Building Consensus and Compromise on Uber in Taiwan

Tackling challenges together This case study is part of a series by CPI and Engage Britain examining innovations in citizen engagement that have used tech-enabled deliberative methods to enhance the policy-making process. There are seven assessed against the Public Impact Fundamentals and two briefings that describe processes for ideas gathering. The series demonstrates that though we are in the early days of using these methods, they offer much hope for more legitimate answers to the big public policy questions of our time. Such methods are also becoming more widespread but work best when used in combination with offline methods. Our paper pulls all the innovations together and offers a view on what we can learn so far.

In 2015, the issue of Uber regulation was addressed in Taiwan through a unique process of citizen engagement called vTaiwan. Uber’s arrival in Taiwan in 2013 presented several challenges, a key issue being the regulation of the company and ensuring fair competition with similar taxi services. vTaiwan brought together citizens and stakeholders to agree an approach to the issue. Using the Pol.is platform, the initiative was able to crowdsource ideas and identify areas of consensus between different parties. Throughout the process, the initiative utilised online and offline methods and different technologies to ensure the process was transparent and open to public engagement and scrutiny. Initially, groups were fiercely divided. However, by the end of the process, recommendations emerged that received almost universal approval. These suggestions were taken forward to talks with Uber, taxi drivers, and the government, which were broadcast live and transcribed. The process resulted in Uber and other groups making important concessions in response to the suggestions, and the government adopted new regulations in line with vTaiwan’s recommendations.[1]

The challenge

The ride-sharing app UberX, one of the services provided by Uber Inc, arrived in Taiwan in 2013. It was very popular with the public, but traditional taxi drivers were losing customers and there was much controversy surrounding its operation. Uber was registered as a technology company, but the Ministry of Transportation and Communication deemed Uber a transport company and ordered it to obey the taxi laws, something Uber initially refused to do. Taxi drivers and many members of the public felt Uber had an unfair advantage, as a result of a number of factors: Uber drivers did not need to have insurance or a professional driver’s licence; their charges undercut a fare structure set down in law for taxi firms; and the company was not paying the same taxes as local firms.[2] Uber’s arrival eventually sparked protest and civil disobedience from taxi drivers. 

The initiative

Because the arrival of Uber in Taiwan proved controversial, it required an effective response from the Taiwanese government. Communication between Uber Inc and the Ministry of Transportation and Communication had been problematic, and Uber Inc’s status as a technology company had generated challenges to traditional policy approaches. The Taiwanese minister for digital affairs, Jaclyn Tsai, therefore invited a civic tech community called g0v (pronounced “gov zero”) to support the government in addressing the issue. 

g0v is a decentralised community of coders, NGO workers, civil servants, and volunteers who develop digital tools to support a more open government. Based in Taiwan, they had been operating since 2012, but rose to prominence in 2014 following their role in supporting decision-making during the Taiwanese student protests, known as the Sunflower Movement. After attending a g0v hackathon in December 2014, Jaclyn Tsai asked g0v to build a platform that enabled citizens to engage in rational discussion online. The g0v community developed a new public engagement process called vTaiwan – the v stands for “vision”, “voice”, “vote” and “virtual”. It focuses on forming public participatory policy about digital technology.[3

In August 2015, at the request of several government authorities, including the Ministry of Transportation and Communications, the Ministry of Economic Affairs, and the Ministry of Finance, vTaiwan hosted the Uber case. The process was designed to support open and transparent deliberation, addressing what constitutes fair regulation of Uber in relation to competing services.[3][4] It brought together citizens and stakeholders who represented different perspectives on the issue. The process was broken down into four stages: the objective, reflective, interpretive and decision stages.

  1. Objective Stage

During the initial two-week objective stage, g0v researched the issue to help define the policy challenge and identify and contact relevant stakeholders. They agreed on a date to launch the engagement process, and were given a URL that everyone could share at once to reach their constituents.[5] The community gathered relevant facts, evidence and research on the topic and prepared material for the public. This included interpreting legal jargon and translating it into something more readily understandable for the purposes of presenting a topic description and initial statements to initiate public discussion.

  1. Reflective Stage

The reflective stage uses the Pol.is platform to crowdsource ideas and gather public opinion. Facebook adverts and social networks were used to target participants and draw them on to Pol.is, including reaching out to affected groups such as Uber drivers to ensure their perspective was included. The most prominent feature of the Pol.is platform is its visual and structural expression of patterns in support for user-generated opinions.[6] Participants vote on other users’ suggestions, where the options are to agree, disagree or “pass”. Participants may also contribute their own ideas. Pol.is provides visual feedback in the form of a map which highlights areas of consensus, as well as representing non-mainstream opinions.[7] There is a word count limit of 140 characters on contributions, and a key feature of the platform is that participants cannot reply directly to others. These design features prevent trolling and enable scalable communication.

The reflective stage took 4 weeks, during which time 4,500 people participated and voted on 145 comments. Participants included taxi drivers, UberX drivers, and passengers of both Ubers and traditional taxis. Two broad groups quickly emerged from the process identified on the basis of the statements they supported, one pro Uber and another (twice as large) that was fiercely anti-Uber.[5] The organisers decided that for an idea to progress to the next stage, it needed to achieve 80 percent approval among all participants. This was calculated on the basis that the participants were initially split 60-40, so the majority and at least half of the minority group would need to agree.[5] As a result of this, people would compete to define moderate statements that crossed the divide between the groups. From this process, 6 recommendations emerged which received approval from over 80 percent of contributors. There was a general consensus on the need to regulate UberX and protect established public-private transport. 

  1. Interpretive Stage

In the interpretive stage, online participants joined a two-hour public meeting with academics, industry experts, active online users from the Pol.is survey, and representatives from the following four stakeholders:

  • “The Association of Taxi Drivers in Taipei... 
  • “Taiwan Taxi, the country’s foremost taxi fleet
  • “Uber Inc...
  • “The Ministries of Transport, Economic Affairs and Finance.”[6]

The meeting combined the results of the Pol.is survey with more detailed discussions and idea exchange, enabling the development of firm proposals. There was further opportunity for public input and scrutiny at this stage through a process described as telepresence.[6] Using streaming service technologies such as LIVEhouse.in, the meetings were live-streamed and transcribed, and the public could participate remotely through online chat rooms and digital whiteboards, which could then feed back into the meetings. Over 1,800 people watched this event or participated remotely, for example asking questions via the chatroom.[5] Faced with public pressure and consensus around the demands, Uber conceded to almost all the recommendations before the legislative process began.

  1. Decision Stage

In the final decision stage, proposals emerging from the process were developed into a draft bill and sent to parliament. There was a delay of a few months at this stage due to a transition in the post of minister for transport, and the following proposals were eventually ratified in 2016: 

  • Taxis no longer need to be painted yellow
  • Taxis can display medallions in different ways
  • App-based taxi fleets cannot pick up passengers from the street randomly
  • People may not charge less than the standard taxi fare
  • The app-based fleets are subject to public auditing to ensure they display correct driver and car identification, how the fare is calculated including surge pricing and they must display average ratings from all customers.[5

Using Pol.is, the reflective stage generated the following six recommendations

  1. The government should set up a fair regulatory regime on transport instead of protecting certain groups with vested interests.
  2. On the issue of paying taxes in Taiwan, Uber has the responsibility to put forward a responsive mechanism to convince Taiwan’s community of its good faith.
  3. UberX should follow the practices of taxis and require their vehicles to display the registration certificate, licence and complete driver’s information at a visible place in the vehicle.
  4. Transport is similar to food medicine in that it should be subject to more stringent definitions and checks than other general service platforms.
  5. Private passenger vehicles should be registered. They should be limited to two shifts per day in order to achieve the car-pooling effect. An additional passenger protection insurance cover should also be purchased. 
  6. I think it should be permissible for a for-hire driver to join multiple fleets and platforms.[11]

The public impact

As a result of the conversations on Pol.is and face-to-ace meetings, the stakeholder groups made the following concessions:

  • Uber agreed to provide its international liability insurance policy to Jaclyn Tsai and, if needed, release it for public review
  • Uber agreed to mandate all drivers to register and obtain professional driver’s licences, and provide the necessary support
  • If legalised in some areas, Uber was willing to pay for UberX car permits, as well as transport taxes
  • The Taipei Taxi Association also expressed a willingness to work with the UberTAXI platform under mutually agreeable terms
  • The Taiwan Taxi Fleet promised to offer better services, if taxi pricing were to increase in response to market demand.[3]

On 23 May 2016, the administration pledged to ratify the recommendations emerging from the process and amend the Regulation on Automobile Transportation Management. The following legislative changes came into effect on 25 October 2016: 

  • Taxis no longer need to be painted yellow
  • Taxis can display medallions in different ways
  • App based taxi fleets cannot pick up passengers from the street randomly
  • People may not charge less than the standard taxi fare
  • The app based fleets are subject to public auditing to ensure they display correct driver and car identification, how the fare is calculated including surge pricing and they must display average ratings from all customers.[3]

The only issue on which representatives of Uber did not agree concerned their tax status. If they committed to being a taxable entity in Taiwan, then drivers would become their employees rather than being self-employed. At the time, Uber was fighting a legal battle on this issue in California and, had they made this concession in Taiwan, it would have been likely to sway the legal proceedings in California against them.[6] This issue has been a recurring one for the company’s operations. 

The process appeared to ease tensions between traditional taxi drivers and Uber, although new government legislation in 2019 may cause Uber to exit the Taiwanese market.[8]

Following the successful pilot of Pol.is with Uber, the vTaiwan process was applied to Airbnb regulations and internet alcohol sales.[3] To date, 26 issues have been discussed through vTaiwan, 80 percent of which have led to decisive action from the government.[9]

Written by Martin King

What did and didn't work

All cases in our Public Impact Observatory have been evaluated for performance against the elements of our Public Impact Fundamentals.

Legitimacy

Public Confidence Strong

The vTaiwan initiative emerged in the wake of a political crisis in Taiwan that had greatly undermined confidence in the Taiwanese government.[10] Jaclyn Tsai was keen to restore faith in political decision-making by using digital technology to enable open public engagement. She chose to reach out to g0v, the grassroots community, because of its experience and credibility.[10] The process through which the Uber case was discussed enabled strong involvement from citizens and various stakeholders.[3] The discussion was transparent, and the recommendations put forward gained the support of over 80 percent of citizens who participated.[3][5] Since the Uber case, participation in the vTaiwan process has increased and public engagement processes have expanded, suggesting confidence in the process and its organisers.[11]

Stakeholder Engagement Strong

“All stakeholders displayed a remarkable willingness to cooperate and work with each other”, says Audrey Tang.[3]

The organisers of the vTaiwan process were careful to engage the relevant stakeholders, including Uber Inc, other taxi companies, and the drivers and passengers of both services. The initiative also engaged representatives from the relevant government departments and the broader public. The use of Pol.is enabled a diverse range of voices to be represented with a transparent indication of levels of support for different ideas. Stakeholders and the public were involved throughout the process, notably through the use of technology such as Pol.is and through the two-hour, live-streamed face-to-face meetings.[6]

Stakeholders demonstrated a strong commitment to the process, as evidenced by their capacity to make concessions and work together. For example, Uber agreed to several points of action including coaching drivers to obtain professional driver’s licences, providing insurance, and paying for car permits. The Taipei Taxi Association expressed willingness to work with the UberX platform, and the Taiwan Taxi Fleet promised to offer better services.[6

Political Commitment Strong

Several government authorities displayed strong commitment to the initiative, dedicating time and resources to resolving the areas of disagreement, notably the Ministry of Transportation and Communications, the Ministry of Economic Affairs, and the Ministry of Finance. There is also evidence that Ministry of Transport staff developed a positive relationship with stakeholders during the process.[3][5] In addition, Jaclyn Tsai was strongly committed to the vTaiwan process that hosted the Uber discussions.[10]

It should be noted, however, that not all cases on vTaiwan have received the same level of commitment from government as the Uber case. One interviewer involved in the process observed that they have struggled to maintain autonomy from the government in terms of what issues get discussed.

Policy

Clear Objectives Good

In response to the challenge presented by Uber, the objectives of the initiative were to open up taxi competition to Uber in a way which was acceptable to the Taiwanese population and perceived as fair by other taxi services.[2][3] The initiative’s objectives were to also open up decision-making and enable greater citizen participation and transparency in the legislative process.[1][4] vTaiwan aimed to bring together citizens and stakeholders to set the agenda for debate using the Pol.is platform. In this sense, the specific agenda and recommendations were not set out clearly at the start of the initiative but emerged through the process in an open and transparent way. The final recommendations were described as very clear and coherent.[5]

Evidence Strong

The initiative established clear stages and methods for ensuring that the best and most relevant evidence was gathered for the Uber case. The vTaiwan process itself was developed on the basis of research into existing practices and years of experimentation within g0v.[11][5] Before its application in this case, it had been trialled on other cases, for example vTaiwan had hosted debates on regulation regarding crowdfunding platforms.[4] The initial stage of the process identified and reached out to stakeholders and those with expertise in the area, gathering and preparing research and legal reports relevant to the case. The Pol.is platform’s capacity to crowdsource ideas enabled 4,500 people to generate and scrutinise proposals relating to the regulation of Uber, improving the strength and quality of recommendations and providing clear evidence of the levels of support for each point of view. Finally, the meetings between Uber Inc, Taiwan Taxi, the Association of Taxi Drivers in Taipei, and government representatives were open to the public and followed by at least 1,800 people, ensuring greater accountability and transparency regarding the decisions made.[2]

Feasibility Good

In the Uber case, vTaiwan’s use of technology, notably the Pol.is platform, enabled it to support participation on a scale that would not otherwise have been achievable or would have been extremely expensive. The process also relies on support from volunteers, which has further reduced costs. 

vTaiwan is currently limited to addressing issues relating to digital technology, such as Uber. In order for the initiative to expand – in terms of addressing a wider range of topics and engaging a larger, more diverse group of participants the process will probably require more full-time staff, and the recommendations deliberative events would need to be more binding on the government.[11]

The success of the initiative relies on strong commitment from political actors, and it should be noted that not all cases on vTaiwan have been as successful as the Uber case. If the government refuses to discuss a public issue, the topic will not go through the vTaiwan process. A draft bill of selling alcohol online, developed through vTaiwan with strong public support, was rejected by the Legislative Yuan due to its conflict with existing e-liquor policies.[2]

Action

Management Fair

The initiative resulted in positive action from all major stakeholders. On 23 May 2016, the government ratified most of the recommendations emerging from the process. This included ensuring that Uber drivers were correctly registered and insured and that appropriate checks were carried out to confirm they were displaying the correct identification information and not undercutting fares. A key issue that Uber did not give way on – and the government did not enforce – was taxation, in part because this would have impacted a court case they were involved in.[2

It has been argued that the biggest limitation of vTaiwan is that the government is not bound by the discussions. Consequently, it is vulnerable to being what Jason Hsu, a former activist who helped bring vTaiwan into being, called a “tiger without teeth”.[11] Although the government implemented recommendations in the Uber case, there are limited means to ensure recommendations are well managed or acted upon. 

Measurement Fair

The success of the initiative can be measured by the extent to which it was able to create a level playing-field for Uber and other taxi services. In this respect, the results were mixed. The initiative did result in government legislation and decisive action from stakeholders. This included ensuring that Uber did not undercut existing metered pricing, opening Uber to public auditing, and ensuring that taxi drivers have professional licences, together with  greater flexibility for how taxis are presented (e.g removing the requirement that they are painted yellow, please see The Public Impact above). One of the biggest problems that Uber poses to lawmakers is its stance on taxation, and the failure or unwillingness of the Taiwanese government to address Uber’s tax advantages can be seen as the biggest limitation of the process.[2

A further measure of the initiative’s success is the extent to which the process opened up the legislative process and managed to engage citizens. In this respect, the initiative ensured that meetings between Uber and government officials took place in the open, being live-streamed and transcribed, rather than taking place behind closed doors, as is more typical of such meetings.[2] The initiative managed to engage 4,500 participants in the reflective stage and 1,800 participants in the interpretive stage.

Alignment Good

The initiative involved bringing together actors with different interests, including groups with business interests that were in direct competition with each other. The aim of the initiative was to facilitate a consensus between all stakeholders on the regulation of Uber. The stakeholders demonstrated good faith and a willingness to cooperate, during the initiative and also in the implementation of the measures.[3][5]