Pay-for-Performance and Bureaucratic Representation in Public Organization

How to cite

Pedersen, Mogens Jin; Favero, Nathan; Park, Joohyung. 2021. "Pay-for-Performance and Bureaucratic Representation in Public Organization." AEA RCT Registry. AEARCTR-0007632. https://doi.org/10.1257/rct.7632-3.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Mogens Jin Pedersen* , Nathan Favero , Joohyung Park

* Primary investigator

Registry ID
AEARCTR-0007632
DOI
https://doi.org/10.1257/rct.7632-3.0
First registered
Last updated
Status
Completed
Trial dates
2021-05-15 – 2021-05-22
Countries
United States of America ()
Registry record
http://www.socialscienceregistry.org/trials/7632

Abstract

Many public organizations use Pay-for-Performance (PfP) schemes as a means for promoting the performance of the public services. Does the widespread use of PfP in the public sector have unintended consequences in terms of job attraction? Does PfP affect the prospects of representative bureaucracies? Using a within-subjects conjoint survey experimental design among a non-probability sample of US residents (n = 1,500), this study examines whether PfP (vs. fixed pay) make individuals of particular race, gender, and age—e.g., people of color, women, and older people—less attracted to a public sector job.

Intervention

The interventions consist of a within-subject conjoint experiment embedded in an electronic survey.
The design is a paired profiles conjoint in which job arrangement profiles for two jobs—A and B—are presented next to each other in a conjoint table.

The first column of the conjoint table lists a total of eight job attributes. The second and third columns list the job attribute values for jobs A and B, respectively. All job attribute values are assigned at random.

The exact text for the eight job attributes and their respective attribute values appear below. Unless noted otherwise, attribute values are assigned with equal probability for each attribute.

Attribute 1: “Total pay: Expected pay (including bonuses), compared to similar jobs elsewhere”
Attribute values (3):
• Slightly above average
• About average
• Slightly below average

Attribute 2: “Performance bonuses: How much of the expected pay is bonuses that depend on performance”
Attribute values (4; “fixed salary” assigned with 50% probability, the others with 16,7% probability each):
• A large part of your potential pay (20%)
• A moderate part of your potential pay (10%)
• A small part of your potential pay (5%)
• No performance bonuses; fixed salary

Attribute 3: “Job performance evaluation: How your performance is measured”
Attribute values (4):
• Attendance numbers for Project HOPE events
• Satisfaction surveys of Project HOPE event participants
• Changes in community crime, poverty, and blight
• A supervisor evaluation of your work

Attribute 4: “Current community involvement: Current participation levels for the program”
Attribute values (3):
• Frequent participation
• Moderate participation
• Rare participation

Attribute 5: “Community income: Average income in target community”
Attribute values (3):
• High income
• Average income
• Low income

Attribute 6: “Community demographics: Racial/ethnic makeup of neighborhoods”
Attribute values (4):
• Mostly white
• Mostly African American
• Mostly Hispanic
• Multiracial

Attribute 7: “Overtime work: How often you will work extra evening hours”
Attribute values (3):
• Frequently required
• Occasionally required
• Never required

Attribute 8: “Key job task: Most important job qualification”
Attribute values (4):
• Analysis identifying community needs
• Teamwork with peers and supervisors
• Coordination with community groups and organizations
• Direct interaction with community residents

Experimental design

The conjoint survey experiment is carried out among a non-probability sample of US residents. Participants are recruited via Prolific, with enrollment limited to individuals with current residence in the US (using Prolific’s prescreening feature). In order to ensure a racially diverse sample of respondents, we will recruit 750 White respondents and 750 non-White respondents. This quota sampling approach is accomplished using Prolific’s prescreening feature. Specifically, two identical studies are created in Prolific, except that one study limits enrollment to only White participants (using Prolific’s demographic category “Ethnicity (Simplified)”) and the other study limits enrollment to non-White participants.

All survey respondents are presented with a job choice task. First, respondents are exposed to an introductory text describing the task and providing basic information about a city government job (as Community Active Worker on a community empowerment program called Project Hope). Next, respondents are exposed to a paired profiles conjoint in which two specific job arrangement profiles—A and B—for the Community Active Worker job are presented next to each other in a conjoint table.

The first column of the conjoint table lists eight job attributes. The second and third column list the job attribute values (for those eight job attributes) for jobs A and B, respectively. All job attribute values are assigned at random.

The exact job attributes and job attribute values appear under ‘INTERVENTIONS.’

As our outcome measure, all respondents are asked to indicate their choice between the two job arrangement profiles (“Which of the two jobs would you personally prefer?”). Response options are “Job A” and “Job B.”

Respondents are presented with similar paired profiles conjoints (i.e., involving the same job attributes and random assignment of job attribute values for the Community Active Worker position) two more times. Thus, each respondent will see a total of three pairs of job profiles.

Based on existing theory and research, we derive and test the following hypotheses:

H1: PfP (vs. fixed pay) affect individuals’ job attraction.

Moreover, we theorize that socio-demographic characteristics (race, gender, age) moderate the effects of PfP (vs. fixed pay) on individuals’ job attraction. In particular, we derive and test the following hypotheses:

H2a: PfP (vs. fixed pay) has negative impact on job attraction for racial minority individuals (vs. racial majority individuals)
H2b: PfP (vs. fixed pay) has negative impact on job attraction for women (vs. men)
H2c: PfP has negative impact on job attraction for older individuals (vs. younger individuals)

Design details

Randomization method
Randomization is carried out by simple randomization by computer (randomization based on a single sequence of random assignments).
Randomization unit
The individual survey respondent
Clusters
1,500 survey respondents (750 White respondents and 750 non-White respondents)
Observations
1,500 survey respondents (750 White respondents and 750 non-White respondents)
Arms
Providing exact sample sizes for all potential constellations of job attribute values is less meaningful given our design and research focus. However, our main treatment of interest—PfP—have the following sample size by treatment arms: • “A large part of your potential pay (20%)” = 1,500 • “A moderate part of your potential pay (10%)” = 1,500 • “A small part of your potential pay (5%)” = 1,500 • “No performance bonuses; fixed salary” 4,500 For our main analyses, however, we operationalize “PfP” using a binary variable capturing “Fixed salary” (n = 4,500) (0) and “PfP” (n = 4,500; all three PfP attribute values) (1).
Primary outcomes
Job attraction (self-reported; based on responses to a survey item)

Materials

Cited by

No citation data available for this registration yet.

Keywords