1. BA 511 Politics & Games A Sampling of Managerial-Related Issues

2. Firms as Governance Structures Coase-Williamson-Radner Coase-Williamson-Radner  Aside on transactions costs/nature of firm  Markets:  Strong incentives & contract dispute  Weak administrative control/planning  Firms:  Strong on administrative control  Weak on incentives & contract dispute  Optimal mix make-buy; optimal planning; market in firm; pre-contract & post-contract issues 2

3. Decision Mechanism Problems Case: 1912 Election Case: 1912 Election  Wilson 42%; Roosevelt 27%; Taft 23 %; Debs 6%  LA “David Duke”; 2008 GOP Primaries … Arrow Theorem: No mechanism (voting; market) can satisfy basic conditions: Arrow Theorem: No mechanism (voting; market) can satisfy basic conditions:  (i) no dictator; (iii) all relevant preferences/options matter; (iii ) consistency outcomes with order/arrangement; (iii ) consistency outcomes with order/arrangement; (iv) consistency from individual to aggregate preferences;  AT: violation inevitable; which is most acceptable?  Mechanisms subject to manipulation Management Implications? Management Implications?  Proposal order; proposal voting rules; decision chain;

4. Costs of Decisions: No Magic in 50% Rule Percent in Agreement Costs of additional Parties to agreement Benefits of additional Parties to agreement 100% 0%M*

5. Policies (rules) v. Discretion Tradeoffs Cases: Cases:  NWA Detroit Tarmac Debacle  Coach K: “people set rules to keep from making decisions”  End-of-year budget bait-switch Econ Issues Econ Issues  How costly is discretion?  Influence on incentives/forward-looking behavior  How widespread/systematic is the event/problem  How costly is adherence to policy  Specifying contingencies; writing rules (TC)  Paradox: Rule for departing from rule? 5

6. Change Politics & Org Inertia Cases: Cases:  Branch Rickey: Brooklyn Dodgers;  Jack Welch: GE; UTA President Basic Principles Basic Principles  Optimal rate of change  Interest group dynamics  Who can you count on  Higher level management  “Lieutenants” 6

7. 7 Politics & “Games” Examples of Settings and Behaviors Group/individual Cooperation Dynamics (inter/intra group, unit) Group/individual Cooperation Dynamics (inter/intra group, unit)  Prisoner’s Dilemma; Hostage’s Dilemma  Promises/Threats/Commitments Cooperation/Non-coop: Prisoner’s Dilemma; Hostage’s Dilemma Cooperation/Non-coop: Prisoner’s Dilemma; Hostage’s Dilemma  Pricing games; Inter/Intra-unit or group dynamics; Location Games Location Games  Geographic; Proposal Bidding-Bargaining; Bidding-Bargaining; Employment Processes (hiring; monitoring) Employment Processes (hiring; monitoring)  Signalling-Filtering Misc Misc  Pricing, Ads, Competition, Entry, Exit  Parent-child; Spouses; Siblings Many governance (and market) settings involve strategizing of 2 (or more) players where their decisions are “actively interdependent” Not just “playing against market” with fixed prices/behavior Think chess, poker, or rock-paper-scissors, not roulette

8. Essential Features of Games Case: Consider Poker or “Chicken” Case: Consider Poker or “Chicken”  Players  Information/beliefs of players  Actions/strategies available to players  Payoffs to moves/paths  Timing of moves/sequence or together  Manipulation of any of these elements by players  Solving: Putting self in other’s shoes 8

9. Solution Concepts & Game Basics: Thinking Ahead—Reasoning Back Case: Apollo 1 Aftermath Case: Apollo 1 Aftermath  Tom Hanks’ HBO series-- From the Earth to the Moon– U.S. space program from Mercury through moon landings. Strategic segment involves the aftermath of the Apollo 1 deaths of 3 astronauts during a launch pad tests:  A capsule fire during a routine test.  The fire resulted from a spark in wiring  The test under highly pressurized, pure oxygen capsule; contractor (North American) sent repeated warnings to NASA  Capsule reached temperatures over 1000 in 15 seconds  NASA planned to lay substantial blame on NA  NASA planned to lay substantial blame on NA  NA chief engineer argues for exposing NASA memos  NA executive, “no, we’re not and goes on to respond; we’re going to just take it”  Can you make sense of the boss’ decision? 9

10. 10 Manipulating Elements of the Game and Countering Manipulation Cases: Cortez ship burning; Poker betting Cases: Cortez ship burning; Poker betting  (Credible) Commitment & actions/beliefs Cases: Retail Stores; My Daughter Cases: Retail Stores; My Daughter  Limiting available actions  Expanding available actions Case: Proposal Consideration by Committees/Task Forces Case: Proposal Consideration by Committees/Task Forces  Agenda Control  Amendments

11. 11 First or Second Mover Advantage? Case: Princess Bride Case: Princess Bride  First Mover Advantage if manipulation of possible through changing game or beliefs of rival Cases: Sailing; NCAA Football Overtime Cases: Sailing; NCAA Football Overtime  Second Mover Advantage if information becomes available by rival’s move Poker? Poker?  Tradeoff: info manipulation v. info gathering

12. Smart but Not Too Smart Case: Email Phishing Schemes Case: Email Phishing Schemes  Too good to be true, probably is Case: Colts in 2010 Super Bowl Case: Colts in 2010 Super Bowl  Dominant strategy but a lot of mixing  Don’t outsmart yourself 12

13. “Location” Games (with extensive information and single dimension) Where to setup shop if consumer/voters positioned uniformly (or normally) along a road, given that competitor is trying to setup shop in best location also? Where to setup shop if consumer/voters positioned uniformly (or normally) along a road, given that competitor is trying to setup shop in best location also?  Simple Solution: Move to the middle (median), otherwise, competitor can locate just to the “busier” side and capture everyone on that side Examples: Variety of retail stores; primary & general election races; Examples: Variety of retail stores; primary & general election races; 13

14. “Location” Games (with very limited information) Where to setup shop if consumer/voters clustered in 10 large cities, you will locate in 5 and a rival firm will locate in 5 but agreements that divide cities are strictly prohibited and punishable by large punitive fines? Where to setup shop if consumer/voters clustered in 10 large cities, you will locate in 5 and a rival firm will locate in 5 but agreements that divide cities are strictly prohibited and punishable by large punitive fines?  Solutions: use of “focal points”  Stanford-Harvard MBA “Divide the Cities” Game  Variants: T. Schelling (Strategy of Conflict) where to meet in NYC? 14

15. Median Location/Voter Model Breakdown (multiple dimensions) Customer /Voter Preferences - Metrics No single dominant position due to asymmetry of preferences X: Traffic 0 Y: Income

16. Prisoner’s Dilemma Criminals arrested; interrogated w/no cooperation Criminals arrested; interrogated w/no cooperation  Choices: Confess/Don’t confess  1 Confesses = low/high sentences  2 Confess = moderate sentences  0 Confess = low/acquittal  Outcomes: cooperative (positive sum) solution v. competitive solution How to arrive at cooperative solution? How to arrive at cooperative solution?  With no/limited info & explicit cooperation?  With cheating? 16

17. Prisoner’s Dilemma-like Games Hostage’s Dilemma Hostage’s Dilemma  Multi-person version of PD Oligopoly Games Oligopoly Games  (pricing, ads, entry, …)  Cooperation (maybe implicit) leads to higher profits than competition 17

18. 18 Game Basics: Bargaining Tactics Case: Builder-Homeowner Case: Builder-Homeowner Understanding Incentives/Tradeoffs & Info Understanding Incentives/Tradeoffs & Info  Info asymmetry  Flex-price: flexibility of changes; no “hold-out problem”; wrong incentive for info problem  Fixed-price: Incentive to monitor & control expenses; “hold-out” problem on changes; incentive to cut corners Case: Car-buying Case: Car-buying  Signaling  Patience is best signal of patience

19. Ultimatum Game (and related) theory and experimentation Ultimatum Game (and related) theory and experimentation  Split of pot if 2 parties agree on split; 1 makes offer-1 accepts or declines offer;  Variations: size of pot; depreciation of pot; anonymity; repetition; wealth of participants; …  Money matters but not all that matters  Typical outcomes: bigger than 99:1, less than 50:50  Patience is a virtue  Patience is the best signal of patience  Tradeoffs in most bargaining situations 19

20. 20 Bargaining Tradeoffs (Home Building-Purchasing Case) Builder-Homeowner Builder-Homeowner  Builder info advantage  Buyer Choices:  Flex Price w/fixed percentage  Fixed-Price w/negotiated changes  Info/Incentive Tradeoffs

21. Insight on Solutions “Nash Equilibrium”: outcome where opponent doing best possible “Nash Equilibrium”: outcome where opponent doing best possible  Sequential  “Rollback”: Look ahead to last period and work back  Simultaneous  Iterative: step-by-step analysis of best choice given a decision by other  Repeated Simultaneous  Rollback + Iterative 21

22. Iterative Solutions to Simultaneous Game (PD Example) Payoffs = (Coke profits, Pepsi profits) Decisions: Price Low or Price High Pepsi Decision LowHigh CokeDecision Low 10,10 10,10 1,20 1,20 High 20,1 20,1 3,3 3,3

23. Solutions to Simultaneous First Iteration: Coke considers best choice if Pepsi sets low price (column 1) Pepsi Low Coke Low 10,10 10,10 High 20,1 20,1 Best choice for Coke, if Pepsi Sets Low Price

24. Solutions to Simultaneous Second Iteration: Coke considers best choice if Pepsi sets high price; Low is dominant strategy for Coke; Low better than high in both iterations Pepsi High Coke Low 1,20 1,20 High 3,3 3,3 Best outcome for Coke, If Pepsi Sets High Price

25. 25 Solving Sequential Games “Life must be understood backward, but … it must be lived forward.” - Soren Kierkegaard (Consider Chess as Example)  Diagram a game tree – simplify if needed  Start with the last move in the game  Determine the best course(s) of action for the player with the last move  Trim the tree -- Eliminate the dominated strategies  Repeat the procedure at the prior decision node(s) with the trimmed tree

26. 26 An Example: Market Entry Game Essentials: Game Essentials:  Players: Current firm (F) with large market share faces a potential entrant (E)  Timing: Potential entrant moves first  Moves: Potential entrant (enter-stay out) Current firm (accept passively-fight)  Information: full information  Payoffs: (see game tree)  Rules: Fixed (to simplify game for now)

27. 27 Scenario in Game Tree E out in F fight acc (0, 100) (-10,-20) (20,75) Payoffs = (E, F) expressed as profits (mil $)

28. 28 Looking Forward… Entrant makes the first move: Entrant makes the first move:  Must consider how F will respond If enter: If enter: Current Firm better off if accepts; so trim “fight” branch from tree Current Firm better off if accepts; so trim “fight” branch from tree F fight acc (-10,-20) (20,75)

29. 29 Now consider entrant’s move with tree trimmed Now consider entrant’s move with tree trimmed Solution = (In, Accept Passively ) Solution = (In, Accept Passively ) … And Reasoning Back E out in F (0, 100) (20,75) acc