Third, we provide (gross output based) growth accounting results by industry aggre- gated consistently into value-added based growth accounting for the UK market sector, using the approach of Jorgenson et al. (2007). Thus we can examine the contributions of different industries to overall growth. This then speaks to the question of, for example, how much manufacturing versus financial services contributed to overall TFP growth.
On specifically UK data, our work is closely related to the industry-level work by Basu et al. (2004). They incorporated software as a productive asset and looked at productivity and TFPG in 28 industries from 1990 to 2000. They did not have data however on other intangible assets and so whilst they were able to document software and hardware spending across industries, they were not able to look at other co-investments in innovation. However, Oliner, Sichel and Stiroh (2008) considered a proxy for intangible investments applying the Basu et al. (2004) framework to US data. As will be clear, our work builds on these studies and we rely heavily on their important work on measuring software and also tangible assets, now embodied in official UK data collection. Likewise, our work is also closely related to the EU KLEMS project (O’Mahony and Timmer, 2009). Their dataset includes software, and we extend their framework with additional intangibles, explicitly setting out the industry/market sector aggregation.