This report describes the results of 14 case studies carried out with organisations in Christchurch and Auckland to further inform and refine the Business Behaviours Module of MERIT. The Business Behaviours Module has been developed based on extensive statistical analysis of the responses of 541 Canterbury organisations to a 2013 survey on their experiences following the 2010/11 Canterbury earthquakes. We needed to test the applicability of the Business Behaviours Module to other types of disruption and to other urban locations. To do this we undertook a series of paired case studies, in two study regions, Canterbury and Auckland. Seven organisations that completed the Economics of Resilient Infrastructure (ERI) Canterbury business behaviours survey were initially recruited. Selection was based on their reported levels of disruption and to get representation from a variety of sectors, size and type of organisations. For each Canterbury case study a corresponding ‘pair’ organisation in Auckland was selected, with similar demographic features. A semi structured interview was carried out with the organisations exploring their responses to a number of actual and hypothetical infrastructure disruption scenarios.

Overall, the case studies revealed few geographic differences in how organisations in Christchurch and Auckland respond to disruption. Sector had far greater significance in predicting likely response to a disruption than location. With only a few exceptions, once an initial understanding of the business had been developed, the interviewer could have answered 80% of the remaining questions accurately. This supports the idea that the Business Behaviours Module developed from the Christchurch earthquake data is transferable to other urban contexts. However, the survey and case study data are based on large urban environments and further research is needed to ensure these findings are applicable within smaller communities with a greater rural interface.

The case studies revealed areas where the Business Behaviours Module needs to change. For example, there is a need to adjust the relationships for very short-term infrastructure disruptions; in these cases the business behaviours model predicts better performance than reality in the short term, followed by a slow recovery back to full operability. In reality, short term disruptions (e.g., a 36 hour electricity outage) creates a more pronounced impact on organisations during the period of disruption, but they are then able to recover back to full operability almost immediately after infrastructure services resume. To account for this we will adjust the Business Behaviours Module to differentiate between short-term and longer term disruptions, with a step function included for very-short term disruptions. Another change needed is to adjust the operability curves so that once an organisation has achieved close to full operability, then full operability is assumed. Otherwise the asymptotic nature of the operability curves means that an organisation approaches, but never returns to full operability.

Through the case studies we also identified a number of subtleties around how organisations responded to the Christchurch business behaviours survey. We will now use these insights to stress test the Business Behaviours Module relationships. For example, we would like to recheck the model for the influence of relocation and temporary closure on reported levels of disruption due to water, sewage and electricity outages. We will also review the weightings of disruption levels in the calculation of experienced disruption, in particular the weighting of ‘slight’. We also need to review the water infrastructure relationships for sectors not critically v ERI Research Report 2016/01 dependent on water, such as transportation, postal and warehousing, professional services, wholesale trade, information, media and telecommunications, and financial and insurance services.

Taking our case study organisations through a variety of different scenarios also revealed interesting differences in how the effects of different types of infrastructure disruption are ‘felt’ by businesses. For example, disruption to port services is experienced predominantly as an increased cost of doing business, rather than as an inhibitor to an organisation’s ability to operate. As such, a better place to model the effects of port disruptions may be in the Economic Module within MERIT rather than the Business Behaviours Module. The same may apply for road based freight transport, but the overall effect of road disruption (which goes beyond just freight movements to include the inability of staff and customers to travel) needs further exploration.

The case studies also revealed a number of potential business behaviour ‘levers’ to include within MERIT. For example, our case studies emphasised the importance of the Earthquake Support Subsidy to the Canterbury recovery. The provision (or not) of government support for businesses during recovery will be an important lever to include in future versions of MERIT. There is also potential to include an on/off variable for availability of mitigation measures for particular types of infrastructure: that is if mitigation is in place, there would be no disruption to operability.

The case studies support the case for including the potential for sustained productivity improvements following disaster events within MERIT, but there is a need to better understand the extent (percent) and nature (production recipe, restructuring, staff working hours) of these productivity gains before they can be incorporated.

The next step in refining the Business Behaviours Module is to carry out the further analysis suggested in this report and implement the recommended changes. Further testing and validation will then take place when a testing version of MERIT, incorporating the Business Behaviours Module, is available in early 2016.