Sugar Policy and Sustainability – Numeric Calorie Labels, Symbolic Calorie Labels, and No-Sugar Subsidy

Last registered on November 03, 2025

Pre-Trial

Trial Information

General Information

Title
Sugar Policy and Sustainability – Numeric Calorie Labels, Symbolic Calorie Labels, and No-Sugar Subsidy
RCT ID
AEARCTR-0017161
Initial registration date
November 02, 2025

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
November 03, 2025, 10:38 AM EST

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

Locations

Region

Primary Investigator

Affiliation
National Chi Nan University

Other Primary Investigator(s)

Additional Trial Information

Status
On going
Start date
2025-09-22
End date
2025-12-31
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Sugar-sweetened beverages (SSBs) are widely consumed globally, with intake levels exceeding the recommended daily limits for free sugars in many countries. As the largest source of added sugars in modern diets, SSBs provide calories without essential nutrients, contributing significantly to poor dietary quality and various health problems, including obesity, type 2 diabetes, cardiovascular diseases, dental caries, and osteoporosis (Malik & Hu, 2022; Brownell et al., 2009; Malik et al., 2013; Mozaffarian et al., 2011; Li et al., 2021; Peñalvo, 2024).
In response to these public health concerns, SSB taxes have been implemented in over 45 countries to promote better nutrition, improve health outcomes, and generate revenue for public health initiatives. Meta-analyses have shown that such taxes are associated with higher prices for taxed beverages and significant reductions in sales, with consumption decreasing by up to 12% for every 10% increase in price (Cabrera Escobar et al., 2013; Andreyeva, 2022). Despite some studies demonstrating the effectiveness of SSB policies, SSB taxes face challenges, including cultural variability, regressive economic impacts, and uneven effectiveness across regions and beverage types.
In Western countries, most studies on SSB taxes have focused on carbonated drinks, with limited attention to other beverage categories (Gollust et al., 2014; Bélanger-Gravel et al., 2019; Pell et al., 2019). In Asia, taxation policies similarly target standardized beverages (ready-to-drink SSBs), such as canned or bottled drinks, as these are easier to regulate due to consistent sugar content. For example, Thailand implemented a six-tiered tax rate in 2017 for drinks containing over 6 grams of sugar per 100 milliliters, while Malaysia introduced a similar policy in 2019, taxing beverages with sugar content exceeding 5 grams per 100 milliliters. Malaysia’s 2024 budget further proposes increasing the tax, reflecting the region’s ongoing commitment to sugar reduction (Phulkerd et al., 2020; Ku et al., 2024). However, these policies do not address non-standardized beverages like handmade drinks, which are more prevalent in Taiwan and some other Asian countries or even the café shops in Western countries. Additionally, the effectiveness of such policies has shown mixed results. For instance, a study in Singapore found that modest price differences highlighted through tax and subsidy messaging had little impact on high-sugar beverage sales in vending machines (Seah et al., 2018).
Taiwan represents a unique case for SSB taxation due to its distinctive beverage culture, dominated by hand-shaken drinks like bubble tea. Between 2009 and 2020, annual revenue from beverage shops in Taiwan more than doubled, from NT$43.3 billion to NT$97.8 billion . Bubble tea, often made with tea, sweetened milk, and tapioca balls, has become a cultural phenomenon. These beverages are typically customized by consumers in terms of sugar levels and ice content, complicating the standardization of sugar content and posing challenges for taxation policies based on fixed sugar measurements (Chang et al., 2021).
Sweetened beverages in Taiwan contain an average of 9.19 grams of sugar per 100 milliliters. A single medium (500 ml) or large (700 ml) serving often exceeds the recommended daily limit for added sugar intake, which is set at 10% of total daily calories—about 50 grams for a 2,000-calorie diet . For example, a 700 ml bubble tea with “full level of sugar” customization contains over 52.5 grams of added sugar, already surpassing the daily limit in one drink (Tseng et al., 2010; Pan, 2020; Shih et al., 2020).
A national survey found that 42.2% of adults aged 19 to 44 consume at least one sugary beverage daily, with males drinking more frequently than females. Teenagers in Taiwan consume nearly twice as much sugary beverages as their peers in other countries. This overconsumption contributes to significant health risks, including obesity, diabetes, and dental problems. The resulting healthcare costs are substantial, with an estimated annual expenditure of NT$87 billion for primary diagnoses related to sugary drink consumption and NT$233 billion for combined primary and secondary diagnoses (Chang et al., 2021).
Nudging strategies have been widely used to encourage healthier food and beverage choices (Grech and Allman-Farinelli, 2015; Hendren and Logomarsino, 2017). These interventions are often categorized into two main types: priming, which involves increasing the visibility, accessibility, and availability of healthier products, and salience, which uses prompts such as front-of-pack calorie labels, traffic light labels, or descriptive labels to attract consumer attention. Recent research suggests that front-of-pack labels, including text warnings, graphic warnings, sugar content indicators, and Health Star Ratings (HSR), can reduce the selection of sugary beverages among young adults (Billich et al., 2018).
However, the effectiveness of these approaches varies. For instance, Hoenink et al. (2021) found that a traffic light sugar labeling system on store shelves did not significantly reduce unhealthy beverage sales or increase healthier beverage choices. Other studies have examined the use of traffic light labels, symbolic labels, and numeric labels to nudge healthier food choices (VanEpps et al., 2016; Haasova and Florack, 2019; Franckle et al., 2018; Ellison et al., 2014). Most findings suggest that symbolic or traffic light labels are more effective than numeric labels. For example, Mauri et al. (2021) compared traffic light labels with sugar teaspoon imagery in an online experiment and found that sugar teaspoons were more effective for prompting healthier choices, though this effect was limited to smoothies. To date, no studies have applied sugar cube imagery to handcrafted beverages, making this a novel aspect of our research.
In this study, we aim to compare the effects of three informational strategies—numeric calorie imagery, traffic light imagery, and sugar cube imagery—on reducing the added sugar content in handcrafted beverages. The sugar cube imagery will visually represent the amount of sugar added to each drink, which is designed to provide a clearer understanding of sugar content and its implications for health. Additionally, we propose a novel sugar policy—a no-sugar subsidy, as opposed to a sugar tax—to reduce sugar intake within Taiwan’s unique dietary culture.
Furthermore, there is a growing recognition of the interplay between health-conscious behaviors and pro-environmental actions. For example, individuals adhering to health-conscious dietary patterns tend to report lower greenhouse gas emissions, driven by reduced energy intake rather than specific dietary choices (Asvatourian et al., 2018). Encouraging individuals to take responsibility for their health may also foster positive attitudes toward sustainability, including recycling and energy-saving behaviors (Castellini et al., 2023; Urbanovich and Bevan, 2020). This potential alignment between health and environmental objectives provides an opportunity to design interventions that simultaneously address these dual goals.
To test these concepts, we will conduct a three-stage field experiment in a bubble tea store, aiming to understand the effects of informational and financial interventions on consumer behavior:
1. Baseline Data Collection: For one month, we will gather data on beverage sales, including details of added sugar levels, without implementing any interventions. This will establish a benchmark for consumer behavior under normal conditions.
2. Information-Based Interventions: Over three weeks, we will display different informational prompts at the ordering counter, each presented for one week. These include numeric calorie labels, traffic light labels, and sugar cube labels to indicate the added sugar content. By testing these strategies, we aim to evaluate their effectiveness in reducing sugar consumption across approximately 4,000 cups purchased.
3. No-Added Sugar Subsidy: For the final two weeks, we will introduce a NT$5 discount per cup for consumers who opt for no-added-sugar beverages. In addition to collecting sales data on approximately 2,500 cups, we will recruit 500 participants to complete surveys capturing data on dietary habits, physical and mental health, pro-environmental behaviors such as using reusable cups, and beverage purchasing patterns. These surveys will provide additional insights into how financial incentives interact with consumer characteristics to influence decision-making.
This comprehensive approach aims to address gaps in health promotion and environmental sustainability in consumer behavior research. By examining the impacts of sugar content information and a no-sugar subsidy on handcrafted beverage purchase behavior, the findings will inform policymakers on how to design effective health and sustainability strategies tailored to Taiwan’s unique beverage market.
External Link(s)

Registration Citation

Citation
Huang, Chiu-Lin. 2025. "Sugar Policy and Sustainability – Numeric Calorie Labels, Symbolic Calorie Labels, and No-Sugar Subsidy ." AEA RCT Registry. November 03. https://doi.org/10.1257/rct.17161-1.0
Sponsors & Partners

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

Request Information
Experimental Details

Interventions

Intervention(s)
Information-Based Interventions: Over three weeks, we will display different informational prompts at the ordering counter, each presented for one week. These include numeric calorie labels, traffic light labels, and sugar cube labels to indicate the added sugar content. By testing these strategies, we aim to evaluate their effectiveness in reducing sugar consumption across approximately 4,000 cups purchased.
No-Added Sugar Subsidy: For the final two weeks, we will introduce a NT$5 discount per cup for consumers who opt for no-added-sugar beverages. In addition to collecting sales data on approximately 2,500 cups, we will recruit 500 participants to complete surveys capturing data on dietary habits, physical and mental health, pro-environmental behaviors such as using reusable cups, and beverage purchasing patterns. These surveys will provide additional insights into how financial incentives interact with consumer characteristics to influence decision-making.
Intervention (Hidden)
Intervention Start Date
2025-11-17
Intervention End Date
2025-11-30

Primary Outcomes

Primary Outcomes (end points)
Participants' choices of suagr levels: full sugar (52.5 grams, approximately 105% of the daily recommended added sugar intake), half sugar (35 grams, approximately 70% of the daily recommended added sugar intake), and light sugar (25 grams, approximately 50% of the daily recommended added sugar intake).
Participants’ binary decision to choose no-added-sugar options (yes or no).
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
We will conduct a field experiment in Taipei, the most populous city in Taiwan, at one of the bubble tea chain stores. The store owner has agreed to collaborate in this academic research project aimed at promoting health and sustainability.
The experiment will consist of two components:
1. Sales Data Collection with and without Information Intervention
This component is divided into three stages:
• Baseline Data Collection (one month): We will collect sales data, including the added sugar levels for each cup, without implementing any interventions. This will establish a benchmark for consumer behavior under normal conditions. Only walk-in purchases are recorded, excluding online or phone orders.
• Information-Based Interventions (3 weeks): Informational prompts will be displayed at the ordering counter, with each type of information presented for one week. These prompts will include numeric calorie information only, traffic light imagery with numeric calorie information, and sugar cube imagery with numeric calorie information to indicate added sugar content. Approximately 4,000 cups of walk-in purchases during this phase.
• No-Added Sugar Subsidy (2 weeks): During the first week, a NT$5 discount per cup is offered to consumers who opt for no-added-sugar beverages only. In the second week, the most effective informational strategy identified in the previous phase is combined with the subsidy—a NT$5 discount per cup for consumers who choose no-added-sugar beverages. Sales data for approximately 2,500 cups of walk-in purchases will be collected.
2. Detailed Consumer Information Collection
During the final intervention period (no-added sugar subsidy with/without information intervention), we will recruit 500 participants who purchase drinks for personal consumption. These participants will complete a survey to provide additional insights into their decision-making processes.
• Participant Criteria: Participants must be 18 years of age or older.
• Survey Content: The survey will capture data on dietary habits, physical and mental health, pro-environmental attitudes (e.g., the use of reusable cups), and beverage purchasing patterns.
• Incentives: Participants will spend approximately 15 minutes completing the survey and will receive NT$150 gift card as compensation.
• Recruitment Process: Sidewalk signs will be placed next to the bubble tea store to attract potential participants. Interested shoppers who meet the criteria will be guided to small tables near the recruitment site.
To simplify understanding and facilitate further analysis, we focus solely on the amount of added sugar in each cup of beverage, excluding the sugar in toppings and the natural sugar present in the ingredients. Since the majority of bubble tea stores in Taiwan, including the one where we will conduct our experiment, only offer large-sized beverages, this study will focus exclusively on data from large-sized drinks. The sugar content information provided will also specifically relate to large-sized beverages. To account for seasonal variations in drinking behavior, the experiment will be conducted within the same season.
2.3.2 Treatments – Different Imagery Information
In this study, we aim to explore how various types of sugar content imagery influence consumers’ choices for handcrafted beverages. Before placing their orders, consumers will be exposed to one of three information treatments: numeric calorie (NC) information, traffic light imagery with NC information, and sugar cube imagery with NC information. These conditions will create four experimental scenarios: (i) Control condition (no information), (ii) NC condition, (iii) Traffic light imagery with NC condition, and (iv) Sugar cube imagery with NC condition.
Numeric Calorie (NC) Information Treatment
For the NC condition, consumers will view numeric calorie information indicating the amount of added sugar in each beverage and its proportion of the recommended daily sugar intake as shown in Figure 3. According to Taiwan’s Ministry of Health and Welfare and National Taiwan University Hospital, added sugar intake should not exceed 10% of total daily calorie intake. For example, with a daily caloric intake of 2,000 calories, added sugar consumption should be limited to less than 50 grams. Typically, a 700 ml hand-shaken beverage contains the following sugar levels: full sugar (52.5 grams, approximately 105% of the daily recommended added sugar intake), half sugar (35 grams, approximately 70% of the daily recommended added sugar intake), and light sugar (25 grams, approximately 50% of the daily recommended added sugar intake) . This treatment provides consumers with essential data to help them make more informed decisions. We hypothesize that this information will lead to a reduction in the added sugar content chosen for each beverage, drawing from insights in the literature (VanEpps et al., 2016; Haasova and Florack, 2019; Franckle et al., 2018; Ellison et al., 2014).
Traffic Light Imagery with NC Information Treatment
Under this treatment, consumers will view traffic light imagery paired with numeric calorie information as shown in Figure 4. This visualization uses red, yellow, and green lights to represent high, medium, and low sugar content levels, respectively, based on the daily recommended sugar intake. For example, a red light signifies sugar content levels above half of the daily recommended intake, while a green light signifies levels below 25% of the daily recommended intake. This approach is supported by findings from Marks (2020) and guidelines from Taiwan’s Ministry of Health and Welfare. We hypothesize that this treatment will reduce the added sugar content more effectively than numeric information alone.
Sugar Cube Imagery with NC Information Treatment
In this condition, consumers will see a representation of sugar cubes indicating the number of cubes corresponding to the added sugar content in each beverage as shown in Figure 5. For example, a full-sugar beverage displays an image of 10.5 sugar cubes (based on 5 grams per cube), a half-sugar beverage displays an image of 7 sugar cubes, and a reduced-sugar beverage (30% of the added sugar level) displays an image of 5 sugar cubes. This visualization is designed to make sugar content more tangible and impactful, drawing on research by Mauri et al. (2021) and official recommendations from Taiwan’s Ministry of Health and Welfare. We hypothesize that the sugar cube imagery will reduce added sugar intake more effectively than numeric information alone.

Experimental Design Details
Randomization Method
The randomization adopts a cluster design based on weekly treatment assignments. Each treatment condition — numeric calorie label, traffic light imagery with numeric information, sugar cube imagery with numeric information, and the control condition (no information) — is implemented sequentially for one week at the same bubble tea store. During each week, all customers visiting the store are exposed to the same information treatment, ensuring internal consistency and avoiding contamination across conditions.
Although individual consumers are not randomly assigned, the weekly rotation of treatments effectively randomizes exposure across different consumer groups and time periods. The treatment order is pre-determined but counterbalanced to minimize systematic bias from weekday effects or seasonal variation.
Importantly, the experiment takes place in a popular tourist area in Taipei, where most customers are one-time visitors rather than regular patrons. This substantially reduces the likelihood of repeated exposure to multiple treatments or learning effects, thereby strengthening the validity of the cluster randomization design.
Randomization Unit
The unit of randomization is the week, corresponding to a cluster-level assignment.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
1500 cups of purchases
Sample size: planned number of observations
13,500 cups of purchases
Sample size (or number of clusters) by treatment arms
Baseline: No interventions. Around 6000 cups purchased will be collected.
Information-Based Interventions: Over three weeks, we will display different informational prompts at the ordering counter, each presented for one week. These include numeric calorie labels, traffic light labels, and sugar cube labels to indicate the added sugar content. By testing these strategies, we aim to evaluate their effectiveness in reducing sugar consumption across approximately 4,000 cups purchased. Around 1500 cups per week.
No-Added Sugar Subsidy: For the final two weeks, we will introduce a NT$5 discount per cup for consumers who opt for no-added-sugar beverages. In addition to collecting sales data on approximately 2,500 cups, we will recruit 500 participants to complete surveys capturing data on dietary habits, physical and mental health, pro-environmental behaviors such as using reusable cups, and beverage purchasing patterns. These surveys will provide additional insights into how financial incentives interact with consumer characteristics to influence decision-making.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
China Medical University & Hospital Research Ethics Center
IRB Approval Date
2025-09-18
IRB Approval Number
CRREC-114-037

Post-Trial

Post Trial Information

Study Withdrawal

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

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials