Descriptive Evidence. Table 2 reports a general profiling of occupational groups across several characteristics. Compared to the benchmark group of low-skilled non-green occupations, green-growing manual jobs exhibit a similar degree of geographical dispersion, offer a 20% wage premium and employ much older workers on average. Notably, green manual occupations are disproportionately biased towards male workers (more than 90%). As Covid-19 exposed occupations represent the lion’s share of low-skilled occupations, it is not surprising that they are very similar to the benchmark group. Low- skilled occupations that are RTI-intensive and with a high-computerization probability also have characteristics similar to those of the benchmark group, and thus have similar differences with green occupations. By contrast, brown occupations are much more similar to green occupations in terms of gender composition, age and wages. However, they are significantly more concentrated spatially. This spatial concentration is a well- known concern for high-carbon communities (Fragkos and Paroussos, 2018), which is largely beyond the scope of this paper. Table 2 – Profiling of occupational groups % of total empl in 2017 Number of 6- digit SOC occupations % of male employees Age of employees (average) Average hourly wage in 2017 (US$) Locational GINI coefficient Growing (2009-2017) green manual occupations 5.2% 18 92.6% 43 21.00 0.26 Growing (2009-2017) green manual occupations (weighted with greenness) 1.0% 18 90.3% 42 20.46 0.31 Benchmark: all non-green low-skilled occupations 59.2% 386 50.5% 39 16.97 0.33 Low-skilled occupations at risk because of Covid- 19 34.2% 157 56.4% 37 15.57 0.31 Low-skilled occupations with highest RTI (top 10%) 9.8% 53 32.6% 38 14.14 0.32 Low-skilled occupations with highest computerization probability (top 10%) 7.9% 46 35.3% 38 14.99 0.32 Brown manual occupations 2.8% 79 79.3% 43 21.55 0.60
