escalation
Examples
company A's perception of competitive threat
Systems Thinking Archetypes © 2003 Continuous Improvement Associates
page 7
S company B's perception of competitive threat
company B price cuts S
Archetype Behavior Tragedy of the Commons
Policy Advice
Examples
Ÿ rational action by individuals to improve individual performance results in destroying the ability of the whole system to perform and also destroys the ability of individuals to perform as the system is destroyed.
Ÿ overgrazing on land destroys the land’s ability to grow feed Ÿ understand that overcoming this structure requires Ÿ overfishing depletes the fish stock and the ability of fish to cooperation toward a larger goal that manages comreproduce ... in this case the “market” signal is increased mon resources and benefits competing parties price, which leads to even more fishing and more rapid Ÿ apportion the expense of long-term collective loss to destruction of the commons individuals or limit individual activity (grazing fees, fishŸ increasing individual farm production by increasing land in ing limits, land allowed in production, development production and improving technology has the goal of impact fees) increasing farm income ... but the increased supply in the presence of inelastic demand decreases prices so all farmers go broke without government subsidies or small farmers get bought out by larger farmers (which does not decrease the supply of land) Tragedy of the Commons Ÿ groups benefit more from getting more resources from a (the market for farm commodities) common organizational resource pool, but overload the Farm A Net common resource (e.g., quality, HR, reproduction services) S Income S Ÿ individual engineering teams maximize the electrical funcFarm A tions they’re designing by drawing more on the electrical Commodity power system, but overall exceed the electrical system’s Profit per R1A Unit R2A ability to supply power Farm A Ÿ firms benefit from economic activity that causes pollution, O Production S Farm A Farm A Quantity but increase negative health impacts for all Production Efficiency Ÿ developers profit from more development that uses Farm A S Cost per common infrastructure, but overwhelm infrastructure Unit S O
Farm A Land in Production & Efficiency
Tragedy of the Commons Tragedy of the Commons
(fishing)
S
S A's activity
R1
net gains for A S S B3
R1
S
S total activity
resource or demand limit
S A's fishing
gain per individual activity
O
S
total fishing
S
S
S
R2
S net gains for B
effort O required per fish caught
B4 S
B's fishing S
O S B's profits
S
S Farm B Production Quantity
R2
Systems Thinking Archetypes © 2003 Continuous Improvement Associates
Farm Commodity Price per O Unit Farm B: B3B Produce More, Lower Price Farm B Land in Production & O Farm B Efficiency Cost per Unit S R1B R2B
Total Production
B3 S
B4
B's activity
A's profits S O
sustainable ocean fishing capacity
B3A
O S Farm B Efficiency Farm B Commodity Profit per Unit S Farm B Net S Income
Farm B Production
page 8
Total Demand
Farm A: Produce More, Lower Price S
Archetype
Behavior
Ÿ a growing action encounters multiple “Limits to Growth” Ÿ addressing one limit (this diagram is puts more pressure shown with the on other limits typical price, Ÿ As Forrester said, service quality, “There are no utoproduct quality pias in social features, but systems.” there are others, (e.g., delivery delay, o community participation / citizenship) The Attractiveness Principle
Examples
Policy Advice
Ÿ no company can be all things to all people (lowest price, best product, best service); it must decide and focus on its “value proposition” Ÿ no region can be all things to all people (lowest taxes, lowest housing prices, best quality of life, best jobs)
Ÿ practice “strategic unattractiveness” ... that is, decide on the features that will make the product or region less attractive and balance out the attractiveness of the other features that are more desirable to customers or the other features that are necessary to support the organization’s purpose/mission.
o
invest in service capacity
S service capacity S
B7
service demand
Service Quality Burden S
demand generating activity
S costs O S
Net Revenue (funds available to generate demand) S Product Quality Burden
B8
Service Focus
B2
S
Service Quality Erosion S customer demand
R1
price
B3
product demand
service quality standard
B5
service quality
Overall Product/Service Attractiveness Product Quality Erosion
S
O
S
S
Demand Generation S
O
S
S product quality
O
O
B6
product development quality standard
S Product O product Focus S development invest in product capability development S capability
B4 Scarcity Premium
o
o
Systems Thinking Archetypes © 2003 Continuous Improvement Associates
page 9
Tragedy of the Commons
Farm A Production Quantity
Farm A Efficiency
Farm A Production
O
B4A
S
O
S
B3B S
Farm B Land in Production & Efficiency
B5A
S R1B
O
R2B
O S Farm B Efficiency Farm B Commodity Profit per Unit S Farm B Net S Income S
Farm B Production
B7A S
Chart: Trends in Government Farm Supports
Farm A Tax Breaks
Daryll E. Ray, agricultural economist with the Agricultural Policy Analysis Center at the University of Tennessee, testified before the House Farm Total Committee on Agriculture on 2/14/01 on "Crop Agriculture Faces LongO Tax Commodity Farm Price Breaks Term Price and Income Problems," (see excerpts from his testimony S O Income Supports below; italics and underlings are as in his original). Farm B Tax http://agriculture.house.gov/hearings/h10214w2.htm suggests that we are in denial. We are in denial that anything long-term B7B B5B is to blame for the devastatingly low prices and low market incomes in crop agriculture. We are more than willing to blame agriculture’s probFarm B Loans lems on the Asian Crisis, exchange rates, energy prices, or anything else that comes along. Others blame the level of loan rates, emergency O payments, crop insurance, etc., etc.
S
Farm Commodity Price per Unit
Farm B Cost per Unit
O
Total Demand
Farm B: Produce More, Lower Price O
Low Interest Loans
Farm A Loans
O
S O Farm A Land in Production & Farm A: Efficiency S Produce More, Lower Price B3A Total Production
S Farm B Production Quantity
Farm A Income Protection
S
Farm A Cost per Unit
S
Tragedy of the Commons structure and the Farm Policy Death Spiral
(the market for farm commodities) S Farm A Net S Income S Farm A Commodity Profit per R1A Unit R2A
Technology Support
Farm B Income Protection B4B
B6A Farm A Tech B6B Farm B Tech
the disruption. Other industries self-adjust. Why doesn’t crop agriculture? That is the real question.
growth of crop supply compared to crop demand and b) the price responsiveness of supply and demand.
This time in history and this stage of the farm bill debate cycle provide the perfect opportunity to make a definitive determination of the how the grain markets work. For the first time in nearly seventy years, markets have been free to reveal the true supply and demand behavior of U.S. crop markets.
Total crop acreage (supply) is unresponsive to price declines in short or longer-run.
Farmers have no incentive to reduce production as prices decline. From an individual farmer’s standpoint, there is no rational reason for him/her to leave land idle because crop prices have declined ... . Each farmer produces too litI believe that the market experience of the last tle to affect total supply and therefore price, so four years shows that crop agriculture is just as any reduction in his output means less prone today to chronic price and market income revenue. … problems as it was when farm programs were Demand is unresponsive to price changes. instituted decades ago. My mission in this testiBecause it is essential for life—like insulin for a mony is to explain why I believe that is so. diabetic—price is of little consequence. Food Agriculture’s price and income troubles are comes first. We will pay whatever is required to quickly understood by considering a) the rate of obtain it. But once we have enough, will not buy
The implication being that: once the—you-name-it-disruption—subsides or is remedied, agriculture will be just fine. That is nonsense. There are always disruptions. There are disruptions in agriculture, the auto industry … every industry. At this stage of the farm policy debate, discussion should not center on this or that disruption, but on the ability of agricultural markets to make adjustments irrespective of the exact nature of
significantly more total food, no matter how far the collective price of food has dropped. Is this true for other products? Of course not. Typically, a price drop greatly expands the quantity demanded of an industry product. ...
possible because of taxpayer support, assures relatively rapid shifts in supply. Under this combination of price unresponsive supply, price unresponsive demand and supply shifting faster than demand, prices and income can be Agriculture is unique. Much of that uniqueness expected to be chronically depressed. This is is rooted in two characteristics: (1) cropland will not a short-run problem. be used to grow crops and (2) food is essential Left to itself, crop agriculture would continue its for life but the quantity needed is finite. These downward spiral, bankrupting successive farmand other supply and demand characteristics ers on a given piece of land, forcing bank forevirtually assure that there will be little change in closures, and, in general, wreaking devastation total crop acreage and little change in the quan- on ALL rural areas. It would be a disaster of a tity demanded as prices fall, even by 40 permagnitude that would be well beyond political cent over a four year period. acceptability. Those that believe otherwise also Periodically, crop exports will grow for several believe that supply and demand quickly adjust to lower prices. If that were true, then crop years at relatively high rates, but usually they agriculture would self-correct. But it is not and do not. Technological advances in crop agriagriculture doesn't. It really is that simple. culture, most of which is directly or indirectly
Systems Thinking Archetypes © 2003 Continuous Improvement Associates
page 10
