x

We are happy to announce that all trial registrations will now be issued DOIs (digital object identifiers). For more information, see here.
Positional consumption and status seeking behaviors: Evidence from a Field Experiment in Uruguay
Last registered on November 02, 2018

Pre-Trial

Trial Information
General Information
Title
Positional consumption and status seeking behaviors: Evidence from a Field Experiment in Uruguay
RCT ID
AEARCTR-0003392
Initial registration date
November 01, 2018
Last updated
November 02, 2018 5:51 PM EDT
Location(s)
Primary Investigator
Affiliation
UDELAR
Other Primary Investigator(s)
PI Affiliation
UDELAR
PI Affiliation
CAF Development Bank of Latin America
PI Affiliation
UDELAR
Additional Trial Information
Status
In development
Start date
2017-03-01
End date
2019-02-28
Secondary IDs
Abstract
In order to explain the reasons why people's consumption decisions might be affected by the decisions of other people, in this paper we focus on the role of status-seeking behavior. There is growing evidence that confirms the hypothesis that people care about social status. However, an overview of the different economic models available and recent empirical findings on this topic suggest an overlapping of different strands of the literature. Frank and Heffetz (2011) highlight that evidence on this topic has been accumulating in several different strands of the literature, from empirical studies that explore the role of reference groups on the level of consumption of certain goods, from studies that explore how relative income affects individual well-being (happiness literature) or the evidence from experimental studies on individuals' social preferences or preferences for status. Recent developments in the field of behavioral economics and on the specific role of relative positions, emulative behavior and social interactions have reinvigorated this research line (Frank and Heffetz, 2008).
The idea that peer's decisions play a crucial role in explaining individual consumption decision has a long story in social sciences and particularly in economic models. Positional or relative income concern, transmission of preferences or tastes, imitation, informational externalities, identity or status seeking behavior are some of the channel which could explain that relationship. However, how much these channels affect the magnitude of consumption decision on positional good has received limited attention in empirical economic literature. The main contribution of this paper is to fill this gap by providing new evidence on the role of social status behaviors on consumption on visible goods, particularly analyzing that behavior among groups of youngsters’ friends.
Our original experimental design seeks to provide evidence to answer whether individuals’ consumption of positional goods is affected by the choices of others (specifically, friends), due to status seeking behavior. In order to advance in this sense, we focus on youngster's consumption decision and the role of their peer's choices. The main challenges of our design are how to reproduce consumption decisions in a real situation and how to isolate status seeking consumption from other peer-group effects, such as: information, network economies and identity. The main hypothesis to be tested is whether the consumption of a positional good by individual i belonging to peer group j increases the willingness to pay for that good for the remaining group members j-i.
Main experiment design: A sample of individuals aged 20-22 (followed by the longitudinal study, ELBU) participate in the field experiment. ELBU respondents will be randomly pre-allocated to a “Reference treatment” (RT) or “Reference control group” (RC) and (independently of that) they will be assigned to be eligible to receive 30 tickets for a lottery of a positional (P) or a non-positional good (NP). The price of each good is 350 U$S (Approximately a minimum wage). Each ELBU youngster provides the names and contact information of four close friends. Their friends are visited in a second step, and they participate in a consumption decision among the same two goods: P and NP. ELBU's friends (aged between 19-28) have to assign 10 lottery tickets to a P and NP good. When they make their choice, 50% have their friends as a reference point (information treatment) and 50% chooses without any additional information (control).
A complementary strategy is based on an adaption of Alpizar et al (2005) experimental survey. That strategy allows us to measure the degree of positionality of selected good. Also, we explore if preferences for relative income depends knowledge about what determines the position in the income distribution: effort or heritage.
There are several arguments that justify the relevance on provides evidence on this topic. At the microeconomic level, status-seeking consumption can compete with other valuable allocations that contribute to increase wellbeing in the long run, such as education, health, leisure or savings (Charles et al, 2009; Schor,1998; Bowles and Park, 2005). Furthermore, the race for status could generate wasteful consumption, turning into aggregate social welfare losses, in terms of private expenditure and potential complementarity with public investment (Frank, 2018). Thus, the potential negative effects of status-seeking behavior have a wide set of policy implications, ranging from progressive taxation to luxury goods, to educational and cultural interventions (Bowles and Park, 2005; Cortina, 2011; Crocker and Linden, 1998; Frank, 2016)
External Link(s)
Registration Citation
Citation
Guillermo, Alves et al. 2018. "Positional consumption and status seeking behaviors: Evidence from a Field Experiment in Uruguay." AEA RCT Registry. November 02. https://doi.org/10.1257/rct.3392-1.0.
Former Citation
Guillermo, Alves et al. 2018. "Positional consumption and status seeking behaviors: Evidence from a Field Experiment in Uruguay." AEA RCT Registry. November 02. https://www.socialscienceregistry.org/trials/3392/history/36703.
Experimental Details
Interventions
Intervention(s)
A set of individuals aged 20-22 are randomly assigned into three groups (Treated-Positional, Treated-Non-Positional, and Controls) and each group receives lottery tickets for a specific consumption good. This set of individuals is drawn from a longitudinal study of first graders attending public schools in 2004. One group receives tickets for a positional good (P), a second group for a non-positional good (NP), and the third group are only told about the monetary value of the good and not the type of good. All goods have roughly the same value (US$ 350). We picked a piece of jewelry and a mattress as the P and NP goods by performing a set of focus groups and a pilot with a different set of young adults of the same age. All individuals are further asked to provide the names and contact information of four close friends. Their friends are then visited and asked to assign a 10 lottery tickets between P and NP. The treatment status of these “friends” is given by the treatment status of the friend who referred them.
Intervention Start Date
2018-01-01
Intervention End Date
2018-10-31
Primary Outcomes
Primary Outcomes (end points)
A measure of the willingness to pay for positional versus non-positional goods.
Primary Outcomes (explanation)
The allocation of the 10 lottery tickets between the two goods, bedspring and mattress (NP) and jewelry (P), provides a measure of individuals’ willingness to pay for positional and non-positional goods.
The experiment tests the following hypotheses: (H1) The consumption of a good by individual i increases the willingness to pay for that good of her friend j. H2) This effect is significantly higher for positional than non-positional goods.
Secondary Outcomes
Secondary Outcomes (end points)
1 - Degree of positionality of selected goods.

2 – Preferences for status


3 – Complementary information on subjects’ preferences

We further include some questions to check the respondents’ understanding of the experimental procedure as well as some questions measuring subjects’ degree of risk aversion.
Finally, additional questions collect participants’ attitudes towards income comparisons, preferences for taxation, cash transfer programs, and status. This information, together with participants’ socioeconomic background, will be used to investigate which covariates are associated with preferences for status.
Secondary Outcomes (explanation)
1 - Preferences for status

We use an experimental survey design adapted from Amiel and Cowell (1992) and Alpizar et al (2005). Participants are asked to choose, among several pairs of societies, in which one they would prefer their grandchild to live. In order to test for gender biases, a random sample of the respondents are asked about their granddaughter’s situation. The different pairs of societies are characterized by different levels of average income and grandchild’s incomes. Choosing between these different societies implies a trade-off between individual's absolute and relative income. We repeat the same procedure using, instead of income, the grandchild’s total consumption expenditure as well as her consumption expenditure in different goods (mattress, jewelry) to assess differential levels of positionality of these goods. Another variant of this relative-income experiment consists in mentioning what determines the grandchild’s position in the income distribution: effort or inheritance. This information allows us to test the sensibility of the positionality parameter to individual beliefs about the role of meritocracy.
2 - Degree of positionality of selected goods.

We adapt the experimental questionnaire used in Alpizar et al (2005) to measure the degree of positionality of income and some visible goods. We include a robustness exercise consisting in randomly changing some questions’ order.
Experimental Design
Experimental Design
(i) Main consumption experiment (ii) an attention check question (iii) relative income experiment (iii) relative consumption experiment (iv) questions on the degree of positionality of different goods (v) background questions on socioeconomic status (vi) individual attitudes, demographics, political beliefs, and preferences for redistribution.
Experimental Design Details
Main experiment design: A set of individuals aged 20-22 are randomly assigned into three groups (Treated-Positional, Treated-Non-Positional, and Controls) and each group receives different types of lottery tickets for a specific of consumption goods. The first group receives tickets for a positional good (P), the second group for a non-positional good (NP), and the third group are only told about the monetary value of the good. All goods have roughly the same value, around US$ 350. These individuals are further asked to provide the names and contact information of four close friends. Their friends are then visited and asked to assign 10 lottery tickets between P and NP. In order to select the goods P and NP we organized three focus groups with youngsters from different socio-economic strata, we pre-tested the experiment and survey questionnaries with undergraduate students and assessed from the findings by Heffetz (2011 and 2012) on visibble goods for the US. Based on these elements, we selected jewelry as the positional good and bedspring and mattress as the non-positional good.
Randomization Method
The randomization is done in office by a computer. Each individual has the same probability of being assigned to each group.
Randomization Unit
The unit of randomization is the individual from the ELBU Panel Survey.
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
1300 binomials. A binomial is defined by a youngster and his or her friend.
Sample size: planned number of observations
750 binomials. A binomial is defined by an ELBU youngster and his / her friend. Namely, the planned number of individuals visited is 1500.
Sample size (or number of clusters) by treatment arms
Main experiment: status seeking consumption
350 individuals are assigned to the treatment group. 50% of them receive tickets for a lottery for winning a positional good (P) and 50% for winning a non-positional good (NP).

350 individuals are assigned to the control group. Youngsters of this group only know the value of the prize. They only get to know the good after their friends have been visited.


Relative income and relative consumption experiment.

Youngsters are assigned to two groups: A) respondents choose between hypothetical societies, where their choices determine their grandchild's income and relative income (350 youngsters). B) respondents choose between hypothetical societies, where their choices determine their granddaughter's income and relative income (350 youngsters).

Friends are assigned to two groups: A) respondents choose between hypothetical societies, where their choices determine their grandchild's (granddaughter’s) income and relative income. In this case, the instruction includes an information treatment describing that the grandchild's (granddaughter's) income’s origin is their effort (350 youngsters). B) respondents choose between hypothetical societies, where their choices determine their grandchild's (granddaughter's) income and relative income. In this case, the instruction includes an information treatment which establishes that their grandchild's (granddaughter's) income comes from inheritance (350 youngsters).

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
IRB Approval Date
IRB Approval Number
Post-Trial
Post Trial Information
Study Withdrawal
Intervention
Is the intervention completed?
No
Is data collection complete?
Data Publication
Data Publication
Is public data available?
No
Program Files
Program Files
Reports and Papers
Preliminary Reports
Relevant Papers