McKinsey Global Institute has an interesting article that highlights the importance of microregions as opposed to country averages to account for changes in growth and development.
The concentration of global economic activity looks very different under a microregional lens. For instance, India and Portugal at the country level might have large difference (5x) in per capita GDP, but if we look at Goa (India) and Porto (Portugal) there is not much difference (GDP per capita of $33,000 in 2019). We will see similar pattern in other countries and their cities in terms of life expectancy.
MGI's analysis shows that half of the additional GDP generated from 2000 to 2019 came out of 3,600 microregions from a total of 40,000 as ranked by the increase in GDP per square kilometer. These 3,600 microregions were scattered across 130 countries but cover just 0.9% of the world’s land mass. 27% of the global population lived in them in 2019, totaling two billion people.
In India, they found 270 microregions home to 114 million people in 2019 where GDP per capita grew more than $7,100. The country average excluded them. Microregions with GDP per capita gains of at least $7,100 (or the top 30% globally) were considered.
They also regressed five-year moving average annual growth rates at the microregional level on annual growth at the country level to estimate the explanatory power of country-level growth. The result showed that a country’s GDP per capita growth rate can explain only 20% of the variation in the microregional growth rates in that country.
