With International Human Rights Day coming up on Sunday, I’ve been thinking a lot about a Greek economist named Andreas Georgiou. In 2010, Georgiou was living in Maryland and working for the International Monetary Fund when he saw a call for applications for a job in Greece heading a new statistical agency. At that point, Greek financial and debt statistics were a joke in Europe. The European Union has a system where countries compile statistical information and submit it to EU-level bodies for review. These bodies had frequently expressed concerns about the figures Greece submitted, refusing to validate or certify the numbers. So, under pressure, Greece launched a new independent national statistical agency, and undertook revisions. Georgiou was hired to lead the effort. He took the mandate seriously, recalculating debt figures according to European and national guidelines, and showed that the numbers Greece had been reporting were far rosier than reality. This was news the country didn’t want to hear. The reality Georgiou helped bring to light led to the austerity measures than many Greeks feel have wrecked the country.
Soon, Georgiou became a scapegoat. Politicians argued that it was Georgiou himself who caused the crisis. He’s now fighting criminal charges in Greece for doing the job he was hired to do. Some of the crimes he’s being prosecuted for make no sense from an American perspective. For example, apparently simple slander involves saying something that’s true but unpleasant. More serious charges include trials for complicity against the state. As Greece appears to have no concerns about double jeopardy; prosecutors have opened second and third trials when their initial efforts led to not guilty verdicts.
Georgiou’s case is a tragedy in itself. It is also a cautionary tale. National statistics are the data that describes a country. In a functioning democracy, these numbers are produced by national statistical agencies that are distinct and independent from the parts of the government that use the data to set policy. This independence is enshrined in Principles and Practices of a Federal Statistical Agency, a wonky but important report from the National Academies, and similar international documents.
Independence is crucial to ensuring that numbers aren’t tweaked to support policy goals. It can also become a target of governments that would rather produce numbers to support their decisions than base their decisions on measured data.
I learned of Andreas Georgiou’s case through my work with the American Statistical Association Committee on Scientific Freedom and Human Rights, which has written several letters to the Greek government protesting the prosecutions. (So have many international statistical organizations.) Before I joined the committee, it similarly tracked the case of Graciela Bevacqua, in Argentina. The former head of the National Institute of Statistics and Census (INDEC) division responsible for compiling consumer price indices (CPI) in the country, Bevacqua also faced charges for doing her job. Thinking about all this reminded of a conversation I had years ago with a researcher who worked with Iraq’s national statistical office to conduct a survey after the US invasion. He told me that one of the challenges he faced was getting staff to try. They’d become so used to Saddam’s government reporting whatever numbers the regime wanted that they had grown wearing of trying to generate solid statistics.
Once, I would have thought of these as issues to worry about in other countries. However, last January, in Science, Jeffrey Mervis documented concerns from the scientific community that the Trump administration would attempt to sabotage the 2020 Census and other Bureau data collections activities. Many of the concerns Mervis listed appear to be being born out, and in recent months, writers at organizations as diverse as Rolling Stone and the Brookings Institution have expressed similar worries. A failed Census would matter for a lot of reasons, the most obvious of which is that the Census is used to set districts for the House of Representatives. And if the Census moves forward in a crippled fashion, that would result in undercounting of groups such as immigrants and minorities. That is, the kind of people who are unlikely to vote Republican.
Beyond redistricting, Census data is used by the government and researchers for a host of purposes. Data from the Census Bureau have been used to study everything from African American grandparents raising grandchildren to estimates of the size of the gay and lesbian population to changes in areas with concentrated poverty, housing affordability, and many other important issues.
Of course critics are not saying they are targeting the collection of data on disadvantaged groups; they are targeting the Census’ budget. Collecting and crunching national statistical data is not easy, nor is it cheap. However, this data is core to understanding the true nature of a country as large and diverse as the United States, to supporting our representative system of government, and to making progress on just about every social issue that matters in a vibrant democracy.
Robin Mejia is a statistician and writer who manages the Statistics and Human Rights Program at the Center for Human Rights Science at Carnegie Mellon University.