Mapping the transparency of a pandemic

Alina Mungiu-Pippidi and her students explore governments’ reponses to Covid-19.

Students in the spring 2020 class “Corruption and transparency mapping”, taught by Hertie School Professor of Democracy Studies Alina Mungiu-Pippidi, wanted to find out how governments are performing during today’s pandemic. The course focuses on rating transparency across countries. Participants decided to dedicate one session to surveying government information policy on the coronavirus in 16 countries around the world. Countries covered were Germany, France, UK, US, Italy, Ireland, Romania, Ecuador, Brazil, Colombia, Mexico, Tunisia, Georgia, Chile, South Korea and China.  

Q: What inspired the class to evaluate the current information on the pandemic?

The news is full of numbers about the Covid-19 pandemic, but what does it all mean? Information such as the US surpassing Italy’s high number of deaths, or global deaths breaching a key “threshold” of a million, have little meaning without context. But when we learn that 300,000 South Africans (about 5% of the population) perished from Spanish flu over six weeks in fall 1918, the high mortality rate in a short time strikes us as significant. When we hear that the epidemic killed more people worldwide in 12 months than four years of war, it’s easier grasp the magnitude of the toll.

The trouble is, these figures come from historians. Understanding what’s happening in real time is a different proposition. During the Spanish flu, it wasn’t in the interest of governments fighting in World War I to publicize the spreading disease, nor the fact that American troops who came to the rescue in Europe most likely brought it with them. It came to be known as the “Spanish” flu because journalists in neutral Spain, where the press was not subordinated to war propaganda, covered it earlier and more extensively. 

The 21st century has brought an explosion of accessible communication technology, which powers government transparency as never before. As democratic governments lost their monopoly on communication, they needed to be more accurate and straightforward in sharing information to earn public trust. The obligation to tell the truth, especially about events that might end in human casualties, is no longer a questionable norm, although few constitutions mention it. People also no longer assume the government knows better than citizens what is good for everyone. Governments must now broadly share evidence supporting their decisions. This brought the class to the question of how governments are performing in today’s pandemic crisis.

Q: How did the students go about their evaluation?

The students evaluated whether governments were doing a good job in communicating, and if public information was organised to help people understand the causes, spread and risks of the pandemic, as well as possible solutions. They rated governments’ main websites using five criteria: 

  • Accessibility: how easy is it to find? Did it cover the main languages in a country, with some summary in English for non-native speakers? 
  • Comprehensiveness: is the information comprehensive? Is some important information or issue missing or misrepresented? 
  • Organisation of information: is it organised in a clear and logical way to facilitate fast understanding? 
  • Interactivity: does it have a Q&A section?  Does it include an interactive component (like a call number?) Does if facilitate sharing through social media? 
  • Actionability: does it make clear what everyone should do? 

The class also first rated their own expectations about the behaviour of these governments on a 1 to 5 scale, and then checked these expectations later against their transparency results. They found the results were mixed, and to some extent quite surprising.

Q: What sorts of things did they find out?

Although many governments clearly were trying hard, students found that no single government could beat the private website “Worldometer” in terms of comprehensiveness and organization of information. The site uses “web-scraping” to collect data from all over the internet, and it tops global Google search lists as a sought-after site. Its tables are missing no essential information, including the number of tests performed – data that students found many governments failed to provide readily.

The website of Germany’s Robert Koch Institute – not the government’s main site, but the institution officially tasked with handling the virus – was found to be comprehensive but technologically unsophisticated– using pdfs, for example, instead of interactive dashboards. RKI publishes a daily report in English and German (pdfs) communicating, among other things, prevalence (number of cases per population, as well as in absolute figures), incidence (new cases), deaths (absolute numbers, as well as mortality, % of total cases) and reproduction rate (how many people one person with the virus infects on average), which has been Germany’s chief policy indicator. However, the report doesn’t list daily the tests performed, which could serve as a rough confidence error: in a country with little testing, most statistics might be misleading, but Germany has tested a lot, at least compared to other European countries. Worldometer remains the best performer aligning all indicators per capita so to allow comparison. Although the number of tests is very important, many governments were found not to clearly report on them in the first weeks of the pandemics. China, for example, has not disclosed the total number of tests it has performed. In countries with a free media this had become one of the main tools used by journalists to assess the performance of governments.

Nevertheless, the students found that China’s communication to the public looks quite modern and comprehensive, compared to initial perceptions that the government was not being transparent. But can one trust the figures? According to news reports, Wuhan authorities first ignored the warnings of a whistleblower doctor who subsequently died of the virus. The mortality rate in Wuhan – a key indicator for those outside the virus’ epicenter – at first appeared low, but the government then suddenly doubled the number of deaths reported. 

Many governments the students checked did not specify if deaths reported were just those in hospitals versus or if they were total coronavirus deaths. A report by the Corriere della Sera at the end of April compared total deaths a year ago with deaths this year, showing that large numbers might have been missed by official statistics: from the countries surveyed, UK tops the list, with nearly double the deaths compared to a year ago. Germany is missing from this report, but Italy and France both chose to report just hospital deaths in daily reports. Comparing deaths in 2020 versus 2019 from respiratory infections – when data will become available for all countries – might thus provide a test of transparency in the future.

While all the countries evaluated provided accessible and actionable general information, students found other transparency issues. The Italian government suspended its Freedom of Information law, arguing that the administration was too busy during the epidemic to answer information requests. However, one Hertie School student wrote to ANAC, the Italian anticorruption agency, and received an answer the same day. The Romanian government issued an order to prefects not to release to journalists directly the number of deaths and other county statistics. Romania notified the Council of Europe on the suspension of certain rights under the European Convention on Human Rights, including transparency and freedom of the press, due to the military state of emergency. Hungary and Romania closed private websites accused of misinformation, which in the Hungarian case led to a reprimand from the European Commission for infringing freedom of speech.

Nevertheless, students found a number of countries to be very transparent – specifically, Germany, the UK, Ireland, South Korea and Georgia. Some countries, namely Brazil, Mexico and Tunisia, where corruption has been an issue, performed better than expected. Brazil, the US and the UK had excellent websites, but communications from their heads of states also diverged from some official government communications, showing that top elected politicians and government experts were not always aligned, created blurred messages. 

In their survey, the students also encountered many innovations. Some governments, like France and Colombia, used cartoons to display information and reach audiences. Ecuador created a search engine using taxpayer identification numbers where relatives could find out if family members had been taken to hospital were deceased, and where the body was. In Tunisia, hospitals’ needs are aggregated on a webpage where direct donations can be made. Many governments made efforts to dispel fake news and created special sections to address fact and fiction about the virus. The class also did not find any government website acknowledging that official messages had changed seriously, in particular on controversial topics like the usefulness of masks in Europe or comparisons of the coronavirus with the flu in US.