The movie “The Big Short”, based on the book of the same name by Michael Lewis published in 2010, tells the story how the housing market collapsed because the banks sold shit based on bad numbers. Just a few people cared to crush the numbers to see that the housing market was built on said shit:
Mark: “Mortgage bonds are dog shit. CDOs are dog shit wrapped in cat shit?
Jared: “Yeah, that is right”
The premise of the movie is, that no one cared about the trustworthiness of the numbers for several reasons, which I will return to later.
This blog is about Africa, and the subject regarding numbers are bad if not shit is relevant here as well. A lot of numbers are floating around concerning Africa. A lot of numbers are created right now.
I want to ask, if much of the information about Africa, in reality, is dog shit wrapped in cat shit, later to be served as trustworthy to the audience by credible agencies?
If yes, is information about Africa a shit bubble waiting to burst?
This is important! Numbers matter because they are our umbilical cord. Without credible numbers, we are screwed. These numbers are used by NGOs, the UN, companies, governments, and you and I. Numbers are the bedrock of our knowledge about African and each African country. And about your own country.
If the numbers about Africa are wrong, the Africa we are presented does not exist, and policy makers, journalists, market analysts, everybody, end up producing policies and strategies based on shit. Such policies and strategies will fail.
Trustworthiness of the World Bank?
Metadata is the data in how the numbers came to be. How many people were asked, when where they asked, how were they asked, who were these people, and how were they located? Did they use quantitative or qualitative research? Or both?
It defines definitions. If you said, that you interviewed the youth, what is the definition of youth in this context?
See it as an ingredient list. When you buy a food product, you like to know what is inside. If the company did not include an ingredient list, you have every reason to doubt the content of the product, you are about to purchase. The same is the case for numbers.
It is the metadata that indicates if the numbers used are reliable or unreliable.
That is the core problem, for long periods of time, we do not have any metadata for several African countries. But we still have numbers. How is that possible?
In the movie, the credit rating agencies gave triple AAA ratings to loans that actually belonged in the trash. In Africa, the World Bank and other institutions give data credibility. But are these numbers credible?
Credibility of the World Bank and the International Monetary Fund (IMF)
The World Bank and the IMF get their numbers from the African statistician officers and their own people. According to economist Morten Jerven, the officers in the most African countries are:
5) Not skilled due to the above-mentioned problems.
These institutions know this, but they forget to tell the readers that when reports are released.
Jerven especially criticises the per diem allowances. It provides the statisticians with a reason to stay in the field collecting data, but there is little incentive to actually sit at the desk to analyse the collected data.
He directly states, that data from Chad and Malawi are virtually non-existent, but the institutions do not even add a footnote telling the readers about this fact (Jerven 2013:101).
To make matters worse, to be able to measure a reliable growth and hence the size of a country’s GDP, you need a base year. The recommendation is, that the base year is revised every 5th year. Some African countries have not changed their base year for 20 years or more. Some countries do not even have a base year. Meaning, the majority of African countries’ GDPs are nothing more than a guess.
To make matters worse, several base years are from the 1980s when African countries were severely affected by the economic crises. The growth rates at that time were unusually slow and a given growth rate is not representative of the country in question in how it normally performs. The consequence is that several African countries’ GDPs are significantly higher in reality than shown on paper. It is even higher if we include remittances, which are three times greater than aid given by foreign donors.
When Ghana revised its base year from 1993 to 2006 its economy grew by +60%. When Nigeria did the same in 2012, its economy became Africa’s biggest overnight. After 2012, the new estimate said that Nigeria’s economy grew by the size equivalent of 52 times Malawi-sized economies (Jerven 2015:105).
Imaginary Growth Rates?
Another result is when rebasing occurs infrequently and growth appears, then we do not really know anything else than the economy was bigger than expected. We do not know when the growth took place.
It is like you step on the scale. You went on the scale in the year 2000 and your weight was 120 kg. You need to lose weight. In 2017, you step on the scale again and this time your weight in 70 kg. You have lost 50 kg. That is great news. But we only know that you have lost weight somewhere in between the year 2000 and the year 2017. In theory, you could have gained weight the last year. A man then walks in asking you to accumulate your weight loss retroactively to have happened within the last few weeks. Suddenly, it appears you have had an amazing weight loss in the recent past.
But we both know that is not what happened, but that is exactly what the IMF asked Nigeria to do.
The IMF asked the Nigerian government to apply the growth retroactively, to be sliced out to the latest previous years. When it was done, it appeared Nigeria had experienced a rapid and recent growth. The reality is, that we do not know. We know Nigeria has experienced growth, but not when the growth occurred. This is done elsewhere as well. Meaning, the so-called African lion economies might be a statistical illusion.
The variations in base years, growth rates, and more means, that when various institutions rate African countries on how poor or rich these countries are measured by each country’s GDP, these institutions do not agree. Jerven went through Maddison, WDI (World Bank), and the Penn World Tables.
Guinea is ranked as the 7th, 31st, and 35th poorest country in Africa dependently which source you use.
Liberia is ranked as the 22nd, 7th, and 2nd poorest country in Africa.
Because the numbers are so bad, the above-mentioned institutions derive at different results. Due to limited access to the metadata, the readers have no idea which data set is the most trustworthy -if any.
So the numbers are flawed if not at times useless. The worst example Jerven mentions is the case of the Food and Agriculture Organization of the United Nations (FAO). Even FAO provides data on crops in West Africa, the reality is, we have absolutely no idea how much food is produced in West Africa(Jerven 2013:78).
General Problems in Data Collection Resulting in Poor Number
Bias naturally occurs, when data is collected, but people using the data seem unaware of this fact.
Eg. bias is present when we measure corruption. Transparency International knows this, why it is called the Corruption Perception Index. People’s perceptions are flawed. If you are told that this football team is good and the other team sucks, you are more likely to overstate the greatness of the team, you believe is good, and understate the performance of the other team.
It is called confirmation and memory bias. We like to portray the world as we are told.
When you are asked about the level of corruption in Denmark and Nigeria, you are more likely to underestimate the level of corruption in Denmark and overestimate the level of corruption in Nigeria. This is the case because, you have been told, that Denmark has a low level of corruption and Nigeria is extremely corrupt. Hence, we should stay sceptical about corruption ratings. They are inherently flawed.
And this is just the tip of the proverbial iceberg.
When measuring GDP, women’s work is forgotten. That is especially worrisome in informal economies. If you hire a servant to clean your house, cook your food and so on, that work is included in the GDP. If you marry your servant, and the person does the same job, but now it is no longer credited as work. Consequently, work done my women are excluded and women become invisible. Data excluding up to 50% of the population should be viewed with much more caution than what we are witnessing.
What Does It Mean That Numbers Are Bad?
That current implication is that the Sustainable Development Goals (SDGs) stands on shaky grounds. How do we know the data collected to check if the goals are fulfilled are carried out properly? Furthermore, the SDGs undermine the statistical work. Instead, to fund better numbers, more is done to overburden the personnel. It is the wrong path.
For now, the data available is a mixture of educated guesswork and blind assumptions. At times, the data consists of even extrapolated data. Extrapolated data can roughly be compared to CDOs or as “dog shit covered in cat shit” to paraphrase the movie. Based on limited data from a few African countries, suddenly scholars make universal claims. You cannot do that! Because you have data from Ghana, that does not mean we know anything about Lesotho.
You cannot do that! Because you have data from Ghana, that does not mean we know anything about Lesotho.
Because you have data from Ghana, that does not mean we know anything about Lesotho.
When the data collecting is carried out by local statisticians, what is not counted? If you have to use your limited funding to investigate issues relevant to donors, like the SDGs, what areas are forgotten? What is not counted? It might not be relevant to donors, but it could be relevant for the local government or the people.
This is not a critic of the statisticians, they work the best they can and they are doing amazing considering the limited tools and funding they have. They need proper funding, proper personnel, proper tools, and proper respect.
They need proper funding, proper personnel, proper tools, and proper respect.
Some Africans countries are trying to improve this sector, like Uganda. Also, the new Ghanaian government has announced that they are going to reform the statical database. African governments are not enemies, they have an interest in having access to trustworthy numbers.
However, sometimes it is better to be aware, that you do not have the needed data than to act based on wrong or no data believed to be true.
Cultural Analyses – Hofstede
Another set of data are the ones from cultural studies. Scholars trying to provide the readers with an overview how Africans are. This is based on various cultural indicators. Sometimes, these cultural analyses are done by people, who clearly have no idea how pluralistic and large several African countries are. The data they produce is clearly extrapolated and based on limited sample sizes (
Sometimes, these cultural analyses are done by people, who clearly have no idea how pluralistic and large several African countries are. The data they produce is clearly extrapolated and based on limited sample sizes (how to lie with statistics). A more popular scholar in this field is Geert Hofstede. His cultural analyses on African countries are, to be frank bullshit.
First of all, because the numbers are weak. His sample sizes are ridiculously small, and his methodology severely criticised. The samples are not representative, and he does not even have samples from several African countries, but he still produces data. That is problematic.
The consequences are unbelievably laughable if it was not so tragic.
Nigeria: First of all, he sees Nigerians as a monolithic group. In reality, Nigeria consists of 200 million people divided into various ethnic groups. Nigeria is more diverse than Western Europe. That is completely forgotten, or more likely, Hofstede doesn’t know. Then he points out, that Nigerians do not like to work. Simply false. The level of inequality in Nigeria is huge, and for people to survive, they are willing to work 16 hours straight under the burning sun. They are a hardworking people. Most of them. We all have that lazy nephew. But that is the case everywhere. Our lazy nephew is the exception, not the rule.
He also states that Nigerians do not have a long time orientation. Also false. It takes practice to become good at something. The Nigerian national football team, called The Super Eagles, won the African Cup in 2013. That requires long time orientation. To be good at football, dedicated commitment is required. We also have Nigerians trying to seek greener pastures in other African countries and/or in Europe. Why? To find work, make a saving, and to start a family. That is long time orientation per excellence. These cultural traits presented by Hofstede reflects how the researcher views Nigeria, but it does not reflect how Nigeria and Nigerians are.
Angola: According to Hofstede, Angola is a place, where “conflicts are resolved by compromise and negotiation.”
The reality is, that the country is ruled by the dictator, José Eduardo dos Santos, who through a civil war fought his enemies ruthlessly. Today, Angola is one of the most corrupt countries in Africa, where dos Santos and a handful of people control everything.
If someone applies Hofstede’s cultural dimensions in order to explain an African country, assume bullshit.
A Shit Bubble?
It appears the information about African countries is severely flawed bordering garbage and at times actually turns into actual shit. And there is a bubble.
Programs are still using the African dummy, that does not exist. People still use these poor numbers, despite the fact the providers of the numbers know that the provided numbers often are nothing but guesswork. The problem is, that we have several people benefitting from the poor quality of numbers.
Western countries and Institutions can use these numbers in the need for SDGs, and there is a risk to lose credibility and face if the providers turn out to be wrong. If the foundation is wrong, then the rest of the projects funded by donors stand at best on shaky grounds.
We shall neither be blind to the systemic ideology in looking for what is missing, aka the subtraction approach. Why is Nigeria not like Denmark? Why is the growth so slow in Senegal?
To look for what is missing in Africa is an old sport. Africa has always been seen through the negative mirror. Lack of light, lack of hope, lack of a future, lack of progress. Perhaps why the African dummy was set in the 1970s, where African countries’ economic situation was bleak and not in the 1960s, when their economies were rather bright.
Westerners have always tended to look for what was missing by using ourselves as the point of departure. This approach led to ironic consequences. At one point in history, even Africans were missing from the statistics, because Whites were so busy that they “forgot” the non-whites.
During the Colonial Era, several African countries, like North and South Rhodesia (modern day, Zambia and Zimbabwe), the value added productions made by Africans were excluded from the colonial statistics. The work done by millions of people did not count. That means historical growth statistics are inherently systematically and historically flawed.
To look for what is missing simply appears natural. The history matters-economics have been extremely influential in that aspect, unfortunately. Daron Acemoglu and James A. Robinson are among the more vocal voices, likewise is Nathan Nunn and his work on the argued causation between slavery and current poverty. None of it has been proven. Poverty traps have been mentioned by Collier and others. Also debunked by economists like Jerven and Easterly. To be clear, poverty traps do not exist!
Instead, to look for what is missing in Africa, we should look for what is present. We should not ask why the growth is slow, we should ask what promotes the growth that is present.
Instead, to ask why Nigeria is not like Sweden (the answer is simple because Nigeria is not Sweden), ask what can be done to improve Nigeria. What can be done to bring forth the best in what makes Nigeria Nigeria. Furthermore, because X worked for Sweden, it does not mean it will work for Nigeria or any other country, and vice versa. We shall learn from each other, but also realise one size does not fit all.
African countries and institution have an interest in keeping this bubble intact too. The numbers indicate rapid recent growth. It might also be shameful to admit, that in spite inviting donors to the country, where numbers are presented by banks in glamours settings, the quality of the groundwork is pitiful.
Let the Bubble Burst
To return to the original analogy to the movie “The Big Short”, this shit bubble is different than the housing bubble. When the housing bubble burst, the average person suffered. In this case, it is the unburst bubble that keeps affecting the average person negatively.
Poor numbers make it impossible to do anything properly. Numbers are the foundation of the house. If the numbers are bad, the house itself is weak. When the bubble bursts, we can take an honest look at the numbers, improve the quality, that several African countries have started doing. Good governments need reliable numbers to carry out necessary policies.
To know if existing policies are working. E.g. if a government reformed the agricultural sector, they need to know if the reform worked. Do we produce more food?
Without honest numbers, governments do politics while blindfolded. That harms the people. Western donors launch projects after projects guided by poor numbers, that as well negatively affects the people.
Mythological reasons are giving why the African continent is not like the rest of the world. All based on false premises.
This bubble has to burst before sustainable progress can take place.
The good news is that it is happening. Especially African economists have begun airing their critical voices.
It is scholars and the general audience in the Western world that have to wake up. Some Western scholars have begun to crush the numbers. Actually, a few began to air critique decades ago, but it appears nobody was listening. I hope this time, that the critics can have a positive impact. At least, we should begin to look at the statisticians, to assist local governments in improving the quality of the work. The African think tank, the Afrobarometer has already cited Jerven’s work, why their Lived Poverty Index is better than using the numbers used to calculate poverty and inequality in African countries than GDP and GNI.
Why accept a bubble in the first place?
In the movie, the banks did not care about the distorted numbers, not because they were stupid, but because they did not care. It also turned out for good reasons. When the housing bubble burst, they were bailed out by the taxpayers.
When it comes to the shit bubble about information on Africa, I honestly do not think, that journalists, economists, and governments are that cynical. But it is also a fact, that the predominant part of people engaged in the usage of poor numbers are not affected if a project or strategy fails.
In the movie, the banks did not care, and they had little to lose, making them more prone to gamble. The lack of negative feedback undermines a willingness to correct possible wrongs and it reinforces a sort of laziness to dig into the numbers supplied. No one read the CDOs.
The donor and aid industry suffer from the same problem. Ritva Reinikka has done an excellent piece titled “Donors and Service Delivery“ featuring in the anthology “Reinventing Foreign Aid” from 2008. He writes about the dangers that donors are not affected by the harm caused by faulty programs and strategies. If a project fails, a new one is set up. If a project goes horribly wrong, it is the local population that suffers. The tragedies inflicted by wrong strategies imposed on the continent during the neo-liberal era were all felt by the African people, not the ones behind the scene. That is a huge democratic problem that furthermore reinforces laziness and lack of scrutiny.
Statistics, present as the past, are flawed. At times the numbers are based on nothing but guesswork. The people hired to collect the numbers in each African country are impaired by lack of proper funding, causing the numbers to be poor to non-existent at times. Several African countries have begun to correct the situation, though, but it is still far from satisfying.
Unfortunately, it is the poorest countries that need proper numbers the most, and the same countries are the least equipped and capable of paying the costs to run an efficient statistician office.
When companies and you accept numbers presented by the World Bank and others at face value, you risk reproducing bad numbers. Do not forget, that several numbers do not carry any metadata whatsoever.
It is also concerning, that several numbers and ideas are built on a false premise from the very beginning, such as the African dummy.
Especially, that Western economists still tend to look at African countries and find out what is missing, instead to look at what is present. It creates a flawed growth literature, where Africa continues to be victimised, that further nurses the white savior syndrome.
The conclusion is, when it comes to numbers about an African country, do not accept it as face value. More should also be done to empower the statisticians, whereas we keep making their workload heavier negatively affect the quality of the numbers produced.