In a recently published paper, we added a driver that is missing so far in the SSP framework - the evolution of the sectoral structure of economies. Structural change scenarios represent a well-known characteristic that accompanies the process of economic growth and development - the reallocation of economic activity between the three major sectors agriculture, manufacturing and services.

The need for scenarios of structural change

The existing SSP scenarios of GDP, widely used in IAMs, are often too coarse-grained for studies by the impacts, adaptation and vulnerability (IAV) community. We took up a challenge of developing scenarios of how economic structures might change over time: this will be crucial for sectoral assessements of impacts of climate change, as well as for more detailed analyses of adaptation and mitigation strategies. Our new projections can be used, for example, to refine energy demand scenarios and as an input to analyses that address the impact of climate change and climate change mitigation at a sectoral level. The projections are available on the country level, as well as aggregated to 12 world regions.

We compute future trajectories of changes in the economic structures based on the concept that all nations follow similar developments on stages of economic growth. We construct scenarios for the sectoral shares of labor, value-added and energy based on historical data and an econometric approach. These scenarios are also linked to the original GDP scenarios and implicitly capture properties of the underlying narratives. We use updated SSP scenarios of GDP per capita harmonized to the most recent economic data and short-term growth projections covering the Covid shock (Koch and Leimbach, 2023). This helps us avoid introducing uncertainty into the projections in periods without real-world uncertainty.

Even though using the updated GDP projections for the construction of structural change scenarios reduces the comparability with the original scenarios, the nature of the structural change scenario variables as dimensionless figures allows combining them with different GDP scenarios and metrics.

We project key variables of economic activity until 2050: sectoral shares of employment, value added and final energy use for the sectors agriculture, manufacturing and services. The five SSP scenarios are available on a country and region level.

Challenging roads ahead

Our results are consistent with the SSP narratives: the transformation process is the fastest in an SSP5 world, and slowest in an SSP3 world. Scenarios for developing countries reproduce structural change patterns (e.g., hump-shape of manufacturing labor share), observed for developed countries in the past. In SSP5 and SSP1, developing countries are projected to experience immense employment reductions in agriculture until 2050. On the other hand, under SSP3 and in particular in Sub-Saharan Africa, it turns out that there is hardly any structural change but almost constant share of agricultural activities. This potential lack of structural transformation and build-up of industrial capacities may intensify the problems developing countries have with adapting to climate change and investing in climate change mitigation.

We also measured differences in labor, value added and energy shares between scenarios that include the Covid shock, and scenarios that do not. In general, the differences are small, larger in 2020 than in 2050, and larger for the labor shares than for the value added and energy shares. We find a maximum difference of 1.5 percentage points for India in 2020. Without the Covid effect, less labor is projected to work in the agricultural sector which in 2020 is the sector with largest labor share in India. Overall, we find that the projections of structural change are robust against the near-term Covid shock.

Future research directions

Our approach covers the major determinant of structural change, but there are certainly more. Including additional factors could easily extend the applied methodological approach, but the major limitation is the availability of scenarios that differentiate additional independent variables along the SSP dimension. Furthermore, future research on structural change scenarios has to go beyond the econometric approach and projection methods based on historical patterns. Not all factors that will impact structural change in the future will be covered by historical data, as for example, the impact of the digital transformation on sectoral energy consumption or the process of deindustrialization of developed countries based on uncertainties about the global geopolitical situation. Future research on structural change scenarios should be directed to overcome these limitations. A further sectoral disaggregation could help to increase the range of application of structural change scenarios, particularly for climate impact and vulnerability studies.

References