ANALYSIS: Tracking Disinformation, Trust and Security in Taiwan

ANALYSIS: Tracking Disinformation, Trust and Security in Taiwan
Credit: Pixabay

What you need to know

Taiwan's battle against disinformation opens the door to familiar questions about the importance of freedom vs. security.

The Organization for Economic Cooperation and Development (OECD) released a report entitled “How's Life in the Digital Age?” last month, exploring the impact of digital transformation on individuals. In particular, it focuses “on individual, rather than economic and societal impacts of the digital transformation […] to assess how the digital transformation affects well-being at the individual level.” In other words, the report looks at how individuals like you and I are directly affected by digital changes, and not just on broad economic terms. A critique of Taiwan’s approach in its digital strategy has been that it is too focused on the economic potential of digitalization, but lacks focus on the human aspect. OECD’s report could thus provide insight as to how Taiwan’s digital strategy could incorporate the needs of individuals better.

The “How's Life in the Digital Age?” report is part of the OECD’s Better Life Initiative which measures well-being via indicators in 11 dimensions, broadly encompassing wealth and inequality, jobs and education, healthcare, work-life balance, political engagement and subjective well-being and security. Accordingly, the report compiles cross-country comparison data of indicators such as the unequal access to digital tools, the risks of job polarization which could result from automation, and job strain and satisfaction that can arise from digital use, among others. The report is quite a long one (and is worth a read!) but in this article, I will focus on its relevance to some of the most current issues that are ongoing in Taiwan right now: on the relationship between “fake news” and trust, the spread of false information by foreign governments, open data and its accessibility, and the threat of digital insecurity to digital transformation. This article will highlight the key data from OECD’s report in these areas, and provide comparison data from Taiwan, as well as expand on the issues discussed, to explore their impact in the context of Taiwan and parts of Asia.

Disinformation and the effect on trust in government

The porousness of the Internet has led to it being used as a medium to spread “fake news.” This is one issue highlighted in OECD’s report, which makes an interesting comparison between “fake news” and confidence in the government. The report avoids the use of the term “fake news,” however. According to the United Nations Educational, Scientific and Cultural Organization (UNESCO), “fake news” is a “straightforward or commonly-understood meaning” but such a term is an oxymoron, it says, because the concept of “news” itself should refer to information that has already been verified and should therefore consequently not be labeled as “fake.” More apt terminologies would thus be “disinformation,” which the OECD uses, or “misinformation,” which the Reuters Institute for the Study of Journalism uses. Whereas both relate to information that is false, UNESCO differentiates between the two based on whether the information is “deliberately created” to cause harm, thus “disinformation,” or whether it is not intentional, and therefore “misinformation.”

OECD’s report found that “there does seem to be a relationship […] between the level of exposure to disinformation and trust in government across countries” and that “self-reported experiences of disinformation are higher in countries where trust in government is lower.” The report highlights that Denmark, Germany and the Netherlands had the lowest levels of exposure to “completely made-up news.” at 9 percent, 9 percent and 10 percent, respectively. Greece, Mexico and Hungary ranked as the highest, with 44 percent, 43 percent and 42 percent who were exposed. Correspondingly, citizens in Denmark, Germany and the Netherlands had high levels of trust in their governments, at 57 percent, 62 percent and 67 percent, respectively, whereas the trust levels in government in Greece, Mexico and Hungary were 14 percent, 26 percent and 38 percent, respectively. OECD therefore surmises: “Self-reported experiences of disinformation are higher in countries where trust in government is lower.”

The relationship between disinformation and trust in the government is somewhat more complicated in Asia, however. OECD’s report does not cover Taiwan but in the Reuters Institute Digital News Report 2018 (published by the Reuters Institute for the Study of Journalism), where the data on disinformation is obtained from, 26 percent of Taiwanese respondents had reported being exposed to completely made-up news. This compares with 17 percent in Japan, 19 percent in Singapore, 30 percent in South Korea and 44 percent in Malaysia. Taiwan ranked somewhere in the middle among the countries surveyed. In terms of institutional trust, the Asian Barometer Survey 2010-2012 found that only 35 percent of Taiwanese trusted political institutions. Taiwan was categorized with Japan (36 percent) and South Korea (32 percent) as liberal democracies where trust averaged 35 percent. On the other hand, in “non-democracies” like China (75 percent), Singapore (81 percent) and Malaysia (74 percent), trust averaged 78 percent. Unlike in the OECD countries where lower levels of reported exposure to disinformation shows a relationship with higher levels of trust in government, in Asia lower levels of reported exposure to disinformation seems to be more related to having lower levels of trust in government, while in Malaysia, there are high levels of political trust but also high levels of reported exposure to disinformation.


Figure 1: Relationship between political trust and reported exposure to disinformation. Source: Asian Barometer Survey 2010-2012 (data on political trust), Reuters Institute Digital News Report 2018 (data on reported exposure to disinformation)

The 2014-2016 survey shed more light on the form of trust in these countries: In Taiwan, 61 percent of the respondents were considered as critical, where institutional trust is not only low, but where perceived corruption is high, whereas in Japan and South Korea, 43 percent and 38 percent of the respondents, respectively, were seen as being demanding (where trust is low but perceived corruption is also low). In authoritarian regimes like China (78 percent) and Singapore (83 percent), their respondents were considered supportive (where trust is high but perceived corruption is low). As such, the relationship between disinformation and trust in government is not as straightforward in the East Asian countries: whereas the reported levels of exposure to disinformation is relatively low in both Japan and Singapore, the reported level of trust is much higher in Singapore on the one hand but low in Japan, on the other. On the contrary, the reported level of trust is also high in Malaysia but so is the reported level of exposure to disinformation.

Paola Rivetti and Francesco Cavatorta explained that in authoritarian regimes, where citizens “‘trust’ the government to take a specific course of action, […] this does not mean that the action of the government will be ethical or legal or just” but that “actions of the institutions of the state become entirely predictable and therefore trustworthy because one knows exactly what will occur in a given scenario.” Rivetti and Cavatorta defined this as “negative trust” where the authority can be trusted to “act predictably” though not necessarily ethically or legally, as opposed to “positive trust” where the authority is trusted to take “ethical, legal or just actions.” In short, citizens in these authoritarian regimes “trust” their governments insofar as they expect their governments to be consistent in carrying out actions, though not necessarily in the interests of the citizens. Moreover, there is also the element of fear and repression: in their study of Zimbabwe, Omar García-Ponce and Benjamin Pasquale found that “in the face of a government that regularly perpetrates violence and repression for electoral ends, citizens may falsify their political preferences, with the aim of enhancing their personal security.” They added: “If respondents believe the state is using the survey to guide its (often violent) tactics and campaigns, respondents should maximize their personal security by avoiding survey responses that are critical of the […] state more generally.” In its study, Reuters also asked a question on whether respondents were concerned with whether openly expressing their view on the Internet could get them into trouble with the authorities, and pointed out that in authoritarian countries like Turkey, Singapore and Malaysia, concern was the highest, at 65 percent, 63 percent and 57 percent, respectively. When comparing this with the level of political trust in Asia, there does seem to be a relationship: in the authoritarian countries of Singapore and Malaysia, political trust was highest but concern with getting into trouble with the government was also highest, which might explain why the relationship between reported exposure to disinformation and political trust in Asia follows a different relationship as that of the OECD countries. (Note that this survey was done in February last year before Malaysia had a change of government in May, and the situation might have now changed after the electorate decided to abandon the former authoritarian regime which had ruled for more than 60 years.)


Figure 2: Relationship between political trust and concern with getting into trouble with the government. Source: Asian Barometer Survey 2010-2012 (data on political trust), Reuters Institute Digital News Report 2018 (data on concern with getting into trouble with the government)

On the other hand, even though low institutional trust exists in Taiwan, Japan and South Korea, this lack of trust in Taiwan differs in that citizens also believe that corruption is high, and this coupled with the high polarization in Taiwan’s media system and the high penetration of social media and interaction with news online could have an impact on the trust in the government. In fact, in a survey by Wang Tai-Li (王泰俐), Professor of Journalism at National Taiwan University, she found that about half of Taiwanese could not tell if six falsified reports were true or not, even though the higher-income were better able to identify falsified reports: 61 percent of respondents who earned between NT$50,000 and NT$100,000 a month could identify falsified reports while among those who earned below NT$10,000, it was 37 percent. Interestingly, younger voters were less able to identify falsified reports (43.5 percent were able to do so among those aged 20 to 30) than older voters (59.6 percent among those 50 to 59). Wang speculated that this could be because younger voters “spend more time on social media, where [falsified news] most often circulates,” the Taipei Times reported her as saying.

Dissemination of disinformation from foreign governments and the impact on disinformation exposure

But yet another reason for Taiwan’s low institutional trust and relatively high levels of reported exposure to disinformation, as compared to other high-country income countries, could also lie in the dissemination of false information in Taiwan by foreign governments. In data recently released by V-Dem under the Digital Society Project, it was found that among 179 countries, Taiwan ranks as the country which is most exposed to the dissemination of “misleading viewpoints or false information” by foreign governments, aimed at influencing the domestic politics in Taiwan. Countries were scored by experts on a scale from 0 to 4, where a score of “0” refers to a country having been exposed to the dissemination of false information on all key political issues by foreign governments, while a country which has never been exposed to any such false information by foreign governments would score “4.” Taiwan scored 0.264, followed by Latvia with 0.524 and Bahrain with 0.973. When comparing the relationship to reported exposure to disinformation, you do also see a general pattern where countries with higher levels of foreign intervention in their information sources do also have higher levels of reported exposure to disinformation. On the other hand, the dissemination of disinformation by the Taiwanese government, both internally and externally, is low.


Figure 3: Relationship between exposure to dissemination of false information from foreign governments and reported exposure to disinformation. Source: Asian Barometer Survey 2010-2012 (data on false information from foreign governments), Reuters Institute Digital News Report 2018 (data on reported exposure to disinformation)

An immediate assumption one might have is that the level of foreign intervention in Taiwan’s information sources is high because Taiwan’s media freedom is “too free.” Taiwan Media Watch chairman Lai Ting-ming (賴鼎銘) remarked last year that “a lot of people say free speech in Taiwan is too free,” but he also said that “freedom of expression is the most precious thing about Taiwan.” Indeed, when you compare countries with their levels of press freedom as measured by Reporters Without Borders, countries which have higher levels of press freedom actually also have lower reported levels of exposure to disinformation. It would therefore suggest that having greater press freedom would help guarantee a greater ability to discern disinformation.


Figure 4: Relationship between reported exposure to disinformation and press freedom. Source: Reuters Institute Digital News Report 2018 (data on reported exposure to disinformation), Reporters Without Borders (data on press freedom)

But if press freedom is not the issue, then what is it? Reuters also asked a question on trust in news media, and when we do a comparison, we do see somewhat of a pattern where in countries where trust in news is high, reported levels of exposure to disinformation would also be lower. In Reuters’ survey, Taiwan actually ranks as 6th lowest in terms of trust in newsf among 37 countries compared. Therefore, it is not “too much freedom” that Taiwan’s news media is suffering from. Rather, it is about the quality of news that Taiwan provides which affects the trust in them. To this, Professor Wang, in an interview with Liberty Times, said: “With regards to Taiwan, some media agencies are citing unverified or greatly biased information directly from the Internet in their reports. This is seen as one of the greatest sources of false news,” which she said is also “due in part to inflammatory comments by politicians and biased reporting from agencies.”


Figure 5: Relationship between reported exposure to disinformation and trust in news most of the time. Source: Reuters Institute Digital News Report 2018

Gary Rawnsley and Ming-Yeh T. Rawnsley wrote about how countries which transited to democracy had faced “media wars” where the “demise of state repression and control” would lead to “an open playing field [which would] generate new public spaces exposed to both political and commercial competition.” But they also pointed out how such “transitional systems [would also] sacrifice the democratic ideal for profit and commercial growth,” which has led to “a rise in sensationalism and a potential loss of quality” in Taiwan’s news media, they cited a 2010 Freedom House report as saying. But such reliance on commercial growth has also led to vulnerabilities in Taiwan’s news media system, which has led to the ability of foreign governments, particularly of China, infiltrating Taiwan’s media landscape and taking over ownership of some media companies, and therefore the possibility of information (and disinformation) being spread via foreign interference. In fact, this is even done with the encouragement of China’s government, such as with Want Want Group’s purchase of China Times (中時電子報) in 2008, which changed the newspaper from being one of Taiwan’s “most liberal papers” to being one “widely regarded as pro-Beijing”. It was even reported by a senior Taiwanese government official that because the Publicity Department of the Central Committee of the Communist Party of China (CCPPD) was worried that China Times would be taken over by Next Media (壹傳媒), that the CCPPD actually conspired with a senior Kuomintang (KMT) leader to convince the current China Times owner Tsai Eng-meng (蔡衍明) to purchase the newspaper.

But it is not that Taiwan does not have the existing structures to deal with the irresponsible spread of disinformation. It does and the National Communications Commission (NCC) had said previously that the Radio and Television Act and the Satellite Broadcasting Act require broadcast media to “check the facts of news reports that they investigate and apply the principle of equality”, which otherwise failure to do so and thereby “harm[ing] public interests and disrupt[ing] public order, could result in a fine of up to NT$2 million (US$64,832)”. However, by March 2019, the NCC had only issued fines of NT$200,000 (US$6,493) to television stations Eastern Broadcasting (東森電視) and CTiTV (中天電視) under the act, leading to its former chairperson Nicole Chan (詹婷怡) recently resigning due to criticisms of the NCC’s inaction in curbing disinformation. But before Chan left, she said: “The Taiwanese media’s ability to regulate itself has completely failed. I myself have been on multiple TV channels’ journalistic self-regulatory committees. The majority of committee members were rubber stamps.” She then added that Taiwan might need to have an “outside regulatory body” to provide oversight of the media, as this “is the best choice among bad solutions, and at least is better than doing nothing,” she said.

Perhaps, then, it is not free press but an unregulated free market that Taiwan should be worried about in its fight against the spread of disinformation.

In survey results released by the Professor Huang Kun-huei Education Foundation last weekend, it was found that 74.1 percent of Taiwanese thought that national security is more important than freedom of the press while 16.6 percent thought the opposite. However, this might be a false dichotomy because as can be seen from the comparisons above, countries with greater press freedoms also reported lower levels of exposure to disinformation, which suggests that open information could enable the citizenry to access information to help them better critically appraise news, and to fend off disinformation. Nonetheless, national security is a real concern and if the quality of news is the issue here, then NCC should step up their enforcement role as given under the current regulations. The eroding trust in Taiwan’s media and the foreign intervention in Taiwan’s information sources is therefore not due to an overly free press but a complex negotiation between a country still undergoing democratization whilst having its independence challenged by a non-democratic entity obsessed with inducting the former into the latter, while using the country’s free market system against its will.

But as OECD’s report highlighted, even though there is an inverse relationship between exposure to disinformation and trust in government, it is unclear whether the lower trust is due to disinformation or whether there are “deeper institutional and societal factors” that could have created better governance and trust, and therefore resilience against disinformation. Moreover, the question of trust and its relationship vis-à-vis regime-type and foreign intervention warrants further investigation. Also, there is still the question of how data on disinformation disseminated by the state can be captured and quantified for comparison purposes, or whether this is even doable amidst the secrecy. It would also not be far-fetched to think that governments would have internally collected data to assess the effectiveness of their disinformation campaigns on trust in the (local or foreign) government.

Open data as a double-edged sword against disinformation

Nonetheless, the comparisons above show that press freedom and open information is a tool to defend against disinformation. This is also what OECD’s report highlights: that the declining trust in the government can be mitigated by open data and a willingness by the government to share information. In terms of open data, Korea, France and Japan ranks tops while for citizens who have used the Internet for visiting or interacting with government websites, the Nordic countries, the Netherlands and Estonia came out tops, with at least three quarters of the population having done so in the last 12 months prior to the survey being conducted. The usage was above 80 percent in the Nordic countries and as high as 89.2 percent in Denmark.

Taiwan actually does not rank that badly in terms of open data. In fact, it comes up tops in the Global Open Data Index. However, in spite of the availability of open data, Taiwanese do not seem to be taking advantage of it as much: the Survey on 2018 Individual/Household Digital Opportunity Survey in Taiwan shows that only 60 percent of Taiwanese had “contact experience” with e-government services, and only 27.8 percent had used e-government online applications. One reason for the relatively lower access of e-government services in Taiwan could be due the lack of integration of e-services with the digital infrastructure. McKinsey & Company pointed out that even though Taiwan’s “government is one of the most digitally advanced sectors in Taiwan, […] it is not at the level of international best practice.” McKinsey highlighted that, “for example, numerous government agencies still print out official electronic documents for review before approving them online.” McKinsey added quite matter-of-factly: “No matter how well-designed a website might be, it won’t improve people’s experience with government services if they must stand in long lines, deal with cumbersome paperwork, or make endless calls just to apply for an identity card, register a new vehicle, file taxes, or set up a business.” In other words, Taiwan’s digital strategy cannot simply be an upgrading of the digital infrastructure without the accompanying systems integration to make use of the infrastructure.

Then, there is also the question of trust in the data that the government provides. A report in CommonWealth Magazine earlier this month showed that, in the pollution emissions readings that the Environmental Protection Administration (EPA) collected at the smokestacks of 111 factories in Taiwan, via the continuous emission monitoring system (CEMS), “out of every four readings showing higher-than-allowed emissions, three were declared invalid, meaning those smokestacks never officially exceeded pollution standards,” which led the magazine to question if the data has been “tainted.” CommonWealth Magazine also found through their own on-site investigations at the Taichung City Refuse Incineration Plant and Chiayi’s Chiahui Power Plant that emission readings which exceeded permitted levels were “coded as invalid, seemingly to make the numbers look better than they should have been.” These 111 factories where emissions readings were collected account for 73 percent of Taiwan’s sulfur oxides and nitrogen oxides emissions, CommonWealth Magazine reported. Their report generated a great deal of controversy, leading the EPA to attempt to clarify that emissions readings were declared invalid “due to regular maintenance on the automatic monitoring devices” and that even so, 98.05 percent of its data was still made available in 2017, with a 97.88 percent data availability in 2018, which it said is “higher than the national threshold of 75 percent, which is based on US standards”. EPA also blamed the data issues on its aged monitoring devices (some devices were older than 15 years old), leading to the Action Coalition for Healthy Air in Taiwan to question why the agency had only replaced five of its data collection stations as at the end of last year, even though it had said in 2012 that it would buy new equipment. But this episode with the EPA has also led to questions as to how “open” Taiwan’s open data really is.

The societal risks of digital insecurity on trust

OCED’s report also highlighted another dimension of trust and how it interacts with digitalization. It says: “Trust in digital tools and applications are essential for reaping the well-being benefits of the digital transformation.” Unlike the way cybersecurity is often seen as a national security issue, OECD pointed out that the risks of digital security can impact on the well-being of individuals, albeit in a more direct way than physical threats. Specifically, the OECD pointed out: “If people do not feel secure online, they will be more reluctant to engage in the digital economy, inhibiting this from unlocking its full potential.” This is especially important as one of the government’s plans is to transform Taiwan into an AI International Innovation Hub. OECD’s report highlighted two indicators to illustrate digital security risks: In terms of the indicator of reported online security incidents, the top three countries with the most reports were Luxembourg (28.6 percent), France (28.4 percent) and Hungary (28.1 percent). On the other end, the Czech Republic (7.6 percent), the Netherlands (8.7 percent) and Slovakia (9.1 percent) had the fewest such incidents. In comparison, 44.4 percent of Taiwanese reported having their accounts stolen, 18.5 percent had experienced Internet fraud and 13.2 percent had Trojans implanted in their websites – these were also among the top five Internet security incidents that Taiwanese reported experiencing in the Survey on Broadband Internet Usage in Taiwan. It is not possible to find directly similar data in Taiwan but this survey by the Taiwan Network Information Center provides relevant data. OECD’s report also looks for comparable data, in the absence of directly similar data: In the reference data that the OECD uses for Mexico considers data on “virus infection” and “fraud with information” as well. In terms of privacy abuse, South Koreans had the highest number of reports of having experienced abuse of their private information, at 6.0 percent, followed by Chile (5.4 percent) and Luxembourg (4.3 percent), while the Czech Republic (0.7 percent), Lithuania (0.9 percent) and Latvia (1.1 percent) had the least. In Taiwan, 14.8 percent of respondents indicated having their personal privacy exposed on the Internet while 10.6 percent reported having their online identities stolen – these were also the third and fifth most reported Internet security incidents.

However, OECD’s report also notes that the ability to deal with cybersecurity incidents differs even among digitally advanced countries, which means that even highly-digitalized countries can face difficulties managing digital risks. For example, the OECD ranks Sweden and Denmark as having high digital opportunities, but they are also ranked as having high digital risks. On the other hand, the OECD also explained that “Internet diffusion does not mechanically bring about higher risks” as the correlation between digital risks and access to information and communication technologies is low, so the key therefore lies mainly in whether countries would be able to identify digital risks and devise strategies to manage them. Similarly, as I explained above, as greater press freedom does not show a correlation with greater reported exposure to disinformation, but in fact press freedom might even facilitate the fending off of disinformation, then the key is not about limiting press freedom, but to ensure the enforcement of rules to protect the quality journalism and news.

However, the OECD points out that there are currently only three countries which have developed a “sound regulatory approach for digital environments,” which the OECD cautions: “The key adverse effects [of] digital transformation, for example [in terms of] the sources and consequences of extreme use or the spread of misinformation online, may not be addressed.” It would seem that even among the most digitally-connected countries that risk strategies have not been adequately developed to address the potential adverse impact of digital transformation, which Taiwan should take note to integrate early on into our digital strategy. An interesting point to note is that OECD’s report found the strongest cross-country correlation between the prevalence of digital security incidents (reported online security incidents and reported abuse of their private information) and the risks of digital transformation as a whole (such as exposure to disinformation and job polarization risks). As such, keeping track of the digital security incidents that occur in a country would also be able to aid in assessing the effectiveness of the country’s digital strategy.

In closing, OECD’s report therefore helps shed light on the relationship between disinformation and trust, and how these are mediated by open data and digital security. The comprehensive data collection in OECD’s report also provide a framework as to how Taiwan can develop a deeper understanding of the impacts of digital transformation to the individual and society, which could aid in developing corresponding policies to deal with these. I have also sought to further show how the relationship between exposure to disinformation and political trust in Asia is not as clear-cut as it is in the OECD countries, but could also be subjected to the types of political regimes in question, especially in Asia where a mix of authoritarian and newly-democratic countries exist side-by-side, and how the spread of disinformation by foreign governments could undermine democratic development, such as that of Taiwan. Moreover, even as open data can be used as a tool to circumvent the effects of disinformation, governments may counter such use by presenting a façade of open data, while simultaneously sabotaging attempts at allowing the data to be truly available. The relatively high prevalence of digital security incidents in Taiwan is also noteworthy as this could impact on the trust that individuals have of the government’s ability to ensure digital security, and their willingness to interact with new digital technologies, and therefore the extent to which Taiwan’s digital plans can materialize. OCED’s report therefore adds value with its approach in seeking to understand the impact of digital transformation on individuals, and it would therefore be worthwhile for Taiwan to conduct a similar endeavor as the OECD by refining existing research and identifying data gaps (and how they can be filled) to also chart a similar social risk and impact assessment of digital transformation in Taiwan.

Roy Ngerng is an Assistant Researcher at the Risk and Society Policy Research Center.

Read Next: Taiwan Takes Centerstage in Global Fight Against 'Fake News'

Editor: Nick Aspinwall (@Nick1Aspinwall)

If you enjoyed this article and want to receive more like it in your news feed, please be sure to like our Facebook page below.