In a presentation I titled “political statistics and statistical politics” I could not envisage the world of official statistics encountering the lows of more recent times. Perhaps I naively underestimated the toxicity of politics’ proximity to statistics despite the safeguards and experience in imposing such defences. Perhaps it was the civil disposition I enjoyed during my tenure, while not compromising on the facts and what they meant nationally and globally as it will soon be obvious. But it was not to be.
El Salvador put its foot in it more recently. I first learnt of the impendingsubjugation of the El Salvador Statistical Institute under the central bank of El Salvador a few weeks back. A no-go area in any book of an official statistics practitioner. A chill ran down my spine in a moment of deja vu, eight years after the UN Fundamental Principles of Official Statistics were passed by the UN General Assembly in 2014 as the basic international law for statistical practice for members of the UN.
Knowing the former head of ELSTAT, the Greek Statistics office, still faces prosecution a decade on, one knows statistics and politics are never a happy match. Andreas Georgiou, an MIT graduate who was recruited back to his home country to fix Greece’s national accounts, found himself enemy number one to the Greek people, a cross he carries to this day as the vexatious litigants are on his tail. This includes accusations of simple slander. This is described as what the defendant said is true, but that truth is demeaning to the plaintiff. Georgiou said his predecessor and the subordinates at ELSTAT had provided a false government debt statistic and he undertook a professional surgery to reach that conclusion and corrected it. The Greek government would have none of it, yet they also clamoured for accession to the EU. The only key to that, is accurate and transparent statistics that Eurostat have to validate and audit. It was Eurostat director-general Walter Rademacher who would wag his finger with displeasure at the Greek statistics, so would the IMF.
I once thought I should forget about ever having cheese, steak and lamb from Argentina because I was on a statistical warpath with them.
Statistics offices have never been this low. We have seen the worst where Stalin executed his Soviet statistician for returning numbers of a population that were lower than Stalin’s expectations. In recent times the Montenegro statistician was fired for delaying to undertake the census because of Covid-19, which disrupted census preparations. Former US president Donald Trump was eager to receive the count of the population before handing over power, but because of Covid-19 the US Bureau of the Census had pointed out the numbers would be released beyond the statutory date.
The statistical conundrums are hard to numerate. I once thought I should forget about ever having cheese, steak and lamb from Argentina because I was on a statistical warpath with them. In 2012, I was in New York at the UN Statistical Commission, where I was a member of the panel on violations of Fundamental Principles of Official Statistics. In the presentation I found against myself for violations and this was on the CPI debacle of 2003. There were four other countries I found against, which included Canada on its census of 2011, Argentina on CPI, and Indonesia on the confidentiality of records. Argentina did not take kindly to my statements, and the Argentinian ambassador to the permanent mission of the UN accosted the SA ambassador on the matter. So on that fateful wintry day in New York, ambassador Basso Sanqu summoned me to his office at 5.30pm. As I stepped in his office he gave me the note from the Argentinian ambassador. The note asked whether the SA government gave me the mandate to present about the Argentinians. I said to ambassador Sanqu, first, I reported and presented myself to the mission upon my arrival here in New York, which means I represent my country. Second, the Ten UN Fundamental Principles of Official Statistics, including the Statistics Act of SA, empower my position to pronounceon abuses of statistics. That practice enjoins me with my counterparts through the UN Statistical Commission to act on standards and methods.
In this regard principle 4, on pronouncing on abuses, principle 8, on international co-ordination, principle 9, on international concepts and classifications, and principle 10, on bilateral and multilateral co-operation, dress my professional capacity with powers to pronounce myself on matters of statistical practice everywhere.
Ambassador Sanqu thanked me, and I hopped into a yellow cab back to my hotel. Little did I know the following day the director of the UN Statistics Division Paul Cheung would be recovering from a visit he got from the Argentinian ambassador. He said, my friend, you put me in trouble, the ambassador was in my office this morning complaining about your presentation. I told Paul to ask the ambassador to talk to me. He was in the presentation and should have raised his concerns, instead he called me “mastido mastido” as his entourage marched out waving and hurling insults.
The simple but terrible facts are that Argentinian finance minister Guillermo Moreno had removed the head of consumer price index (CPI), and the government started manufacturing their own CPI. Graciala Bavacqua became the Andreas Georgiou of Argentina. The government hurriedly promulgated a law that barred any production and publishing of CPI by anyone other than the government-manufactured CPI. The sentences and fines were draconian. Bavacqua continued to publish her CPI and was threatened with prison. More damaging about Argentinian numbers was the prospect of including these numbers of Argentina in the International Comparison Program (ICP) statistics. I was a board member of the World Bank-led ICP — a discussion on its own. Back to El Salvador, the International Statistics Institute (ISI) and the International Association for Official Statistics (IAOS) have petitioned the decision by the government of El Salvador to dissolve the Dirección General de Estadística y Censos (DIGESTYC) and to transfer the functions of the National Statistical Office to the Salvadorean central bank. The law for the dissolution and transfer of DIGESTYC was introduced end of July 2022 by the government and has already been adopted by parliament.
As in the past with Greece and Argentina the ISI, where I served as the vice-president, and IAOS expressed their grave concern on the practices there. The ISI and IAOS have also done so with El Salvador. Let me illustrate how grave the issue could be in a SA context. Prof Charles Meth, an economist of then University of Natal, penned a scathing critique of the moral hazard of shared responsibility over components of the national account compilation between Stats SA and the SA Reserve Bank. Meth had been on this matter for more than two decades from the times of Dr Trienacht du Toit, who was the head of the Central Statistical Services (CSS) in the late 1980s up to 1995. The moral hazard was captured when Helmut Spinner of Karlsruhe Institute addressed the statisticians of the Economic Commission for Europe. In his rendition on the topic of The Changing Order of Knowledge in Information Societies, he argues for the need for separation between policy institutions and producers of ideas and knowledge. In SA the SARB was a player and a referee. It was an interested party in the statistics and policy decisions it would take. So there was a need to ensure the SARB is a recipient of the national accounts figures rather than being a producer and a market player. By 2010, I decided to take the bull by the horns. It required arduous planning to take the expenditure side of the national account from the bank.
First Dr Rashad Cassim, former head of economics statistics at Stats SA, had become head of research at the SARB, which created a good condition for the move. However, the bank was not going to give us their staff to shoot themselves in the foot. From my experience of bringing in new blood into Stats SA through the census, it was clear the strategy would be one through which we train and recruit.
I called the team the marines. The retired deputy chief statistician from Canada, Dr Jacob Ryten, would provide the training. The key was to select through the internship programme. By 2013 the interns for the takeover were enrolled. A task team across the SARB and Stats SA was formed and governor Lesetja Kganyago and I were sponsors. The youthful Michael Manamela, who sadly joined the SARBb eight months ago, led the enthusiastic and youthful team of marines. Three years later in 2017 the expenditure side of the national accounts resided in Stats SA with full capacity, reduced residual and customer delight. The residual was always a pain between production and expenditure sides of the GDP, and represented a threat to Stats SA as a SARB-inspired moral hazard. Now imagine a whole Stats SA swallowed by the SARB. This is what El Salvador has just done, against all sensibilities of global practices in official statistics.
In SA I doubt this can happen. The two institutions are strong. But before I forget my worries about being barred from Argentinian cheese, steak and lamb, I was pleasantly surprised in 2017, when I received a letter through the international relations ministry in SArom Argentina commending me for staying the course of crying for Argentina in all its wild days of grave existence. El Salvador and all other violators of official statistics can also make amends, but it all needs gravitas and a political environment that is not toxic — a rather rare commodity in SA of the past 15 or so years. Time for me to head to Argentina for my cheese, steak and lamb and take a break from the battle of “political statistics and statistical politics”.
Dr Pali Lehohla is the director of the Economic Modelling Academy, a professor of practice at the University of Johannesburg, a research associate at Oxford University, a board member of Institute for Economic Justice at Wits and a distinguished Alumni of the University of Ghana. He is the former Statistician-General of SA
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.