Reform Windfall as Redistribution: A Survey Experiment on Redistributive Preferences in Contemporary China

Last registered on November 25, 2024

View Trial History

Pre-Trial

Trial Information

General Information

Title

Not My Money to Touch: Experimental Evidence on Redistributive Preferences under Market Transition in China

RCT ID

AEARCTR-0013928

Initial registration date

July 09, 2024

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

July 16, 2024, 2:31 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated

November 25, 2024, 1:05 PM EST

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

China

Region

Primary Investigator

Name

Zhexun Mo

Affiliation

Stone Center on Socio-Economic Inequality, GC-CUNY; World Inequality Lab, Paris School of Economics

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Margot Belguise

PI Affiliation

Warwick University

Contact Investigator

PI Name

Yuqian Chen

PI Affiliation

Harvard University

Contact Investigator

PI Name

Yuchen Huang

PI Affiliation

Paris School of Economics

Contact Investigator

Additional Trial Information

Status

Completed

Start date

2024-07-11

End date

2024-09-01

Keywords

Behavior, Governance, Welfare

Additional Keywords

Inequality;redistribution

JEL code(s)

H23; H24; H53; I38; D31; D63

Secondary IDs

Prior work

This trial is based on or builds upon one or more prior RCTs.

Abstract

China’s economic rise is arguably one of the most significant economic events of the past four decades. However, its opening up and liberalization have also created new inequalities. As China’s growth slows and inequality becomes more pronounced, redistributive policies have entered public discussion. A crucial question arises: how does the average Chinese person view those who have become wealthy over the past 40 years? Do they consider their wealth fair, or do they call for redistributive measures? This project builds on a pilot survey experiment we conducted in 2021, in which we found that priming–via vignettes–getting rich via relatively less meritocratic, yet representative ways under market transition in post-reform China - such as inheritance, housing arbitrage, or house demolition compensation - significantly reduces redistributive support, and in particular support for policies taking from the rich, while one may have a priori expected the opposite to happen. To verify the results of the pilot, we run a RCT on a representative sample of 2,000 Chinese respondents, replicating the priming treatment on a randomly-drawn subsample of 1,000 respondents. To better understand the results, we ask treated and control respondents what factors they think are the main sources of wealth in today’s society, as well as the sources of wealth of the individuals in the examples.

External Link(s)

Registration Citation

Citation

Mo, Zhexun et al. 2024. "Not My Money to Touch: Experimental Evidence on Redistributive Preferences under Market Transition in China." AEA RCT Registry. November 25. https://doi.org/10.1257/rct.13928-2.0

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

Treated respondents will be presented with three vignettes. These vignettes capture stories of wealth acquisition through luck (that is to say not by personal effort) which were frequent during China’s economic transition. The first story involves an enterprise owner who became rich by joining a real estate hunting group, the second story involves a person who became rich by inheriting his family’s business, and the third story involves a family who became rich after being compensated for their house’s demolition.

Intervention (Hidden)

Intervention Start Date

2024-07-12

Intervention End Date

2024-07-30

Primary Outcomes

Primary Outcomes (end points)

Support for redistributive policies; Opinions on government duty

Primary Outcomes (explanation)

For redistributive policies, we ask the respondent’s opinion on 10 real-stake redistributive policies that could be implemented or have been implemented in history and calculate their average z-score for tax-rich policies, help-poor policies and all policies to avoid multiple hypothesis testing. For government duty, we similarly ask 4 questions on government duty (generally reducing inequality, uniform standard of admission, providing jobs and whether it is just for the government to regulate wealth and income) and compile a z-score.

Hypothesis on Primary Outcome: Given the previous survey, we expect that the overall preference for redistribution will decrease in the treatment group that has seen the vignette. Particularly, we expect the tax-rich index to decline more than the help-poor index. In the previous study, we conjectured that the representative vignettes were perceived as fair in the transitional context, and thus seeing them evoked a sense of deservingness among the respondents that led them to be more hesitant in selecting redistributive policies in the following redistributive questions, particularly tax-rich policies. Please see the explanation section for the secondary outcomes for analysis of the mechanism.

Secondary Outcomes

Secondary Outcomes (end points)

Beliefs about the sources of wealth in today’s society, Attribution of the sources of wealth in the vignettes to different factors

Secondary Outcomes (explanation)

Beliefs about the sources of wealth in today’s society

Secondary outcomes (explanation): Respondents will be asked to indicate on separate Likert scales to what extent they think being rich in today’s society is attributable to: 1) Individual ability and effort, 2) Pure luck, 3) Opportunities present in the specific time period, 4) Connection, privilege and inequality of opportunity. For transparency, we will treat the answer to each of these questions as separate outcomes.

Hypothesis on Secondary Outcome: We hypothesize that the Chinese public regards all the opportunities present in the transitional period, and maybe also pure luck, as fair, despite significant randomness and a lack of extraordinary effort or ability. The only unacceptable form of inequality are privilege and structural inequality of opportunities, which dominated the pre-reform period and still play a large role in producing wealth today.

This fairness view is slightly deviant from the standard meritocratic view, where only merit (item (1) effort and ability) are considered just and not the rest (including item (2) (3) (4)), yet an “anything but political privilege” mentality is rationalizable given China’s historical trajectories.

If the representative vignettes are indeed considered examples of period-specific chances, then seeing these examples temporarily primes the respondents to believe that wealth today is due to period-specific opportunities rather than the (unfair) connection and privilege. We expect to see the following results:

- compared to the control group, the treatment group attributes the sources of wealth in today’s society more to opportunities in the transitional period and less to connection, privilege, and inequality of opportunities.

In addition, we would like to test the following,

- The respondents (both treated and control) indeed attribute the examples overwhelmingly to period-specific opportunities, instead of ability and effort or privilege.
- A higher attribution of contemporary inequalities to fair sources of wealth ( (1) and (3), maybe pure luck (2) too) is negatively correlated to demand for redistribution, holding other factors (such as demographics) constant.

Experimental Design

The experiment will be conducted through a survey. After being asked if they consent to participate in the experiments, respondents will be asked to answer a series of demographic, socio-economic, and subjective well-being questions, as well as questions about their views on inequalities. Treated respondents will then be presented with the vignettes (described in “intervention"). Treated and control respondents will then be asked a series of questions on whether they agree or disagree (from “strongly agree" to “strongly disagree") with specific policies to reduce inequality (primary outcomes), and on what are, according to them, the main sources of wealth in today’s society (secondary outcomes). Treated and control respondents will then also be asked, for each of the vignettes, to what extent they attribute the wealth of the individuals in the vignette to individual ability and effort, pure luck, opportunities present in the specific period, or connection, privilege and inequality of opportunity.

Experimental Design Details

At the end of the block of questions on respondents’ personal characteristics (demographic, socio-economic, subjective well-being questions and general views on inequalities), respondents are asked how much they would give to a random stranger in China in a hypothetical dictator game. At the very end of the survey, respondents are additionally asked a series of questions about their personal attitudes (reluctance to intervening in other people’s business, fear of making mistakes, preference to delegate difficult decisions, perceived zero-sumness of the world, libertarianism). The direction of the statement with which respondents are asked whether they strongly agree is systematically randomized to correct for acquiescence bias. Responses to those questions will be used to study the prevalence of those personality traits within the Chinese population and their demographic correlates to guide the design of another survey experiment on status-quo conformity.

Randomization Method

The survey will be executed through a Chinese survey company. They are in charge of collection of data as well as the randomisation procedure and randomise by a computer.

Randomization Unit

Individuals

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

2000 respondents

Sample size: planned number of observations

2000 respondents

Sample size (or number of clusters) by treatment arms

1000 respondents for the control group, 1000 respondents for the treatment group

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

With the control group and treatment group each at 1000 respondents, 5% alpha and 80% power, the minimum detectable effect size is 0.125 standard deviations. This is similar to the size of the effect we have previously obtained with the pilot study (between 0.1 and 0.15 standard deviations) and the literature.

Supporting Documents and Materials

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

IRB