Systems Thinking in Design: What if Governments Really Used It?

Transcription

1 Systems Thinking in Design: What if Governments Really Used It? George P. Richardson Professor Emeritus, State University of New York 1

2 2

3 There are Lots of Systems Thinkers Fields that say systems Systems analysis General systems theory Viable systems analysis Living systems theory Soft systems methodology Critical systems heuristics Critical systems science Sociotechnical systems System dynamics (that s me) Fields that imply systems Soft operations or operational research (Soft OR) Strategic options development and analysis (SODA) Strategic assumption surfacing and testing (SAST) Hierarchy theory Interactive planning Cognitive mapping. Cybernetics (that s Bob) 3

4 A Quick Introduction to System Dynamics Dynamics Graphs over time Feedback loops What goes around comes around Stocks Accumulations The endogenous point of view Looking inward. System as cause. Decisions Action Perceptions State of the system 4

5 The Classic Policy Planning Loop Goal Perceived state Perceived gap Planned action to reduce gap Implicit, unstated goals Actual state of the system Changes in the State of the system Intended actions Implemented action 5

6 The Classic Loop with Complications Goal Perceived state Actual state of the system Perceived gap Autonomous changes in the state of the system Changes in the State of the system Planned action to reduce gap Intended actions Implemented action Implicit, unstated goals 6

7 The Classic Loop with Complications Goal Perceived state Actual state of the system Perceived gap Autonomous changes in the state of the system Changes in the State of the system Planned action to reduce gap Intended actions Implemented action Unintended actions Implicit, unstated goals 7

8 The Classic Loop with Complications Goal Perceived state Actual state of the system Perceived gap Autonomous changes in the state of the system Changes in the State of the system Planned action to reduce gap Intended actions Implemented action Unintended actions Implicit, unstated goals Ramifying effects 8

9 It Can Get Really Complicated! <xxx> Govt willingness to act against tobacco - Tobacco taxes Pro-tobacco contituencies Tobacco revenues Tobacco production capacity Govt awareness of tobacco health risk Trend in tobacco company revenues - Tobacco marketing activities - Health care costs Anti-tobacco constituencies Researchers awarness of tobacco health risk Smokers - - People starting smoking People quitting Smoking as a - smoking social norm - Tobacco products availability Public awareness of tobacco health risk Pressure on tobacco companies to reduce marketing activities Funding for tobacco health research Health insurers coverage of tobacco quitting costs Funding for tobacco control programs Tobacco control programs <xxx> 9

10 Prejudice and Minority Achievement (Myrdal, Merton) Prejudice (R) Hope or despair Aspirations of the minority Discrimination (R) Prejudice Achievements of the minority (B) Striving Minority perceptions of the gap Opportunities for the minority Minority efforts to achieve 10

11 Stocks and Flows in Global Climate (A deep insight from bathtubs, faucets, and drains) Thought experiment: Economic activity CO2 annual production Atmospheric CO2 Uptake of atmospheric CO2 Outflow is high, inflow is low (zero), Temperature so stock declines. declines after CO2! Incoming solar heat energy (heating) Global heat energy Outgoing global heat energy (cooling) How long after depends on how big this gap is how long we wait before we act!! 11

12 Stocks and Flows in Health Priorities Society's Health Response General protection Targeted protection Primary prevention Secondary prevention Tertiary prevention Safer, Healthier People Vulnerable People Afflicted People without Complications Afflicted People with Complications CDC: Enhancing Health Protection! Dying from complications 12

13 New York, Chicago & Philadelphia, Philadelphia Chicago New York

14 Populations of Selected U.S. Cities, Atlanta Baltimore Birmingham Boston Buffalo Cincinnati Cleveland Minneapolis Newark Pittsburgh San Francisco St. Louis Washington

15 Dutch Urban Dynamics Population 1,000, , , , , , , , , , Amsterdam Rotterdam Den Haag Utrecht Eindhoven Nijmegen Groningen Enschede Zaanstad Tilburg 15

16 Urban Decay Camden, NJ, USA 16

17 Forrester s City: Endogenous Dynamics 17

18 A Simple Urban Model: Urban1 construction structures demolition construction demolition Population Outmigration Inmigration Net births 18

19 Area A Simple Urban Model: Urban1 land multiplier construction structures 20,000 demolition structure 40,000 30,000 10,000 structures Land fraction occupied Time (year) structures : es construction demolition Population Outmigration Inmigration Net births 19

20 Area A Simple Urban Model: Urban1 land multiplier Land fraction occupied construction structures labor force multiplier demolition Jobs Labor to job ratio Attractiveness from jobs multiplier 40,000 structure 4 M people Selected Variables Labor force 20,000 structure 2 M people construction demolition Population Outmigration Inmigration 0 structure 0 people Time (year) structures : es & Population structure Population : es & Population people Net births 20

21 Area land multiplier construction A Simple Urban Model: Urban1 land multiplier 2 people/job 10,000 structure 100,000 structure 400,000 people Land fraction occupied 1.4 people/job 5,000 structure 50,000 structure 200,000 people 0.8 people/job 0 structure 0 structure 0 people construction structures labor force multiplier demolition Jobs Time (year) demolition Labor to job ratio Labor force Population Outmigration Attractiveness from jobs multiplier Inmigration availability multiplier Households to housing ratio Net births Attractvieness from housing multiplier 21

22 Dynamics of the Little Urban Model Growth, Stagnation & Decline with High Unemployment 2 people/job 10,000 structure 100,000 structure 400,000 people 1.4 people/job 5,000 structure 50,000 structure 200,000 people 0.8 people/job 0 structure 0 structure 0 people Population Labor-to-Job Ratio es Time (year) 22

23 Forrester s Urban Dynamics Core Structure Underemployed Labor Managerial- Professional Underemployed housing Worker housing Premium housing New enterprise Mature businesses Declining industry 23

24 es Urban Dynamics Key Stocks production unit 20,000 15,000 10,000 5,000 housing units 600, , , , Time (year) New Enterprise : Urb Dyn base Mature : Urb Dyn base Declining Industry : Urb Dyn base 600,000 Populations 0 450, Time (year) 300,000 Underemp : Urb Dyn base Worker : Urb Dyn base Premium : Urb Dyn base 150,000 workers Time (year) Underemployed : Urb Dyn base Labor : Urb Dyn base Manager Professional : Urb Dyn base 24

25 Area land multiplier Favorite Policies in the 1960s Low cost housing and Job programs land multiplier BCN Land fraction occupied construction LPH LPBS structures JPBS labor force multiplier demolition Jobs LPF BDN Labor to job ratio Labor force Attractiveness from jobs multiplier HCN construction availability multiplier demolition HDN Households to housing ratio HS OMN Population Outmigration DN BN Net births Inmigration IMN Attractvieness from housing multiplier 25

26 Building low cost housing (t 40) Population structures 400,000 6, ,000 4,500 people 200,000 structure 3, ,000 1, , Time (year) Population : Urban1 Population : More housing constr Little impact, no improvement Time (year) structures : Urban1 structures : More housing constr Labor to job ratio 75, structure 50,000 people/job , Rockefeller College of Time Public (year) Affairs and Policy : Urban1 : More housing constr Time (year) Labor to job ratio : Urban1 Labor to job ratio : More housing constr 26

27 Wildly successful jobs program (t 40) Population structures 400,000 6, ,000 4,500 people 200,000 structure 3, ,000 1, Time (year) Population : Urban1 Population : More housing constr Population : More jobs Time (year) structures : Urban1 structures : More housing constr structures : More jobs 100,000 Short run benefit, no long run effect 2 Labor to job ratio 75, structure 50,000 25,000 people/job Rockefeller College of Time Public (year) Affairs and Policy : Urban1 : More housing constr : More jobs Time (year) Labor to job ratio : Urban1 Labor to job ratio : More housing constr Labor to job ratio : More jobs 27

28 Why Doesn t the Jobs Program Work? Area land multiplier land multiplier construction structures The red loops naturally Land fraction compensate occupied for the increase in jobs and return the Labor-to-Job ratio back to where it was. units labor force multiplier demolition Jobs Households Labor to job ratio Labor force Attractiveness from jobs multiplier construction demolition Population Outmigration Inmigration Households to housing ratio Net births availability multiplier Attractvieness from housing multiplier 28

29 What policies prevent urban decay and improve long run employment? 2 people/job 10,000 structure 80,000 structure 400,000 people Population 1.4 people/job 5,000 structure 40,000 structure 200,000 people es 0.8 people/job 0 structure 0 structure 0 people Labor-to-Job Ratio Time (year) 29

30 Area land multiplier land multiplier BCN Land fraction occupied An unpopular policy: Knocking down housing construction LPH LPBS structures JPBS labor force multiplier demolition Jobs LPF BDN Labor to job ratio Labor force Attractiveness from jobs multiplier HCN construction availability multiplier demolition HDN Households to housing ratio HS OMN Population Outmigration DN BN Net births Inmigration IMN Attractvieness from housing multiplier 30

31 demolition (t 40) (Grey curves) Population structures 400,000 6, ,000 4,500 people 200,000 structure 3, ,000 1, Time (year) Population : Urban1 Population : More housing constr Population : More jobs Population : demolition 100,000 Makes things better! Why? Time (year) structures : Urban1 structures : More housing constr structures : More jobs structures : demolition 2 Labor to job ratio 75, structure 50,000 people/job , Time (year) : Urban1 : More Rockefeller housing constr College of Public Affairs and Policy : More jobs : University demolition at Albany Time (year) Labor to job ratio : Urban1 Labor to job ratio : More housing constr Labor to job ratio : More jobs Labor to job ratio : demolition 31

32 Another unpopular policy: Reducing our biases that favor people structures Area land multiplier land multiplier BCN Land fraction occupied construction LPH LPBS structures JPBS labor force multiplier demolition Jobs LPF BDN Labor to job ratio Labor force Attractiveness from jobs multiplier HCN construction availability multiplier demolition HDN Households to housing ratio HS OMN Population Outmigration DN BN Net births Inmigration IMN Attractvieness from housing multiplier 32

33 Combined, the last two policies eliminate decay and lower unemployment 2 people/job 10,000 structure 80,000 structure 400,000 people Population 1.4 people/job 5,000 structure 40,000 structure 200,000 people es 0.8 people/job 0 structure 0 structure 0 people Labor-to-Job Ratio Time (year) 33

34 A Few Systems Thoughts (Empirical results from sixty years of real-world system dynamics applications) Short run behavior often runs contrary to long run behavior Better before worse, Worse before better High-leverage points in complex systems are hard to find And if we find one, we usually push it in the wrong direction We usually think pretty well in causal sequences We do not think well in causal loops We usually believe our problems are caused by outside forces We d rather not realize we cause or exacerbate our own problems A dynamic systems view doesn t make moral choices easier But it makes them much clearer! 34

35 Policy Resistance of Complex Systems 35

36 Policy Resistance of Complex Systems 36

37 Policy Resistance of Complex Systems 37

38 Policy Resistance in Complex Systems 38

39 Predominant Mode of Analysis Exogenous Endogenous Final Thought The Endogenous Point of View Striving for understanding and leverage, but failing JK Accepting fate, Predicting, Preparing KL Exogenous Achieving understanding and leverage JJJ Confused, Misguided, Misguiding LLL Endogenous True (Predominant) State of Affairs 39