The ethical blind spots in SA's unemployment stats

THIS opinion piece responds to former Statistician-General Pali Lehohla’s article Debating the Labour Force Survey – A Response to Fourie’s Critique.

It serves as a rebuttal to his critique of my earlier article, Why Capitec’s CEO Is Forcing SA to Rethink Its Unemployment Narrative, in which I argued that South Africa’s unemployment figures fail to reflect the lived economic realities of the majority Black population.

Lehohla claims that my article has “amplified the debate” and insists on setting the record straight before it spirals into misinformation and speculation. However, my article did not reject StatsSA data outright. Instead, I argued that South Africa’s high unemployment statistics are shaped by a biopolitical statistical system that invisibly erases informal economic activity and Black labour. This is largely due to restrictive measurement methodologies and the active suppression of the informal sector, unlike in other developing countries.

I proposed the adoption of hybrid metrics and structural reforms to more accurately capture and support this vital, yet uncounted, segment of the economy. My stance aligns with UCT economist Haroon Bhorat, who engages constructively with Fourie’s arguments rather than dismissing them entirely.

Lehohla, however, dismisses Fourie’s estimate of a 10% unemployment rate — based on informal economic activity — as “abracadabra”, “lying”, and the rant of a “random businessman who profits from Black communities”. His anger masks a deeper crisis: South Africa’s economic measurement system, though methodologically sound, is philosophically ill-equipped to account for the informal, digital, and survivalist nature of the majority-Black workforce.

Lehohla defends StatsSA’s unemployment figures based on their adherence to International Labour Organisation (ILO) standards and the Quarterly Labour Force Survey (QLFS). Yet, I argue that this technical rigour obscures vast swaths of economic activity. For instance, a township hairdresser or street vendor without formal records becomes statistically invisible. This creates a profound ethical issue: stark racial disparities in unemployment, with Black South Africans facing an expanded unemployment rate of 40%, compared to just 7% for white South Africans.

South Africa’s high unemployment also stands out globally. Countries like Mexico (55% informal, 4.5% unemployment) and Nigeria (85% informal, 3.34% unemployment) include self-reported informal work in their statistics. In contrast, South Africa’s metrics function as biopolitical instruments that perpetuate apartheid-era exclusion by rendering Black economic agency statistically non-existent.

Bhorat notes that UCT’s Development Policy Research Unit (DPRU) consistently shows South Africa having one of the highest unemployment rates globally (33.6%), but also one of the lowest informality rates (about 16.3%). He highlights how most emerging economies address unemployment not by creating more formal jobs, but by allowing informal work to flourish. DPRU research further suggests that South Africa’s unusually high unemployment is not primarily due to poor job growth or strict labour laws, but because our economy actively suppresses the informal sector.

My advice to DPRU is not to shy away from confronting the moral failures or societal consequences that their data may obscure.

Lehohla’s refusal to engage meaningfully illustrates the difficulty of escaping the grip of orthodox economics and its limitations. Orthodox economics treats the economy — and by extension, social life — as a predictable machine operating in equilibrium. When official statistics diverge from lived experiences, the social contract built on citizens sharing data begins to erode, revealing a deep crisis within the discipline of economics.

Unlike Adam Smith — who grounded market value in ethics and social relations in The Theory of Moral Sentiments — modern economics has severed this moral root, prioritising abstract mathematical models over real-world complexity.

Joseph Stiglitz warns that GDP-centric metrics obscure true well-being. Persistent youth unemployment amidst trillions of rands in township transactions is not merely an error — it reflects a flawed measurement paradigm. Kenneth Boulding adds that modern economics builds on classical works like The Wealth of Nations and Das Kapital, which contain unrealised “evolutionary potential” absent in contemporary models. He cautions that excluding economic history from graduate education produces “idiots savant” — technically proficient economists who lack institutional understanding and historical insight.

A balanced synthesis of modern analytical tools and classical wisdom can help bridge this divide, fostering critical engagement with economics as both a technical and humanistic discipline.

Lehohla’s defence rests on rigid positivism — the belief in the “holy” authority of statistical processes — yet this glosses over the ethical roots of economic thought. For Smith, wealth was defined by the ability to command others’ labour — a social relationship, not a cold data point. Modern economics, however, has decoupled itself from these normative foundations.

As Stiglitz points out, most metrics conceal inequality and human suffering, reducing development to arithmetic rather than justice. This philosophical drift is evident in South Africa: while StatsSA reports rising unemployment, Capitec Bank documents over R2 trillion in township transactions — a vibrant economic reality invisible to official instruments.

This disconnect signals a deeper crisis in economics. Equilibrium models and optimisation problems eclipse historical nuance, cultural dynamics, and power relations. Boulding warned of this technocratic drift, describing modern economists as technicians fluent in calculus but blind to social texture.

In a direct response to me, Lehohla stated: “There is no legacy to protect on my part, Bhungane (my totem), nor language to polish. When a lie is told, there is no reason to give it a different word. It is simply a ‘lie,’ and when an argument does not make sense, it is called nonsense in the English language, and when nonsense is given wheels and wings to fly, it is called ‘rubbish.’ Those who wish to opine should do so from research rather than from a hailer.”

While I may not use his hyperbolic or confrontational language, I am neither uninformed nor inexperienced in public discourse. I have an academic and policy track record that makes me far more than “a hailer.” As many have rightly pointed out, shouting or using aggressive language does not strengthen an argument. We must allow space for multiple viewpoints to ensure inclusive policymaking around poverty, inequality, land reform, and unemployment.

Finally, Lehohla attributes South Africa’s unique unemployment situation to two key factors: agricultural activity tied to land ownership and high levels of economic concentration. He argues that these factors challenge simplistic international comparisons and emphasise the centrality of the land question in shaping employment outcomes.

No. Lehohla is deliberately conflating issues to obscure the fact that his revered unemployment metrics miss the ethical forest for the numerical trees.

Siyayibanga le economy!

* Siyabonga Hadebe is an independent commentator based in Geneva on socio-economic, political and global matters.

** The views expressed here do not reflect those of the Sunday Independent, Independent Media, or IOL.

Get the real story on the go: Follow the Sunday Independent on WhatsApp.