Still Poor Numbers?

Many public decisions and plans rely on essential information such as GDP (composition and growth rates), unemployment rates, inflation, productivity, and income. To create an accurate picture of these figures, countries have official statistics offices that collect, assess and make sense of the numbers. Most people trust these numbers without hesitation. Although there are examples of politicians influencing economic statistics, it is not widespread.

So why am I talking about it? Because Nigeria recently updated its method of calculating the unemployment rate and it made me raise an eyebrow. The good folks at Stears have an excellent explainer (worth reading the whole thing) on the nuts and bolts but here is the short gist:

‘’Employment in Nigeria is now defined as anyone working at least an hour a week, unlike the old methodology where you had to work at least 20 hours a week to be considered employed. This means anyone who works below an hour a week is unemployed. This is called time-based employment.

It’s worth noting that there is no definite way to measure time-based employment. Is working 40 hours a week the best to capture an individual’s productivity? Should it be 20 hours or even 1 hour? We know this measure is not an accurate measure of labour productivity. For instance, there is no way to tell if the supermarket owner working 40+ hours weekly (9 am — 9 pm daily) contributes more than the social media influencer who works 20+ hours weekly (10 am — 2 pm daily).

A positive, though, is that, as I mentioned before, this (one hour a week) increases the working population, that aggregate labour input now corresponds with national output (nominal GDP). With this coherence, estimating the share of labour (an input) to GDP will be easier. For instance, if labour as a share of GDP declines while GDP rises, it tells us that economic improvements do not translate to higher incomes.

But there are glaring limitations.

One is that the methodology change will present significantly lower unemployment numbers for Nigeria than the 33% (34 million unemployed) recorded in Q4’2020. ‘’

I understand the real big changes here, but the fact that Nigeria’s unemployment rate will now be reported as roughly 6% from 33% in 2020 makes no sense to me. In 2012, economic historian Morten Jerven wrote a great book in which he expressed some very serious concerns about most of the development statistics being computed in Africa. In an essay for The Guardian in 2012, he summarised his worries, especially about the GDP rebasing fever that was the rage then:

‘’Meanwhile, in Nigeria an upward revision is pending. Their base year for the national accounts, 1990, is even more outdated than that of Ghana. According to reports from the National Bureau of Statistics (NBS), Nigeria plans to change its base year to 2008. It has been boldly announced that this could lead to a “huge jump” in GDP figures.

This radically challenges our current understanding of economic development in Nigeria and in Africa. According to the World Development Indicators’ most recent data, the total GDP in 2010 was above $200bn (in current US$). Nigerian GDP, before the predicted revision, already accounts for 18% of sub-Saharan Africa’s total (about $1,200bn). The reports in the media, from the IMF and the NBS all indicate that Nigeria’s GDP will increase at least as much as it did in Ghana.

Let us be conservative and assume that the GDP in Nigeria merely doubles following the revision. This alone will mean that the GDP for the whole region increases by more than 15%. The value of the increase amounts to nothing less than 40 economies roughly the size of Malawi’s. The knowledge that currently there are 40 “Malawis” unaccounted for in the Nigerian economy should raise a few eyebrows.’’

My eyebrows were also raised a few months ago when the National Bureau of Statistics (NBS) published the Multidimensional Poverty Index report, which boldly claimed that ‘’approximately 70% of Nigeria’s population live in rural areas’’. Meanwhile data from the World Bank says that the percentage of Nigeria’s total population living in urban areas already reached 53% in 2021.

So, what is going on here? The short answer is that I do not know. Perhaps there is something I am missing and smarter people can explain it to me. But an even bigger fear I have is that more than a decade after Morten Jerven raised alarm about the issue, economic statistics in Africa (and Nigeria) still have a credibility issue. The common retort is that some of these statistical revisions bring us in line with ‘’global methodology’’. But there is plenty of room for errors in how data is collected, sampled, and analysed. Nigeria has not had a census in almost two decades, and some will even say we have not had a non-political census for decades before that.

The second concern I have is that what Lant Pritchett called ‘’defining development down’’ might also be happening to economic statistics. In the bid to be ‘’inclusive’’ or set some less ambitious targets in national development, the urge to revise statistics in ways that meet these targets can be quite tempting. I find it absurd that a 60-something-year-old person can be considered active in the labour market enough to be ‘’underemployed’’.

Even if that is the reality in many countries, the fact that we might be refining economic statistics to put a technical gloss on such a tragic state of nature is something I find truly disturbing.