How markets shape and promote values: a survey study among UK sellers and Chinese consumers

Market treatment: each UK participant will first express his view on a statement related to democratic values, which will be revealed to a Chinese participant that we recruit separately. The UK participant will make a sales offer for a virtual gift card to the Chinese participant, which the Chinese participant can accept or reject.
Non-market treatment: each UK participant will first express his view on a statement related to democratic values, which will be revealed to a Chinese participant that we recruit separately. There is no subsequent market activity.

2020-08-18

2020-08-24

1. UK participants' agreement with the statement. 2. Chinese consumers' willingness-to-trade (binary); Chinese consumers' willingness-to-pay (coded as 0 if someone is unwilling to trade at all);

beliefs about Chinese consumers; general attitudes toward China; incentivized private attitude as measured by a donation allocation decision; price offers

We recruit a representative UK sample via Prolific in the role of sellers. During the study, these UK participants will evaluate the extent to which they agree with the statement, “Foreign countries should stop meddling in the internal affairs of China in the name of human rights, when other countries also confront similar issues. Whether China respects the rights of its own people can only be evaluated from its own citizens’ perspective and by its own standards”.

We plan to implement two treatment conditions, one market treatment condition and one non-market condition. In the market condition, the UK participants will sell a virtual gift card that is worth 10 Pounds to a random Chinese consumer that we separately recruit. In the non-market condition, such exchange opportunity does not exist. Note that, in both conditions, the UK participants’ responses to the statement, as well the content of the statement itself, will be translated into Chinese and disclosed to a random Chinese individual. Importantly, UK participants are aware of the upcoming disclosure (and the market activity in the Market condition) when expressing their views to the statement. Our main hypothesis is that UK participants will shift their response to the statement in a direction that caters to Chinese consumers in the market condition, relative to the non-market condition.

UK participants’ choices will then be matched with Chinese participants that we separately recruit from mainland China through Tencent. These Chinese participants will view each possible response to the statement, and will make decisions for the market activity based on each possible response. That is, they will first decide if they are willing to trade with a seller with a given view at all, and if yes, they will report a maximum willingness-to-pay conditional on that view. Afterward, one of their choices will be randomly selected to count. We hypothesize a strong first stage in both Chinese consumers' willingness-to-trade and willingness-to-pay decisions. In particular, they will be more likely to trade with sellers who agree with the statement. Accordingly, they will also be willing to pay higher prices to those who agree with the statement.

At the end of the experiment among UK participants, we also collect beliefs regarding how Chinese consumers’ react to different responses to the statement, personal attitudes towards China, etc. We also elicit UK participants’ value in an incentivized manner by asking them to allocate a donation of 1 Pound on their behalf between Human Rights Watch, an NGO that promotes human rights, and Confucius Institute, a state-funded organization that promotes Chinese culture. All these additional choices will remain private and not disclosed to Chinese participants. While we have no ex ante predictions on these secondary variables, we will analyze them for exploratory purposes.

randomization done by a computer

individual

No

500 UK participants and 125 Chinese participants*

500 UK participants and 125 Chinese participants*

250 UK participants in market condition; 250 UK participants in non-market condition; 125 Chinese participants*

*Note 1: the above is the effective sample size that we are aiming for based on a power calculation. Here we pre-register two exclusion criteria as follows. We will collect additional data points as compensation in case we exclude too many samples.
- Prolific cannot preclude those who have participated our previous pilot studies from a representative sample at the moment. Thus, we plan to exclude participants who have participated our pilots by manually matching their Prolific IDs after data collection of the UK study.
- We have limited knowledge about Tencent survey platform relative to other platforms such as mTurk or Prolific, which are more broadly used. Based on our pilot, the data quality is quite mixed. To impose a minimum requirement on attention, we thus plan to exclude participants who spent less than 90 seconds completing the Chinese survey, simply because reading through instructions will easily take up more than a minute.

*Note 2: there exists a non-negligible risk of our survey among Chinese participants getting shut down unexpectedly due to censorship, either manually or automatically triggered. If that happens before we collect enough data points, we will have to consider whether we can switch to one other platform in China or just stick to the data in hand.