Quantifying the impact of green leasing on energy use in a retail portfolio: limits to big data analytics

Panel: 8. Monitoring and evaluation: building confidence and enhancing practices

This is a peer-reviewed paper.

Authors:
Ramon Granell, University of Oxford, United Kingdom
David Wallom, University of Oxford e-Research Centre, United Kingdom
Susan Bright, University of Oxford
New College, United Kingdom
Kathryn Janda, Environmental Change Institute
Oxford University Centre for the Environment, United Kingdom

Abstract

The retail sector is a significant contributor to any industrialised economy and, as a result, a major consumer of energy. Larger retailers are normally aware of the contribution energy makes to their operational costs with efficiency activities of different types and scales desirable. Within the commercial property sector implementation is complicated by the different models of property ownership including the predominance of rental as the occupancy model. This introduces a barrier as many of the larger, and hence more impactful, energy efficiency measures need collaboration between the owner and occupier for successful implementation as they may require changes to either the building fabric or plant/equipment.

New mechanisms aim to smooth this possible barrier through the use of environmentally friendly legal instruments. With either a Memorandum of Understanding between parties with existing tenancy agreements or directly inserted clauses specifying mechanisms for collaboration between parties within new lease agreements (a Green Lease) and it is clearly beneficial to be able to provide some quantifiable measure of their impact.

Using data from a UK retail chain we have investigated using a number of different analytical methods how consumption changes after MoU/GL introduction for a number of different classes of stores operated. This included the identification of features which could show a connection, i.e. reduction, in consumption with the introduction of an MoU/GL. It is clear though that with the limited dataset available that it will be difficult to establish a clear causal link between their introduction and statistically significant consumption changes. We will discuss the limitations of the use of Big Data analytics using currently available data, what further information availability is necessary and what additional questions above and beyond just quantifying benefits have been possible with these information streams.