Scotland's productivity gap with international countries: research

Annex: Methodology and limitations

The research adopted a comparative perspective focussing on Scotland and a set of European benchmark regions. The analysis was undertaken for EU NUTS1 regions, for which a rich database was gathered of official data outputs, inputs, as well as a number of other economic and socio-demographic characteristics. A levels accounting framework was implemented to investigate different levels in productivity between Scotland and close comparator regions at the beginning of the period analysed. A dynamic growth accounting approach was then employed to map sources of regional productivity growth in the period of 2009 to 2017. This type of framework cannot address causality, but is an accounting tool that has been employed extensively in the empirical literature to explain cross-country and cross-industry productivity differences. Both the levels and growth accounting allow for the identification of the main contributing factors to labour productivity cross-differences, disentangling the effect of capital investment, human capital or TFP.

One limitation of the methodology is around how labour quality was measured. Labour quality has been measured by changes in the employment shares of groups with different educational attainment levels. Average wages by the same education group were then used as a proxy for worker productivity, to calculate the contribution of labour inputs to labour productivity. A potential limitation of this approach is that it is unable to formally incorporate qualification mismatches beyond those implied by the wages paid. For example, this work did not use survey data to draw from workers' perception on the match between their jobs and education to explore the link between skill and qualification mismatch and labour productivity. Due to small sample sizes, other productivity determinant characteristics such as age and gender were not considered. In addition, as the labour quality measure is a qualification based indicator, it does not capture "on-the-job" training and other soft-skills which can be acquired through professional activity and on-the-job training activities.

To measure capital accumulation, Gross Fixed Capital Formation was used. The limitation of this is that it does not pick up on significant parts of intangible investment (i.e. investment lacking physical or financial properties). Due to data limitations, the analyses that require the use of capital stocks only go to the year 2016.

There were data limitations around skills and organisational factors. There is not yet a comparable EU wide survey comparable to the UK Government's Employer Skills Surveys which is why only UK results are presented. Additionally, it was not possible to examine the role played by factors such as managerial practices, workplace performance or organisational factors on TFP performance due to a lack of harmonised regional data.