The “employment factors” have been used to calculate how many jobs are required per unit of electrical capacity. They take into account jobs in manufacturing, construction, operation and maintenance and fuel. The tables below list the employment factors used in the calculations. These factors are calculated for OECD countries. For other regions, a regional adjustment was used.

Because of its dominance in current electricity supply, regional employment factors were calculated for coal mining in the 2009 analysis. The calculations included figures from national employment data where available, and historic coal production, with most data for 2006/2007. These employment factors have been projected to 2010 using the 2009 GDP growth data from IEA 2009, but the coal production and employment figures have not been updated.

It is important to note that coal is mined using extremely different methods around the world, and employment per unit of electricity also varies according to the type of coal and the efficiency of generation. In Australia, for example, coal is extracted at an average of 13,800 tonnes per person per year using highly mechanised processes while in Europe the average coal miner is responsible for only 1,800 tonnes per year. China is a special case: even though it currently has a very low average rate of extraction per person (700 tonnes per employee per year) this will change very soon, as thousands of small mines close and new super-mines open. For this reason, the model uses US employment factors (above current levels) for future coal production in China (for a detailed discussion of the coal employment factors see Rutovitz and Atherton, 2009).

The factors for gas generation were taken from a publicly available model called JEDI, developed by the National Renewable Energy Laboratory in Washington to help work out local benefits of different types of energy supply.

For nuclear energy, the factors for construction, manufacturing and installation were derived from a Nuclear Energy Institute 2009 factsheet, while operations and maintenance was calculated using Energy Information Administration data. Fuel employment was calculated from Australian census data.

For the renewable energies, employment factors were taken from industry data where available, as listed in Table 7.5, or derived, depending on the maturity of the technology.

A number of ‘adjustment’ factors were used to make the employment calculations more realistic, including:

• regional job multipliers The employment factors used in this model for all processes apart from coal mining reflect the situation in the (typically wealthier) OECD regions, so regional multipliers are applied to make the jobs per MW more realistic for other parts of the world. In developing countries it typically means more jobs per unit of electricity because of more labour intensive practices. The regional multipliers are the ratio of labour productivity in the region to labour productivity in the OECD. The multipliers change over the study period in line with the projections for GDP per capita. This reflects the fact that as prosperity increases, labour intensity tends to fall.
• learning adjustments or ‘decline factors’ This accounts for the projected reduction in the employment per MW of renewable and fossil fuel technologies over time, as technologies and companies become more efficient and production processes are scaled up. Decline factors are calculated from the cost reduction projected in the energy modelling, as generally, jobs per MW would fall in parallel with this trend.
• local manufacturing and fuel production Some regions do not manufacture the equipment needed for wind power or PV, for example, nor do they produce sufficient coal for their needs. The model takes into account the percentage of each technology which is made locally. The jobs in manufacturing components for export are counted in the region where they originate. The same applies to coal and gas, because they are traded internationally, so the model shows the region where the jobs are actually located.