Updated: January 2023
Major differences exist in the quality of both GDP and population data between countries.
When GDP and population data of poor quality are combined to produce GDP per capita data the accuracy of the resulting ratio is further diminished.
Significant variations in the quality of both GDP and population data compounds the problems in assessing the accuracy of measurements of GDP/capita which is the proxy means used for measuring living standards across countries.
World Economics measures variation in the quality of both GDP and population data, by the GDP Data Quality Ratings (GDP DQR) and the Population Data Quality Ratings (PDQR) respectively and employs both scores to assess the overall accuracy of GDP per capita data. In most cases, because of out of date base years, a lack of resources in national statistics offices and a large unrecorded informal economy, official GDP figures are underestimates of actual economic activity, although in some rarer cases GDP may be overestimated as a result of government interference.
Population data is most likely to be underestimated because of gaps between census years, low statistical capacity, high fertility and declining mortality rates and low rates of registration of birth particularly in many African and Asian countries. On the other hand there are also cases when a country’s population is overestimated and even census data can be highly politicised and distorted. For example, in Nigeria, many of the 36 constituent states have over claimed population size to gain bigger subventions from the Federal authorities or political power. In order to maintain their relative power position each state managed implausibly to maintain its share of the total Nigerian population in two censuses fifteen years apart. In the first census as an independent country in 1962 some Nigerian states in the east of the country reported increases in population of over 200% over the decade.
These problems are not confined to emerging market countries. The highly regarded Brookings Institution in the USA recently noted in respect of the planned American Census, “The 2020 Census may be wildly inaccurate – and it matters more than you think.”
World Economics estimates the overall accuracy of GDP per capita data because of these problems by a combined index based on dividing the GDP DQR and the PQR by 100 and multiplying the two figures together to get a score between 0 and 1. This takes into account the fact that that errors are multiplicative so for example, a country like Benin which has a low DQR and a low PQR will have a much lower rating for GDP per capita. The final scores are converted into a ranking between 0 and 100 by multiplying by 100. See Data Ratings for countries and regions.
This data can then be sorted into quartiles and ranked by quality from A to D. For example, Lebanon which has a B ranking in terms of its DQR score, has very unreliable GDP per capita data on account of its low quality population data and ends up with a D rating. These are shown in the data ratings tables.
The compound errors arising from poor GDP and population data means that it is likely that in the case of GDP calculations for DQR and PQR category C and D countries, GDP per capita estimates could be out very substantially.
The massive errors in estimating a country’s GDP per capita possible from multiplied error factors from using poor quality national income data and population data is illustrated by a number of examples below for Brazil and Nigeria.
In the case of Brazil both GDP and population could be a lot higher than official figures, but the errors in GDP are likely to be larger meaning that GDP per capita could be much higher than official data suggests. In Nigeria, although GDP is likely to be substantially underestimated, very poor population data quality means that GDP per capita could be higher or lower than official estimates.
GDP Data Quality Ratings The quality of GDP economic data in Bolivia is relatively poor and according to the DQR the country has only a D ranking with a score of 28.3 out of 100. The reason for this poor rating is that Bolivia produces GDP data using a 32 years out-of-date Base Year, makes little attempt to include data covering its large estimated informal economy and uses Price Indexes based on decade old data. It is estimated that introducing a more recent Base Year alone could add up to 22% to GDP. Adding estimates for the Informal economy could easily add a further 55%. This means that GDP in Bolivia could be 77% higher than recorded.
Population Data Quality Ratings The quality of population data in Bolivia is relatively better than its economic data. The main issue is that the last census was held in 2012, over a decade ago, and according to the World Bank the country only registers 92% of live births meaning that population estimates are becoming increasingly unreliable until the next census is held. This means that if 8% of births are not recorded, ignoring deaths and net migration, the population could be significantly higher than estimated. GDP Per Capita
In 2021 according to the World Bank Brazil generated a GDP per capita of US$9,248 in PPP terms in current prices, but as a result of a the combination of potential economic and population data errors, the real figure could be within a wide range of this figure.
GDP Data Quality Ratings The quality of GDP economic data in Nigeria is relatively better than Bolivia with a DQR ranking of D and a score of 52.1 out of 100. The reason for the D rating is that although Nigeria has a very large informal economy making GDP estimates uncertain, it produces GDP data with a eleven years out of date base year, and it uses the latest System of National Accounts. It is estimated that introducing a more recent Base Year alone could add up to 29% to GDP. Adding estimates for the Informal economy could easily add a further 57.7%. This means that GDP in Nigeria could be as much as 86.5% higher than official data.
Population Data Quality Ratings The quality of population data in Nigeria is very poor and the country receives only D rating with a score of 76.1 out of a 100. There are a number of problems. One issue is that the resources devoted to statistical cpacity are low.
One view is that Nigerian population recorded by census data is overestimated because of the political pressures alluded to above. On the other hand according to the World Bank the country only registers 43% of life births, a very low figure, meaning that population estimates are becoming increasingly unreliable and will continue to be so until the next census is held. This means that if 57% of births are not recorded, ignoring deaths and net migration, the population could be as much as 50% higher than current estimates. GDP Per Capita World Economics has estimated Nigeria's GDP Per Capita to be $9,325 in PPP terms by the end of 2022, a long way behind South Africa at $19,342 despite the larger Nigerian economy.
The combination of errors in the estimates of GDP and of population means that GDP per capita data can be very unreliable as a source of international comparison of living standards.