Hi there. In the last video, we spent some time looking at the growth and distribution of the world's population in the 20th century. We explained the dynamics of change. And we looked at the difficulties in compiling the data. Well in this video, we're going to configure the world in terms of output. GDP or Gross Domestic Product. And we'll look at its distribution and we'll asses the accuracy in measuring it. Now on the 6th of April, 2014, the head of the Nigerian statistical bureau announced, that his country's output was 89% higher than had previously been believed. He was proud to announce that the country had passed South Africa, as the largest economy on the continent. Of course, on the ground, nothing at all had changed. But what about the results of all the earlier calculations and employed the earlier data? Well, we'll return to this question later in the video. Statisticians, don't actually measure output. They measure transactions. If no money changes hands, it's not recorded, and if transactions are not recorded, they don't get counted either. To complicate matters, not all transactions are included. Illegal transactions for example are deliberately, excluded. And, to provide a final touch, they're not interested at all in transactions per se. But in the value added between each transaction. Well, needless to say, constructing national accounts is a hugely complex operation, involving the collection of data from a vast array of sources and compiling them into a coherent set of numbers. We've shown this, in more detail in the visualization, that accompanies this video. So, just imagine then, all over the world statistical agencies are engaged in the collation and compilation of national accounts. In order to make comparison easier, the results are usually expressed in a common currency, most often the Dollar, expressed in current values. And we use this form of calculation, when we're assessing the impact of countries on the world economy. Less so, when we're looking at their relative incomes or their relative spending powers. Now, the first thing to note, is that the distribution of the world's GDP stands in stark contrast the distribution with population. Despite the explosive economic growth of China and some of the other developing countries, the world economy remains dominated in size by Europe, North America and Japan. In per capita terms, the richest countries in the world are also concentrated in these regions. We'll see in a moment, that there are possibilities of wide margins of error, especially among the poorer countries. But even if we adjust for these, it wouldn't significantly change the overall picture. Now, we've observed that calculating GDP statistics is a complex exercise. It requires a sophisticated infrastructure of data reporting and registration, and that costs money. So it'll be no surprise to learn, that the greatest difficulty in obtaining accurate data, is among the world's poorer countries. One obvious problem lies simply in the task of data collection, with an underpaid staff and under equipped offices. With UN agencies demanding answers and statistical bureaus just simply unable to supply them, it didn't wonder that one researcher who described the results as simply random numbers. Now a second source of error lies in the informal sector. For example in the 1990s, officials in Tanzania estimated the size of the informal sector as anything between 30% and 200% of GDP. In the end they decided to hike their GDP estimate by 62% to take this into account. A third source of error, is simply fraudulent reporting, not by businesses, but by government themselves. In the wake of the Euro crisis for example. The Greek government has been accused of misrepresenting it's GDP figures, in order to hide the size of its debt. And a final source, of problems are legitimate re-estimations. And this is where the Nigerian case is interesting. The country hadn't adjusted the baseline for it's calculations for over 20 years. They did also based its GDP on very small sample of businesses, remedying both these deficiencies resulted in that 89% boost, to its GDP figures. But its not the first African country to have done this. So far, 12 countries have reported the results of rebasing exercises. And many more are still to come. All these efforts at improving statistical accuracy must surely be welcomed. But my point is this, let's accept that with some exceptions, the richer countries tend to produce better statistics. But many of the statistical exercises undertaken by social scientists are aimed at the poorer countries. And it's here where the errors are high, where any improvements are haphazard and disjointed. So what is the value of any ranked series of data at any point in time? What is the validity of any effort to look at changes over time, as we do when we talk about economic growth? And what therefore, is the validity of any statement made on this basis? Regardless of how neat and tidy the statistical outcome, might appear to be. As we'll see in this course, social scientists spend a lot of time and ingenuity in constructing data, for different social science variables. But they tend to ignore the possibility that a basic data set like GDP, might get so totally unsuited for statistical analysis. Okay, lets tie all of this together. We've looked at how world output is measured, and we've pointed to the broad outlines of its distribution. We spend a considerable time, looking at the difficulties in estimation, especially among the poorer countries. And we've underlined how this affects the use of GDP indicators, in any statistical exercises. Well, we've tried to configure for you the size of the various economies. We invite you now to look at our visualization of the world map of GDP. In the next video, we'll look at how the world output is estimated so it's better to reflect, well, real world output.
