Dodgy data and missing measures: why good numbers matter (part I)

Earlier this year, an excellent short book, “Poor Numbers: How we are misled by African development statistics, and what to do about it” by Morten Jerven from Simon Fraser University in Canada was published (see this African Arguments piece for a summary). It makes the case that African statistics are often worse than useless, and decisions, rankings and other assessments made based on such poor numbers are usually grossly misleading. Jerven comments (page xi):

“…the numbers are poor. This is not just a matter of technical accuracy. The arbitrariness of the quantification process produces observations with very large errors and levels of uncertainty. This numbers game has taken on a dangerously misleading air of accuracy, and the resulting numbers are used to make critical decisions that allocate scarce resources. International development actors are making judgments based on erroneous statistics. Governments are not able to make informed decisions because existing data are too weak or the data they need do not exist”.

He argues that this appalling state of affairs came about through a long neglect of statistical services in Africa, made worse by the withdrawal of state support during the structural adjustment period. He focuses in on the iconic statistic, the gross domestic product (GDP), and a few countries, including Nigeria, Malawi, Zambia and Tanzania. GDP figures are made up of various elements, and in many countries in Africa, agricultural income is crucial. Yet, as Jerven shows for Malawi, there are all sorts of reasons not to believe the figures, as political incentives in particular result in distortions (in the case of Malawi massively upwards to ‘prove’ the ‘success’ of the politically driven fertiliser subsidy policy). Also, in much of Africa, the informal economy is massive, and very poorly understood. There are ways of assessing informal economic activity, such as through assessing expenditures, but understandings remain often very limited. The result is that in countries where the informal economy is significant (most of Africa), there are large under-estimates in national income.

The consequences of all this are severe, the book argues. Planning and budget allocations are carried out on the basis of flimsy evidence, distortions arise as statistics are influenced by political interests, successes much hailed may be far from such, and in the endless pursuit of targets (driven for example by the Millennium Development Goal process), indicators may be meaningless, or the data simply made up or guessed. The highly popular country rankings on everything from GDP to good governance – including the latest offering coming from IDS (where I work), the Hunger and Nutrition Commitment Index (HANCI) – thus create their own political economy. Informed by dodgy data and the even more dubious process of ‘expert judgement’, many rankings may be worthless. Dudley Seers (quoted by Jerven, p. 36), who went on to become the founding director of IDS, had this to say 60 years ago:

“In the hands of authorities, such international comparisons may yield correlations which throw light on the circumstances of economic progress, and they tell us something about relative inefficiencies and standards of living, but they are very widely abused. Do they not on the whole mislead more than they instruct, causing a net reduction in human knowledge?”

A key complaint Seers was the lack of attention to the ‘subsistence economy’. This he referred to as the “well-known morass which those estimating national income of underdeveloped areas either skirt, rush across or die in” (again quoted by Jerven, p. 37).

Yet such measures and rankings inform opinion, resource disbursement and provide competitive league tables to which governments respond, often exacerbating the poor numbers problem, as yet more dodgy data is conjured up, combined and ranked in ways that make little sense.

Zimbabwe is not covered by the book, but the core argument still holds, as I will explore further next week. The Central Statistics Office, now ZIMSTAT, has been the main source of government data since the colonial era. Compared to many countries, it has impressive capacity and a very strong track record. One thing that could be said of the colonial and Rhodesian authorities is that they were very keen on data. From the Rhodesian Yearbooks to the regular national income and expenditure surveys, data was collected, collated and compiled rigorously and consistently.

Statistics are after all about measurement and control – they are the very essence of the state, as the term suggests. In his brilliant history of statistics, The Taming of Chance, Ian Hacking relates how states were developed alongside statistical services, including cadastral surveys, taxation systems and population counts. In Jim Scott’s terms the ordered, controlling and regulated way of ‘seeing like a state’, is very much wrapped up in counting, surveying and so being able to control, through a form of Foucauldian governmentality at the core of modern states.

While there are clearly negative aspects to this form of state capacity, there are also positive attributes. A committed developmental state cannot allocate funds, direct energies and plan for the future without a good statistical base. Negotiations with donors, steering of investments and prioritisation of expenditures are impossible. Equally, without solid data, political biases, bureaucratic whims and donor influence can overtake planning and budgeting to the detriment of developmental objectives.

Jerven concludes on the state of African statistics: “…the data are based on educated guesses, competing observations, and debateable assumptions, leaving both trends and levels open to question and the final estimates malleable (p. 108)… He continues: “Decisions about what to measure, who to count, and by whose authority the final number is selected do matter” (p.121). Which is why he recommends the revitalisation of African statistical services and, perhaps just as importantly, the improvement of capacity to interrogate and interpret data, including from qualitative insights.

Next week, I will turn to the implications for Zimbabwe more specifically.

This post was written by Ian Scoones and originally appeared on Zimbabweland


Filed under Uncategorized

9 responses to “Dodgy data and missing measures: why good numbers matter (part I)

  1. ABMacLean

    Whatever is true in other countries in Zimbabwe each headman – village level – is a collector and maintainer of meaningful statistics e.g. houses, people. His book is a great statistical source. It is also known who has what land – arable fields. Then there is the stock book per house – a very serious document treated with great responsibility. It would be worth investigating the movement of the statistics in two directions e.g to the chief and to the government and determining which is more accurate.
    There is less dodgy data at the headman level and chief level, probably. It is at least more up to date.

  2. Pingback: Food crisis in Zimbabwe: 2.2 million a risk. But where do the figures come from, and what do they mean? | zimbabweland

  3. Pingback: Food crisis in Zimbabwe - where do the figures come from, and what do they mean?

  4. Com a idade que eu tenho ainda não ouve nelhuma estatistica que divulga-se os numeros verdadeiros A estatistica somos nós que temos que ir há procura da verdade Se existe chefe dentro das aldeias são é que t´^em que procurar a verdade e também são os responsaveis pela segurança alimentar alimentar dentro da sua aldeia

  5. Pingback: Food crisis in Zimbabwe: 2.2 million at risk. But where do the figures come from, and what do they mean? | Zimbabwe Investor

  6. Pingback: Zimbabwe’s gold rush: livelihoods for the poor or a patronage economy, or both? | zimbabweland

  7. Pingback: Spurious statistics: why figures on Zimbabwe’s ‘lost growth’ mislead | zimbabweland

  8. Pingback: Drought politics in Southern Africa - Uncle Sam's Blog

  9. Pingback: Drought politics in Southern Africa | The Povertist

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s