The research was supervised by Edelmiro López Iglesias and Xesús Pereira López
Input-output analysis is a tool widely used by real science to study economic, social and environmental phenomena. Since its conception, (inter)regional input-output analysis has faced the problem of the scarcity of detailed data for the different sub-national scales. This problem is particularly severe in the less developed regions, where there are often insufficient resources to obtain information. In the specialised literature there is a general consensus asserting that the construction of hybrid input-output models -those that combine direct observations with indirect estimations- constitute the most efficient solution to the problem of data scarcity.
In his PhD thesis, ‘Expanding hybrid approaches to construct (inter)regional input-output models‘, the Ecoagrasoc group researcher Fernando de la Torre Cuevas develops new methods for (inter)regional input-output analysis with scarce data, thus expanding the toolbox available to researchers in this field. The work, defended on October 19th, was supervised by Edelmiro López Iglesias and Xesús Pereira López, professors at the Universidade de Santiago de Compostela, within the framework of the Doctoral Programme in Regional Development and Economic Integration. The doctoral thesis achieved the ‘cum laude’ distinction.
More specifically, the study introduces three methodological innovations that make information requirements flexible in order to solve three problems associated with this type of models: updating supply-use tables, deflating supply-use tables from current to constant prices and nesting regional models within world-scale models. For each of these innovations, a detailed mathematical description is made and the corresponding empirical tests are carried out to check the accuracy of the estimates. As a general result of the research, this research shows how it is possible to expand the scope of input-output analysis at the regional level by combining information gathered through direct observation with indirect estimates in a more flexible and efficient way.
The study led to the registration of an algorithm as intellectual property, ‘Modified KRAS: input-output matrix balancing with conflictive information‘, with direct applications for research organisations using input-output frameworks, statistical agencies and institutes at the regional level or in developing countries, and organisations producing multiregional input-output models.