Monday, November 24, 2014

The Problem with Poor Countries’ GDP

SEATTLE – Even in good financial times, development aid budgets are hardly overflowing. Government leaders and donors must make hard decisions about where to focus their limited resources. How do you decide which countries should get low-cost loans or cheaper vaccines, and which can afford to fund their own development programs?

The answer depends, in part, on how we measure growth and improvements in people’s lives.  Traditionally, one of the guiding factors has been per capita GDP – the value of goods and services produced by a country in a year divided by the country’s population. Yet GDP may be an inaccurate indicator in the poorest countries, which is a concern not only for policymakers or people like me who read lots of World Bank reports, but also for anyone who wants to use statistics to make the case for helping the world’s poorest people.

I have long believed that GDP understates growth even in rich countries, where its measurement is quite sophisticated, because it is very difficult to compare the value of baskets of goods across different time periods. In the United States, for example, a set of encyclopedias in 1960 was expensive but held great value for families with studious kids. (I can speak from experience, having spent many hours poring over the multi-volume World Book Encyclopedia that my parents bought for my sisters and me.) Now, thanks to the Internet, kids have access to far more information for free. How do you factor that into GDP?

The challenges of calculating GDP are particularly acute in Sub-Saharan Africa, owing to weak national statistics offices and historical biases that muddy crucial measurements. Bothered by what he regarded as problems in Zambia’s national statistics, Morten Jerven, an assistant professor at Simon Fraser University, spent four years examining how African countries obtain their data and the challenges they face in turning them into GDP estimates. His new book, Poor Numbers: How We Are Misled by African Development Statistics and What to Do about It, makes a strong case that a lot of GDP measurements that we thought were accurate are far from it.

Jerven notes that many African countries have trouble measuring the size of their relatively large subsistence economies and unrecorded economic activity. How do you account for the production of a farmer who grows and eats his own food? If subsistence farming is systematically underestimated, some of what looks like growth as an economy moves out of subsistence may merely reflect a shift to something that is easier to capture statistically.

There are other problems with poor countries’ GDP data. For example, many countries in Sub-Saharan Africa do not update their reporting often enough, so their GDP numbers may miss large and fast-growing economic sectors, like cell phones. When Ghana updated its reporting a few years ago, its GDP jumped by 60%. But many people didn’t understand that this was just a statistical anomaly, not an actual change in Ghanaians’ standard of living.

In addition, there are several ways to calculate GDP, and they can produce wildly different results. Jerven mentions three: the World Development Indicators, published by the World Bank (by far the most commonly used dataset); the Penn World Table, released by the University of Pennsylvania; and the Maddison Project at the University of Groningen, which is based on work by the late economist Angus Maddison.

These sources rely on the same basic data, but they modify it in different ways to account for inflation and other factors. As a result, their rankings of different countries’ economies can vary widely. Liberia is Sub-Saharan Africa’s second-poorest, seventh-poorest, or 22nd-poorest country in terms of GDP, depending on which authority you consult.

It is not only the relative rankings that differ. Sometimes, one source will show a country growing by several percentage points, and another source will show it shrinking over the same time period.

Jerven cites these discrepancies to argue that we cannot be certain whether one poor country’s GDP is higher than another’s, and that we should not use GDP alone to make judgments about which economic policies lead to growth.

Does that mean that we really don’t know anything about what works (and what doesn’t) in development?

Not at all. Researchers have long used techniques like periodic household surveys to collect data. For example, the Demographic and Health Survey is conducted regularly to determine things like childhood and maternal death rates. Moreover, economists are using new techniques like satellite mapping of light sources to inform their estimates of economic growth. Although such methods are not perfect, they also are not susceptible to the same problems as GDP.

Other ways to measure overall living standards in a country are similarly imperfect; but they nonetheless provide additional ways to understand poverty. One, called the Human Development Index, uses health and education statistics in addition to GDP. Another, the Multidimensional Poverty Index, uses ten indicators, including nutrition, sanitation, and access to cooking fuel and water. And, by using purchasing power parity, which measures the cost of the same basket of goods and services in different countries, economists can adjust GDP to gain better insight into living standards.

Yet it is clear to me that we need to devote greater resources to getting basic GDP numbers right. As Jerven argues, national statistics offices across Africa need more support so that they can obtain and report timelier and more accurate data. Donor governments and international organizations such as the World Bank need to do more to help African authorities produce a clearer picture of their economies. And African policymakers need to be more consistent about demanding better statistics and using them to inform decisions.

I’m a big advocate for investing in health and development around the world. The better tools we have for measuring progress, the more we can ensure that those investments reach the people who need them the most.

Read more from the "What Now for Global Trade?" Focal Point

  • Contact us to secure rights


  • Hide Comments Hide Comments Read Comments (14)

    Please login or register to post a comment

    1. CommentedSinthia Sorkar

      Wanna know some secrets about Bill Gates? Go here:

    2. CommentedTheStudent Economist

      Here is my opinion based on both anecdotal and rather subjective judgement. From what I have heard Africa has a patriarchal society where women actually do most of the work. I've heard of some success in programs where banks support women headed businesses. Personally, it would be encouraging to see a greater initiative to provide impoverished women with the support they need to grow small businesses. Surely this would also affect violence against women rates and help communities grow faster than they would with just standard aid.

    3. Commentedfernando fajardo

      Biil said "Now, thanks to the Internet, kids have access to far more information for free. How do you factor that into GDP?." That's not a problem. Using the internet helps one become productive which leads to higher GDP and per capita income.

    4. CommentedJonathan Lam

      Gamesmith94134: The Problem with Poor Countries’ GDP
      If there is no Microsoft, where will we find Window 8? Eight windows?
      In America, there is a 2-3% growth last month, but the PPI is down, so is the new housing. Are we in the disinflation or deflation now? So, Bill, why are you so concern of the GDP? Some barter for necessity, and some marry a wife if he have ten cows. In such a way, no one is read the 0 and1 anymore; it’s all about hardware and software that travels at the speed of light. Statistics fools most of the experts and even economists.
      Graphically, you can use the infrastructure in the Human Development Index which what the region carries; and programs of applications like Multidimensional Poverty Index that how can the local government do for its people? This is your hardware and software in opening the Window that give the resources and energy to innovate and attribute for the new looks of lives other than your Microsoft logo.
      Many lives in minimalism and data of values are foreign to them. However, some may have more cell phone than food, some made their daily lives with lesser than two dollars a daily; but many survives through the toughest environments. So, I am not surprised with “When Ghana updated its reporting a few years ago, its GDP jumped by 60%.” Relatively, how much difference do they make when accountability is undercut?
      “Does that mean that we really don’t know anything about what works (and what doesn’t) in development?” It depends on the nature like droughts and earthquakes and human factors like wars and conflicts. Subsequently, they all need a helping hand like you with the resources and intelligence to put applications to use like the Microsoft office. It helps me a lot with just copy and paste or cut and paste. I went through the toughest time.
      At present, with much of the advanced technology and data service, I think World Bank, Development Bank, IMF and WTO are molding the unknowns into the system that is compatible to us that economists are working closely in understanding better of what valuation is applied to their livelihood and providing us a better picture of what development is being forwarded.
      Mr. Gates, data service can come later if the environment allows? Perhaps, if you concern of what is your goal for development, think of the Aqua duct stretching from Sacramento to Los Angles. Some thought digging a tunnel in the Peru to turn highland to farmland. Why not you? Imagine the water runs from White Tower to where you want. It gives and develops clean water is a good start; or goes where you think your resources can help most. There is no scoreboard in humanity, and GDP is not compatible with our finance system, besides, it is foreign.
      May the Buddha bless you?

    5. CommentedHemant K Chitale

      Quite possibly true to a great extent. Underdeveloped countries would have underdeveloped methods to collect and report statistics. There would be significant gaps in statistcs, some statistics might be fudged at the point of collection.

    6. CommentedDavid Olsen

      I can see how it might be helpful for an accountant in a development agency to have a number that he can quote to back up a decision to allocate funds to a specific country or region, but this isn't about profit maximisation within a firm.

      The question should be : if you take a specific region that needs poverty relief, what actually works? What will actually increase prosperity for a village?

      People are generally resistant to change, but the actions of a neighbour is often the best spur to action. A guy I met in South Africa told me that one person will start selling oranges on the street, the next day there are 10 people selling oranges on the street. Nothing like seeing an idea in action and the fear of your neighbour getting ahead of you to generate action.

      Concrete examples of local people on the ground using a successful low tech idea that all his or her neighbours can copy and claim as their own idea can be more beneficial than a million dollar top-down instruction programme.

      Give one go ahead small farmer the tools, seed money and materials to create a clay lined rainfall dam plus simple irrigation system, and all the other farmers around him will want one as well. Even if the irrigation breaks down, people will still be able to repair a small dam and bring buckets to it. Something will be left from the money spent to aid people once the NGO's have packed up and left for the next most fashionable development idea.

    7. CommentedPaul Peters

      Is it wise to use any statistics at all?
      No bell curve can express power laws over networks, nor does it express the tensional integrity. Plus, countries have two such network distributions, or two side of the spectrum, highly coherent urban areas and the rural areas. It shouldn't be all too difficult to get to a solid measure of the geographically distributed socio-economical cohesion.
      Something showing a relative status and trend movement would suffice, such contrast gradients?

    8. CommentedColin Ong

      As an economist, I have always struggled even with the terms quality of life and standard of living. The reason is that as long as we cannot safely define who represents the average model citizen of the country, GDP estimates will just be fodder for those with strong political aspirations.

    9. Portrait of Sabina Alkire

      CommentedSabina Alkire

      I agree that better tools for measurement are essential. If poverty is acknowledged to be multidimensional then in addition to improving GDP measures, there is some urgency about agreeing a headline indicator for post-2015 MDGs that reflects at least some dimensions of poverty besides income.

    10. Portrait of Chris Blattman

      CommentedChris Blattman

      Like Bill Gates, I'm a fan of Morten Jerven's new book. And I can only assume a rousing endorsement from Gates left Jerven and his publisher in ecstasy.

      I would like to see better GDP numbers–who wouldn’t?–but it’s hard for me to see this making a real difference. What constraint on development would this revelation relieve? Why are better GDP numbers anywhere near the top constraints poor countries face?

      The problem with those of us in the development complex, be we academics or Presidents or foundations or NGOs, is we want the world nicely ordered with levers to pull and a dashboard to monitor. And so we put a lot of energies into levers and dashboards and monitors.

      I think of poverty and political powerlessness in terms of constraints and frictions–the limitless host of things, little and big, that made it more difficult to run a business profitably or turn a profit or invent a new product or get your kid educated or select the leader who serves your interests. States and institutions and norms and technology and organizations reduce these frictions and relieve these constraints. That is the fundamental driver of development. This is the basic logic behind almost every theory of development in your textbooks, from growth models to poverty traps to everything in between.

      Reducing frictions and eliminating constraints is maybe the best thing outsiders can try to help with, freeing entrepreneurs and citizens to do their thing. (Well, I guess we can also help by giving them a big market to sell things to, but that’s another story).

      To the extent that missing information and measurement constrains development, or creates frictions, there’s a long list of more likely candidates than GDP. A sample:

      - Small banks who don’t know the creditworthiness of the mass of potential borrowers,
      - Village leaders who don’t know what funds the local bureaucrats get from the center
      - Citizens who don’t know their MP’s meteoric rise in wealth
      farmers who don’t know prices a district to the west

      I kind of wish Gates would say “we need credit bureaus” or “we need freedom of information acts” instead.

      I’m not even sure information of these sorts are even the most important frictions to address. To the extent we pay them attention or design programs, I think it’s because they seem cheaper and easier to tackle than the harder ones. But they are all a far sight better than better GDP data.

      The litmus test: If we went back in a time machine, and Gates wanted to expand sales or product development or factories in Asia or Africa, would he have called for these things or better GDP data?

      My hunch is that GDP data wouldn't be on his wealth creation radar. Which might mean it shouldn't be on ours.

    11. CommentedFrank O'Callaghan

      The deep point that Gates is getting at is about what we are measuring. Is it 'real'?

      Basically, the quantities to be measured have been those deemed important to the people in power. Reality is always more complex than we expect.

    12. CommentedZsolt Hermann

      The GDP is unsuitable for such measurements for multiple reasons.
      The GDP simply measures the performance of a country within the artificial, unnatural constant growth socio-economic system, and it does not say anything about the true state of its people.
      We can see it clearly in Europe how the superficial statistics, "stimulus", "balancing" only concerns the state of the profit making machinery, completely ignoring that actual state and life quality of the public.
      Moreover as this artificial and unsustainable system is gradually collapsing the GDP will dramatically fall all over the world as even the virtual growth measured today will fall.
      If we want to measure something we need to know what we are measuring.
      Today the whole of humanity has evolved into a global, interconnected and interdependent network.
      Humanity practically behaves as a single, living organism.
      Thus in order to measure, examine, cure anything we have to measure, examine, and cure the whole system.
      Our present approach is similar to how classical, "modern" medicine looks at a patient, treating symptoms, or only a actually suffering organ, body part, without looking at the whole body in a holistic manner.
      As a result the billions of dollars given to charity, donation by philanthropists with good intention either fall into the wrong hands, or simply vanish like water poured into a bottomless vessel.
      The actual problems observed on the ground could be the result of problems in a far away country, related to financial, economical, social, geographical, ecological problems somewhere else, true lasting solutions can only happen by education, positive motivation instead of simple aid programs which education and motivation is also multi faceted, requiring global cooperation, support, and so on.
      Until people start to view the world as a single system, in a holistic manner, sensing, understating the multi-dimensional interconnections, their role, their optimal balance, each and every plan and action will fail, since we are treating the wrong problems with the wrong tools.
      Apart from securing the absolute, basic necessities for those who need it, and this should be generally straight forward, the next step has to be a global education program for each and every human being, helping them to understand today's integral, mutually interdependent world, as only on the basis of such understanding can we build a sustainable future, that is capable of functioning in an harmonious, balanced, natural manner.
      Those who still has money to offer should start investing in such global, integral education.

    13. CommentedWilliam Osterberg

      While Mr. Gates has 'long believed that GDP understates growth' in even rich countries, presumably including the U.S., from past professional experience at one of the major U.S. statistical agencies, I have long wondered whether inflation might be understated, implying that real output is overstated. The problem (one of many) is that companies voluntarily report price information and can be quite sensitive to the range of products covered and how the indexes are calculated. Simpy put, they do not want to be seen as having increased prices 'too much.' Needless to say, the issue is complicated.
      As regards the GDP of poor countries, the activities of multinationals cannot be ignored. Output generated by MNCs located in a developing country is considered part of that country's GDP yet the valuation of that output can be distorted by 'transfer pricing' done for bottom-line purposes.
      It is all well and good to suggest that more effort and money needs to be put into the collection and compilation of developing country statistics. However, the 'improvements' in such statistics needs to be driven by a neutral view of the proper measurment uncontanimated by private interests about which industries should be targeted for development aid.

    14. CommentedHem Acharya

      May be other economic indicators such as gross happiness quotient. Happiness quotient and distribution of per capita gdp or the range in statistical terms could be used while deciding on lower rates donations etc