Forecasting Global Growth to 2050

Forecasts of the future shape of the global economy, if they are at least somewhat accurate, can help planning and policy discussions in areas from global governance through business expansion plans. At the same time, this relies on forecasts in fact being somewhat accurate, and it’s a cliché that predicting things is hard, especially about the future. To allow for that uncertainty, in a new paper, Zack Gehan and I come up with a set of potential scenarios about the future using a reasonably simple approach. Results that hold across those scenarios are (we hope) more robust to uncertainty about the future. 

We generate our central forecast using two inputs: the coefficients from a regression analysis of the historical links between growth, demographic factors, education, and national temperature; and forecasts of those variables to 2050. And we create a plausible range of outcomes using the error terms from that regression to create low and high scenarios for groups of countries. It’s a process that shares some features with the one used to create the five ‘SSPs,’ the shared socioeconomic pathway scenarios that underpin some of the modeling around the extent and impact of climate change. (Although one difference: unlike the SSPs, we try to take into account at least some of the impact of climate change on future economic prospects).  

Our central forecast matches most closely the SSP called “Middle of the Road,” perhaps both reassuring and unsurprising. Our range for global GDP per capita in 2050 is between $16-39,000, and the range for the SSPs is $18-42,000 (current global GDP per capita is around $16,000). For what it is worth, that suggests to us that the SSPs do present a range of potential economic outcomes that seems reasonable –noting  again that, because we followed a somewhat similar approach to forecasting, this is hardly an utterly independent verification. The figure below presents our central forecast (KG) and range (KG+ and KG-) along with the most ‘extreme’ SSP scenarios in terms of global GDP per capita outcomes: SSP3, titled “Regional Rivalry (A Rocky Road)” and SSP5, titled “Fossil-fueled Development (Taking the Highway).”  

It is no challenge to think of events that could leave the world looking far different from any of the scenarios we work with. Even within the scenarios, the range of outcomes is large enough to suggest it is very plausible current global income convergence could dramatically accelerate or it could reverse, for example. The impact of policy change at the national and global level could be very large. That said, some conclusions look to be reasonably robust at least across the range of scenarios:  

  • Demographic change will be an increasing drag on growth particularly in richer (upper middle- and high-income) countries. Education is likely to be a factor favoring convergence globally (because average years of schooling will grow faster in poorer countries). And while climate change (at least as reflected in temperature, at the national level) will be a force for slower growth especially in poorer countries, it is unlikely to be a major driver of global economic trends up to 2050.  

  • The share of OECD DAC (traditional donor) countries in the global economy is very likely to shrink. This reflects the likelihood of (i) relatively slower per capita income growth (both compared to the past and to poorer countries); and (ii) stalling or declining population compared to continued population growth in most low- and middle-income countries.  

  • It is plausible to imagine that $2.15 a day poverty will have effectively disappeared by 2050 (which would be good news, if two decades later than envisaged by the UN Sustainable Development Goals). It is also plausible that more than two thirds of the world will be living on more than $10 a day (up from about 42 percent today). Low income countries may disappear as a group, and the proportion of the world living in high income countries is very likely to more than double from its current proportion of 16 percent.  

  • The shrinking share of the US in the global economy suggests an end to the country’s veto power in both IMF and World Bank decision-making processes unless those institutions move even further away from a voting/shareholding formula based on relative economic size.  

  • Global electricity consumption can be reasonably expected to double by 2050.  While high-income countries will see slow consumption growth, current low-income countries will still be responsible for less than five percent of electricity consumption (more likely two percent) in 2050.  

  • The US is likely to remain the largest military in terms of spending in 2050, but its global lead will considerably diminish, and it is plausible to imagine both India and China outspending the US on defense in 2050.  

For all of the challenges this likely future may present, it is one of a richer planet with more resources to respond to threat from pandemics through climate shocks, containing many fewer people living in the kind of absolute poverty that was the lot of ninety percent of humankind for nearly all of our history.  


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.

Image credit for social media/web: Adobe Stock