A major component of any form of government open data is trust that the information is true and reflects reality. Government statistics are one of the more essential vehicles to achieve this end, playing an integral role in not only in governance but everything from economic activity to overall citizenship. As The Economist puts it, without statistics, “governments would fumble in the dark, investors would waste money and electorates would struggle to hold their political leaders to account.”

In the digital age, the quality of government statistics has become even more vital as open data makes it easier for government to release not just static statistics, but raw information. It also allows for citizens to directly engage government information in a dynamic way. Along with the commercial potential that open data has uncovered, government data has also become even more relevant in the social and ethical arena, including its use as an increasingly vital component of human rights monitoring.

In previous posts, we addressed why and how open data can support human rights advocacy, especially highlighting the benefits that the concept can have on localizing data collection, enriching evidence for human rights violations and reducing biases in data collection. National statistics and the agencies that produce them can be a vital primary resource for those engaged in data-driven good-governance or human-rights initiatives. However, when they are static, national statistics require trust in government integrity, often the very thing human rights organizations are out to verify. Thus, concerns of unreliable government statistics — especially when it comes to human rights information that might unfavorably expose government — poses a huge barrier to accurate monitoring of human rights, one that can and should be addressed by opening the data and the methodologies that have produced the statistics in question.

The problem with national statistics

Across the world, national statistics agencies are often criticized for a lack of quality and reliability in the data produced. In an analysis of several government statistics websites, we found several instances where governments were engaged in practices that presented poor or incomplete data as it relates to human rights. Glaringly, many governments provide little-to-no data in categories related to human rights, despite the presence of a portal to access data. For example, while Saudi Arabia has a portal for government statistics, metrics pertaining to certain human rights are limited or absent (e.g. statistics related to the provision of health care services do not appear on their portal). What’s more, much of that data, such as labor force survey information, is available only in closed or nonmachine-readable formats, such as PDFs.

In this post, we will examine cases in which data has been not only omitted or presented in closed formats, but has even been manipulated to the detriment of public trust. A further example will suggest how open data can help citizens confront government manipulation of data by facilitating accountability and more legitimate human rights monitoring in the process.

Argentina: Manipulated national statistics and the cloud of mistrust it creates

By providing services, among other functions, governments gather information that should reflect the reality of their citizens. The use of government data to power various aspects of the economy or empower evidence-based public decisions has helped spawn a change, improving the day-to-day lives of the average citizen. As a result, the need for quality data from governments has become imperative. However, far too often governments engage in efforts to manipulate or corrupt data to protect their interests. There are myriad consequences, but trust is lost and problems worsen as there is not an accurate depiction of the conditions which are being observed.

Unreliable government data is a challenge that presents itself throughout the world. While many governments have been guilty of undermining their statistics, Argentina presents one blatant and recent example. During the Kirchner family’s Peronist rule, it was common knowledge that figures from the National Statistics Agency (INDEC) on the country’s GDP, employment and inflation were falsified. The government removed officials within INDEC and replaced them with loyalists to the government. Those who published independent figures were threatened with fines and other forms of intimidation.

The actions of the government toward its statistics agency and the government statistics it produced was a unique and embarrassing violation of the human rights of Argentines. Article Six of the International Covenant of Human Rights provides for a person to be able to make a living, through their work. When viewed through this lens, the falsifying of data the government undermined the economic viability of the country, harming investors who owned bonds that were tied to inflation and hindering the ability for people to provide for their families by misrepresenting, for example, the actual price of goods. Argentines will now have to struggle with steep price increases in certain sectors as the market corrects itself.

Beyond the problems of accuracy, the data itself is also closed. In rankings by the Open Knowledge Foundation, Argentina was found to be ranked 105th in quality of national statistics. The data available on INDEC’s portal was not machine readable or available in an accessible, digital format. Perhaps most importantly, it is not regularly updated, providing less accuracy and credibility.

All this reflected a flippant attitude by the former Argentine government toward the data it produces, which is to the detriment of the country. Along with an extended battle with creditors, Argentina was shunned by international financial markets because of this poor data, even getting sanctioned by the IMF, putting further strain on the economy. With the beginning of the Mauricio Macri administration, attempts to push out this bad data are only now being rectified; the government is beginning to open up about these falsifications and attempting to collect reliable figures. This along with an agreement reached with creditors signals the current government’s attempts to bring Argentina back from the financial wilderness.

The example of Argentina demonstrates what happens when a government actively engages in the manipulation of its data or presents it in a closed way. The primary consequence is that many aspects of governance, the economy and society are undermined by a lack of trust and reliability. As we noted, open data has the capacity to be transformative commercially but also in the ability to monitor and protect human rights. Reliable government statistics are an essential component — when that has been compromised, it not only hinders the ability for a country to advance but may compound existing problems.

Taiwan: Open data as a means of independent, decentralized verification

While Argentina shows how the publication of bad data can be stifling to government transparency and facilitate human right violations, by contrast, accurate and open data can be a tool for empowering public oversight when governments attempt to misrepresent data. An example of this comes from Taiwan, which recently grabbed the top spot on Open Knowledge’s 2015 Open Data Index.

Taiwan has several factors that led to its success. According to OKFN, the country possesses a highly active press environment and a high level of engagement with technology (especially with social media). That — combined with a large number of very active civil society groups who regularly hold the government to task — creates “enormous pressure” for transparency. This may explain the enthusiasm for open data by the government. As OKFN noted, the environment for open data in Taiwan facilitates a culture which provides space to “interpret data, check the integrity, report it, or even use it to hold … government accountable in litigation.”

In promoting this accomplishment, the government at the time demonstrated why open data, human rights monitoring and accountability are interwoven. The government released two studies analyzing statistics from the Ministry of Labor and Ministry of Education relating to the employment of higher-education graduates. Both studies reached the conclusion that graduates were paid $3,000 more than what was commonly thought. It seemed that these studies both showcased the openness of the government’s data as well as the their ability to stimulate jobs for young graduates.

However, these results were met with public cynicism. Released close to elections in January, the studies were not accompanied by a call for outside review from civil society, nor did they reflect the prevailing view that the Taiwanese had about the plight of graduates in the country. Despite the election being framed as a referendum on China/Taiwan relations, the issue of most concern to the Taiwanese was the slow economy. With stagnant wages and competition from regional rivals such as China, tales of government success in transparency fell on deaf ears.

The government provided the raw data and methodology that it used to reach these conclusions, but it did so without analysis or input from third parties, damaging any claim it made. Worse still, such verification was very possible to achieve given the open nature of Taiwanese media and data. Engagement and verification from civil society would have at least drawn attention to if not confirmed the government’s achievement, given Taiwan’s robust environment for public discourse.

Taiwan demonstrates that while a government can reveal statistics and even open the underlying data to make an argument, there also needs to be engagement from the media and civil society to provide legitimacy to that data in the form of independent verification. In Taiwan, the government put out data but reached questionable conclusions. This was exacerbated by the absence of confirmation from civil society and the media. In the end, this undermined the authority of the government when it came to those conclusions and reinforced a preconceived view that many Taiwanese had toward it. The timing of the release during a heated election campaign and the positive conclusions it reached about government performance provide a possible indication to why the government went down this route in releasing these studies in question.

The evolution of open data in Taiwan shows that better monitoring based on evidence could play a significant role in guiding how issues related to human rights are being addressed and how governments will be held accountable when citizens have the tools with which to hold them accountable. It also demonstrates that challenging governments when they engage in activities that manipulate data is a multifaceted exercise that can provide the impetus for governments to report accurately on the conditions in their country through national statistics.

Unreliable data presents a danger, reliable data an opportunity

National statistics agencies must be trusted sources for the kinds of data that is crucial for more effective monitoring of human rights. And not only should the data be trusted, but how governments reach their conclusions is equally important: Without accompanying open data with transparent methodologies — even absent clear reasons to suspect manipulation of national statistics — true legitimacy is impossible, especially when statistics differ from prevailing public sentiment. (The UN provided a useful guide on the importance of opening up methodologies and what national statistics offices should consider in order to ensure their quality.)

In Argentina, the manipulation of economic data showed the downside of what happens when governments undermine the veracity of the data they produce, which in turn undermines governance, accountability and human rights. Yet, as we have seen with Taiwan, when reliable data is combined with an environment that supports citizen engagement, verification is expected and in many ways necessary. Open data thus presents itself as an integral piece in the digital age of keeping governments honest in every aspect of what they do.

What these examples also show is that there is more work to be done in engaging/inviting verification from trusted third parties. It also highlights that open data can be a crucial step for getting there. Reliable statistics are a vital component to true transparency and good governance, and it takes the whole of public society — empowered by open data — to get there.