MASTER Administration des Entreprises Apprentissage Stéphane Saussier Saussier@univ-paris1.fr «Environnement Economique de l Entreprise» Séance 5 - Incentives in Organization - SSA 2015-1
Incentive Theory and Agency Theory The normative agency theory (incentive theory) is directly inspired from the positive agency theory Main idea - Manager s discretionary margins are coming from asymmetries of information Problems generated by such asymmetries are pervasive all over society, not only at the shareholder - manager relationships level One solution: contracts can be framed in order to give proper (monetary) incentives to economic actors SSA 2015-2
Altruism vs. Selfish behavior (1) Experimental economics Two anonymous players Player 1 is given 20 $ Player 1 offers an amount of money to player 2 between 0 & 20 $ Player 2 then accepts or refuses the proposition If player 2 refuses, both are leaving the game with 0 gain Best strategy of player 1 Give 1 penny to player 2 ; keep 19,99$! Player 2 should accept. This is not what we observe! Player 2 usually do not accept below 3$! On average Player 1 gives 6$ to player 2 Altruism? SSA 2015-3
Altruism vs. Selfish behavior (2) Dictator game Two anonymous players Player 1 is given 20 $ Player 1 offers an amount of money to player 2 between 0 & 20 $ Player 2 has no other choice than accepting Best strategy of player 1 Give 1 penny to player 2 ; keep 19,99$! This is not what we observe! On average Player 1 gives 4$ to player 2 Altruism!? True whatever the kind of people playing and whatever the countries! SSA 2015-4
Altruism vs. Selfish behavior (3) Dictator game (John List experiment) Steal a dollar if you want! Two anonymous players Player 1 is given 20 $ Player 1 offers an amount of money to player 2 between 0 & 20 $ Player 2 has no other choice than accepting Player one can also take 1$ from player 2 If people are altruist, this should not change results Results 30% of players gave something to player 2 45 % gave nothing 25 % steal a dollar John List U. Of Chicago SSA 2015-5
Altruism vs. Selfish behavior (4) Dictator game (John List experiment) Two anonymous players Player 1 is given 20 $ Player 1 offers an amount of money to player 2 between 0 & 20 $ Player 2 has no other choice than accepting Player 1 is told that player 2 had been given the same amount of money as player 1 (20$). Player 1 can decide to steal the whole money of player 2 (20$) If people are altruist, this should not change results Results 10% of players gave something to player 2 60% took money from player 2 40% took all player 2 s money! John List U. Of Chicago SSA 2015-6
Altruism vs. Selfish behavior (5) Dictator game (John List experiment) Two anonymous players Idem as before but players had to work first to be given money (20$) If people are altruist, this should not change results Results 65 of players 1 gave nothing to player 2 Only 28% took from players 2 Real world is more complex than experimental games previous results has nothing to do with altruism See : "What Do Laboratory Experiments Measuring Social Preferences Reveal about the Real World?"Journal of Economic Perspectives, 2007, 21(2), pp.153-174. (List, John A. & S. Levitt) John List U. Of Chicago SSA 2015-7
Incentive Theory: questioning the impersonal dimension of exchange General Equilibrium Model Arrow - Debreu (51) The introduction of Asymmetric information. The market is not anymore a central place where quantities of a specified good are exchanged. It becomes a local face to face exchange place where economic actors are able to identify themselves and to contract together. Stiglitz 74, Mirrles 71, Spence 74 & J-J Laffont. SSA 2015-8
New questions generated by asymmetric information Asymmetric information gives rise to pervasive problems. Adverse Selection Definition Illustrations Insurance; Labor contracts; Manager-Shareholders Impact The case of insurance: Average cost is a function of the price! Akerlof 1970 and the «Market for Lemons»: markets might be closed SSA 2015-9
New questions generated by asymmetric information Moral Hazard Definition Hidden action Hidden information Illustrations Insurance; Labor contracts - Team; Manager- Shareholders Impact Disentangling Adverse Selection and Moral Hazard effects might be difficult The case of car insurance SSA 2015-10
Incentive Theory s Methodology Exchange between two actors One Principal One Agent One party is more informed than the other one One party gets the whole negotiation power Third parties are supposed to be efficient and costless SSA 2015-11
Organization is viewed as a nexus of contracts The incentive theory postulates that the contract helps To make actors reveal their information For example, bad actors will declare their type To make actors change their behavior Good ones will perform To find out an optimal contract whatever the organizational structure concerned Results applied to private and public firms and interfirms contracts What will change from one organizational structure to another is the capabilities to implement optimal solutions proposed by the theory Condition under which such incentive scheme might be implemented are meet or not See Laffont 2000 SSA 2015-12
Adverse Selection: The incentive theory solution in practice The menu of contracts principle Insurance contracts: partial vs. complete insurance Franchise contracts: manager vs. franchisee Wage: Piece-rate vs. constant wage SSA 2015-13
Assumptions underlying the result Full rationality and complexity The Principal is able to perfectly reconstruct the strategies of privately informed agents Perfect communication Full control of communication channels between agents and mediators Risk of collusion Full commitment Costless Enforcement Those conditions might be more or less respected depending of the organizational structure considered SSA 2015-14
Incentive Contract and Moral Hazard Implicit Assumption There is no way for the Principal to control perfectly agent s behaviour Or this is too costly. SSA 2015-15
Incentive Contracts and Moral hazard Production Technology Production process is summarized by just three variables: (1) the Agent s total contribution to firm value (or, for now, the Agent s output ), denoted by y; (2) the action the Agent takes to produce output, denoted by a; and (3) events in the production process that are beyond the Agent s control (i.e., noise ), denoted by ε. SSA 2015-16
Incentive Contracts and Moral hazard Contracts Assumption 1: y = a + ε with ε N (0,1) Assumption 2: Linear payment scheme, with a fixed part and a variable part W(y) = s + by Those assumptions change results quantitatively, not qualitatively SSA 2015-17
Incentive Contracts and Moral hazard Revenues Up = y - w Ua = w - C(a) With C(a) the cost of effort for the agent C > 0, C > 0 SSA 2015-18
Incentive Contracts and Moral hazard Sequentiality The Principal and the Agent sign a compensation contract (w = s + by). 2. The Agent chooses an action (a), but the Principal cannot observe this choice. 3. Events beyond the Agent s control (ε) occur. 4. Together, the action and the noise determine the Agent s output (y). 5. Output is observed by the Principal and the Agent (and by a Court, if necessary). 6. The Agent receives the compensation specified by the contract. SSA 2015-19
Incentive Contracts and Moral hazard The Agent s choice Max a s + ba - C(a) C (a) = b C(a) E(w) = s + ba Line with slope b s a*(b) a SSA 2015-20
Incentive Contracts and Moral hazard Qualitative results The fixed part does not influence incentives received by the Agent! It is not always in the advantage of the Principal to implement such a contract It is not a free lunch SSA 2015-21
Conclusions What Limits? The willingness of employees to bear the risk might vary a lot Such payment schemes exacerbate differences between workers Stress on employees Natural motivations might be destructed by monetary incentives Individual incentives might be replaced by group incentives (Scanlon plan) Must be complemented by a dynamic reward (careers, ) The Principal must commit to accept ex post inefficiencies! Restricted to the cases where contribution is observable? SSA 2015-22
Facts : Non verifiable contribution might be rewarded Very often, the contribution of the Agent is not observable by the Principal. Or it might not be contractible (i.e. non verifiable) y reflects all the Principals cares about, not simply «output» Contracts cannot always be based on y Contracts might be based on performance indicators p W = s + bp SSA 2015-23
Multi-tasks Models «The essence of the incentive problem is this divergence between the Agent s incentives to increase p and the firm s desire for increases in y» (Gibbons 1995). Consider a simple extension of the basic model, with two kinds of actions, a1 and a2. y = a1 + a2 + ε and p = a1 + Φ y = a1 * a2 + ε and p = a1 + Φ Examples: Professors in Universities; Judges ; Police; This might explain that we often observe very simple contracts in the real world compared to what incentive theory predicts Do not forget organization consequences! (Laffont 2000) SSA 2015-24
Implications (Rapport CAE 2000) Principe 3 : Si le gouvernement veut utiliser une information décentralisée, il doit se soucier de fournir des incitations adéquates pour obtenir, des agents économiques, une transmission véridique de ces informations. P. 127 : «Ce principe peut sembler trivial, mais il est clair qu il n est pas mis en oeuvre dans l administration française et les entreprises publiques. Les incitations des cadres administratifs français sont des incitations d avancement de carrière et des primes qui ont été souvent vidées de leur pouvoir incitatif par le comportement syndical et la «démission» de ceux chargés de les mettre en oeuvre (car eux mêmes n ont pas d incitations appropriées) SSA 2015-25
Implications (2) Principe 4. «Si le gouvernement n observe que des signaux bruités de l effort de ses agents, il lui faut mettre en place des rémunérations en fonction des performances pour motiver ses agents à l effort». «Ce n est pas une tâche facile» «Ce n est pas un free lunch» «Il faut pouvoir mettre en place un tel système en faisant mesurer les performances par des auditeurs fiables, euxmêmes dûment motivés, ce qui est coûteux». SSA 2015-26
Implications (3) Principe 9. «Autant que possible, le gouvernement doit organiser la concurrence dans ses propres services lorsque ce n est pas un obstacle aux nécessaires coordinations» Benchmarking SSA 2015-27
Implications (4) Principe 18. «La privatisation d activités peut se justifier par les obstacles qu elle met à la poursuite d agenda privés par des gouvernements non bienveillants». SSA 2015-28
Conclusions Implications Role of the explicit contracts? Role of the commitment? Role of the reputation of the firm? Application : Laffont 2000 Public firms cannot implement such incentive schemes easily compared to private ones SSA 2015-29
FACTS!? "Data! Data! Data!... I can't make bricks without clay.» Sherlock Holmes SSA 2015-30
DATA? SSA 2015-31
Data SSA 2015-32
Applications and Implications: One Case Study Application Lazear E. 2000, Performance Pay and Productivity, American Economic Review, 90, 1346-1361 SSA 2015-33
Applications and Implications Lazear E. 2000, Performance Pay and Productivity, American Economic Review, 90, 1346-1361 Safelite Glass Corporation Introduction of a new management strategy W = S 9 => W = s + by 19 months 3,000 different workers over a 19-month period SSA 2015-34
Lazear 2000 Consequences Average productivity : a 44-percent increase in output per worker! 50% due to adverse selection effect 50% due to moral hazard effect Variance around individual productivity changed Reduced or increased? Average wage increased of 7% SSA 2015-35
Lazear 2000 Is it a success? Profit and wage increase? Costs of implementation? Costs of control? Impact on quality? Other more optimal solutions? SSA 2015-36
Pour approfondir : http://www.youtube.com/watch?v=3ifgt8eiq3m Ratchet effect Workers involvement SSA 2015-37
Pour approfondir : les entretiens individuels d évaluation Pour quoi faire? Pour qui? Quel impact sur le stress? Voir le rapport du CAS 2011 http://www.webssa.net/node/305 SSA 2015-38
SSA 2015-39
Et les motivations? Dan Pink sur la surprenante science de la motivation http://www.ted.com/talks/dan_pink_on_motivation.html SSA 2015-40