Incentives and Behavior Prof. Dr. Heiner Schumacher KU Leuven 15. - - PowerPoint PPT Presentation

Incentives and Behavior

• Prof. Dr. Heiner Schumacher

KU Leuven

• 15. Identity and Motivation
• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

1 / 27

Introduction

Monetary incentive schemes have a number of problems. First, compensation schemes can only be based on observables that are easy to measure. Second, monetary incentives create opportunities to game the system (recall the lecture on multitask incentives). Third, in tournaments (i.e., competition for promotion) there are incentives to sabotage the colleagues’ work.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

2 / 27

Introduction

Firms therefore may want to create non-monetary incentives that are based on social preferences. In this lecture, we will discuss two approaches from economics (identity and mission motivation) how to create non-monetary incentives in organizations. Both approaches are inspired by psychology and sociology. We will also consider the empirical & experimental evidence on identity and mission motivation.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

3 / 27

Introduction

Overview Identity Empirical Evidence (1): Ingroup and Outgroup E¤ects Empirical Evidence (2): Priming Social Identity Mission Motivation Empirical Evidence: How Much Does the Mission Matter?

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

4 / 27

Identity

In traditional models from organizational economics (as the principal-agent model discussed in class), individual preferences are …xed and utility depends on pecuniary variables only. In contrast, Akerlof and Kranton (2005) argue that individuals have an identity that in‡uences behavior and that can be changed.1 The ability of organizations to place workers into jobs with which they identify and the creation of such identities are central to what makes

• rganizations work.

1Akerlof, George, and Rachel E. Kranton (2005): “Identity and the Economics of

Organizations,” Journal of Economic Perspectives 19(1), 9 - 32.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

5 / 27

Identity

“On plebes’ …rst day at West Point, called R-Day, they strip down to their underwear. Their hair is cut o¤. They are put in

• uniform. They then must address an older cadet, with the proper

salute and with the statement: “Sir, New Cadet Doe reports to the cadet in the Red Sash for the …rst time as ordered.” Plebes must stand and salute and repeat, and stand and salute and repeat, until they get it exactly right, all the while being reprimanded for every tiny mistake.”

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

6 / 27

Identity

Social categories describe types of people (gender, race, occupation, etc.). People often think of themselves in terms of social categories. Identity describes a person’s social category. It captures how people feel about themselves, and how those feelings depend on actions. Utility depends on norms, i.e., ideas of how people of a given social category should behave. In a model of utility, identity describes gains and losses from behavior that conforms or departs from the norms of a particular social category. A person who identi…es with being a member of a particular social category then loses utility insofar as her behavior di¤ers from that of the ideal. An employee who identi…es himself with an organization is called an insider, while an employee who does not identify himself with the

• rganization is called an outsider.
• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

7 / 27

Identity

Akerlof and Kranton (2005) create a simple preference model of identity. Think of a worker who chooses how much e¤ort e she wants to exert (assume that she may exert either high or low e¤ort). Let there be two social categories c, N and O. Her utility function is given by U(y, e, c) = ln y e + Ic tc je(c) ej , where y is her income, Ic her identity utility from being in category c, and tc je(c) ej the disutility from diverging from the ideal e¤ort level (denoted by e) for category c.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

8 / 27

Identity

If the worker has an identity as an insider, the presence of identity utility will reduce the wage di¤erential needed to induce the worker to exert high e¤ort. If the worker identi…es as an outsider, the presence of identity utility will increase the necessary monetary incentives. This creates the following trade-o¤: An organization may choose relatively high monetary incentives and no investments into changing workers’ identity, or relatively low monetary incentives and large e¤orts to change motivation by identity. What can a …rm do in order to increase its workers’ identi…cation with the …rm?

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

9 / 27

Identity

The model also o¤ers implications for optimal organization. Assume that identity is created at the workgroup level. A supervisor may report the workers’ actions to management. However, workers do not like being monitored (recall the hidden costs

• f control), so they will identify less with the workgroup.

They will then require higher pay to perform.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

10 / 27

Identity

With strict supervision, the worker distinguishes herself from her

• supervisor. In reaction, she becomes an outsider.

With loose supervision, the worker views the supervisor as part of the

• workgroup. The worker then identi…es with the workgroup and

becomes an insider. Again, we obtain a trade-o¤ between monetary and non-monetary incentives. There exist several methods to manipulate worker-supervisor relations and workgroup identity: job rotation, workgroup composition, physical arrangements, …rm-sponsored activities, group lunchrooms, etc.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

11 / 27

Empirical Evidence (1): Ingroup and Outgroup E¤ects

In a lab experiment, Chen and Li (2009) show how group identity changes economic preferences.2 They use the “minimal group paradigm” from social psychology, i.e., a method for investigating the minimal conditions required for discrimination to occur between groups (that is, the group distinction per se has no meaning). On the following slides, we consider the most important …ndings of their study.

2Chen, Yan, and Sherry Xin Li (2009): “Group Identity and Social Preferences,”

American Economic Review 99(1), 431 - 457.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

12 / 27

Empirical Evidence (1): Ingroup and Outgroup E¤ects

In the …rst stage, subjects review …ve pairs of paintings (in each pair,

• ne is by Klee and one by Kandinsky).

Based on reported preferences, subjects are divided into two groups, the Klee group and the Kandinsky group. In the second stage, they play a number of economic games with members of their own and the other group. Roughly, these games are other-other allocation games (subjects allocate tokens between two other anonymous participants; no tokens to herself), dictator games with varying endowments and trust-games with varying payo¤ consequences.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

13 / 27

Empirical Evidence (1): Ingroup and Outgroup E¤ects

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

14 / 27

Empirical Evidence (1): Ingroup and Outgroup E¤ects

Other-other allocation games. The average number of tokens allocated to an ingroup member is signi…cantly more than that allocated to an outgroup member. Hence, there is ingroup favoritism in minimal groups paradigm experiments. Dictator games. Subjects exhibit a higher degree of altruism toward an ingroup member than to an outgroup member. Trust games. Subjects are more likely to reward and ingroup member than an outgroup member for good behavior. They are more forgiving toward misbehavior from an ingroup member than toward an

• utgroup member.
• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

15 / 27

Empirical Evidence (2): Priming Social Identity

Clean evidence for the incentive e¤ects of social categories comes from a priming experiment by Benjamin et al. (2010).3 Priming refers to the temporary activation of an individual’s mental representations and the e¤ect of this activation on behavior in an unrelated subsequent task. As priming instrument researchers have used the writing of essays, responding to questionnaires, the “Scrambled Sentence Test” or the subliminal presentation of priming words. Benjamin et al. (2010) use a “background questionnaire” to prime social identity (Asian-American versus white American). This includes questions about languages spoken by the subject’s family and how many generations the subject’s family has lived in the United States (in the control treatment, the questionnaire was about meal plans and cable television subscription).

3Benjamin, Daniel, James Choi, and Joshua Strickland (2010): “Social Identity and

Preferences,” American Economic Review 100(4), 1913 - 1928.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

16 / 27

Empirical Evidence (2): Priming Social Identity

It is known that relative to white Americans, Asian-Americans accumulate more human capital and are more likely to participate in tax-deferred savings accounts. Hence, the hypothesis is that priming social identity results in less impatient choices by Asian-Americans (relative to the white control group). Time preferences were elicited by asking participants to make 46 binary choices between a certain amount of money received earlier (3 USD or 7 USD) and a larger amount received later (either between 3.05 USD and 7 USD, or between 7.10 USD and 15 USD).

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

17 / 27

Empirical Evidence (2): Priming Social Identity

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

18 / 27

Empirical Evidence (2): Priming Social Identity

Social identity a¤ects fundamental economic preferences. Making Asian-American subjects’ ethnicity salient causes them to exhibit more patient preferences. Moreover, Benjamin et al. (2010) …nd suggestive evidence that native blacks become more risk averse and whites more patient when their racial identity is salient. These results show that varying identity has an impact on preferences as suggested by Akerlof and Kranton (2005).

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

19 / 27

Mission Motivation

Many workers are “motivated agents”, i.e., they pursue a certain goal (other than their monetary payo¤) because they perceive intrinsic bene…ts from doing so. Examples: Doctors are committed to saving lives; researches to advancing knowledge (teaching students?); judges to promoting justice; soldiers to defending their country in battle. Some …rms (like universities, charity organizations, armies) can adopt a mission and then may attract “motivated agents”. What is the di¤erence between mission motivation and identity?

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

20 / 27

Mission Motivation

“Shopkeepers do not swear oaths about shop-keeping. Neither do investors swear oaths about investment. But in most places in the world, doctors swear oaths about doctoring.” (Sulmasy 1999)

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

21 / 27

Mission Motivation

Besley and Ghatak (2005) build a principal-agent model where …rms may or may not have a mission, and some worker are “motivated agents”.4 Roughly, they show that there will be assortative matching: On the

• ne hand, there will be a for-pro…t sector which attracts workers

without mission motivation and o¤ers contracts with substantial monetary incentives; on the other hand, there will be a non-pro…t sector which attracts workers with mission motivation and o¤ers contracts with little monetary incentives. This matching economizes on the need for high-powered incentives.

4Besley, Timothy, and Maitreesh Ghatak (2005): “Competition and Incentives with

Motivated Agents,” American Economic Review 95(3), 616 - 636.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

22 / 27

Empirical Evidence: How Much Does the Mission Matter?

It is commonplace in business economics to claim that an

• rganization’s mission matters for employee performance.

There is some survey evidence. Gallup polled workers in the US and found that employees who felt their job was important due to their employer’s mission were more productive. However, there is no clean evidence for this claim in the economic literature. Varying an organization’s mission is obviously di¢cult. Empirical evidence on mission motivation and its interaction with other incentive devices is therefore scarce. One of the few papers that try to tackle the problem is Carpenter and Gong (2013).5

5Carpenter, Je¤rey, and Erick Gong (2016): “Motivating Agents: How much does

the mission matter?,” Journal of Labor Economics 34(1), 211 - 236.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

23 / 27

Empirical Evidence: How Much Does the Mission Matter?

Carpenter and Gong (2013) …rst measure the political preferences of their (potential) student subjects (ahead of the 2012 US presidential elections between Romney and Obama). In an economic experiment two weeks later, randomly selected subjects address and stu¤ campaign letters for one of the two major party candidates. A “match” occurs if a subject with preferences in favor of the Democratic (Republican) party works for the Obama (Romney) campaign. A “mismatch” occurs if a subject with preferences in favor of the Democratic (Republican) party works for the Romney (Obama) campaign. There are two types of treatment: A treatment with …xed wage and additional monetary incentives (piece-rates), and a treatment with …xed wage only.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

24 / 27

Empirical Evidence: How Much Does the Mission Matter?

What is the null hypothesis w.r.t. mission match/incentives? What outcome would you expect given what you know about incentives so far?

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

25 / 27

Empirical Evidence: How Much Does the Mission Matter?

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

26 / 27

Empirical Evidence: How Much Does the Mission Matter?

Matched workers produce on average 15,62 letters, while mismatched workers produce on average only 9,07 letters, a di¤erence of 72%. Between 25% and 45% produce nothing when mismatched; the corresponding number for matched workers is close to zero. Mismatched workers without incentives produce 6,33 letters, while those with incentives produce 11,75 letters, a di¤erence of 86%. So monetary incentives can overcome a lack of motivation (that is due to a mismatch). The corresponding number for matched workers is only 13%. We therefore can conclude that mission can matter substantially.

• Prof. Dr. Heiner Schumacher (KU Leuven)

Incentives and Behavior

• 15. Identity and Motivation

27 / 27