MCU14/0106 & OPW15/0001 Appendix B - Retail Needs Response

4.1.2 URBIS METHODOLOGY

Urbis acknowledges that directly applying the Quantium distribution of trade metric to derive the market
share at Noosa Civic has the potential to be biased. Residents that are closer to the property are more
likely to do smaller more frequent cash transactions whereas residents further away may do larger less
frequent electronic transactions, biasing the distribution of spending away from the primary sector.

To account for this, the following market share methodology is applied with different approaches for
Noosa Civic and the other centres in the network due to the availability of superior data for Noosa Civic.

Noosa Civic
To assess the market shares and ensure no cash bias four approaches were assessed:

Ie Distribution of Spend approach. As detailed in the report, this approach uses Quantium data
and distributes turnover across the trade area sectors and beyond at the tenant level. The
distributed turnover is then divided by the market, as defined by Urbis, to attain market shares.

2. Market Share Approach. This approach accepts the Quantium market shares at the product
group level, which implicitly assumes that the share of ‘plastics’ is reflective of the share of total
wallet (e.g. if the average primary trade area residents spends x% of their electronic spend at
Noosa Civic, then they are assumed to spend the same x% of their total spend at the property
(i.e. including cash). This eliminates the potential bias in the distribution approach.

3. Balance against previous surveys: This approach follows the first but uses the 2012 exit
survey in place of the Quantium data. Noting the bias in exit surveys towards more frequent
everyday use customers.

4. Balance against total market: Combined with the analysis of competing centres as well as
inflows (tourist) and outflows (escape spending to Brisbane etc.), the total trade area market was
balanced (i.e. where does each dollar spent by each trade area resident end up). This was
undertaken to assess for any bias in the Quantium data that would be influencing market shares
and hence the impact analysis.

Approaches 1 and 2 resulted in a generally consistent market share for Noosa Civic, suggesting
that any cash bias is minimal. Furthermore, upon balancing the inflows and outflows of the
market (approach 4), the market share achieved by the centres and indeed the supermarkets
were found to be appropriate.

Finally, the 2012 exit survey was shown to have a lower share of turnover from beyond the trade
area. This is typical of a comparison between an exit survey and the Quantium turnover
distribution, due to the local bias of an exit survey. However, given the alignment of the other
three approaches, which include an assessment of the size of the tourist market (i.e. beyond
spending), the Quantium distribution approach is considered to be robust.

Other Centres

The key mechanism to derive the impact of the Noosa Civic expansion on other centres in the network is
the market share achieved by other centres. To derive this, Urbis initially used the Quantium market
shares by post code, as provided in Table F.1 of the DA report (i.e. the market share retailers in each
post code achieved from residents of each trade area sector, by product group) and then made
adjustments based on productivity, performance, and internal model consistency.

To assess the market shares and move the Quantium market shares from the postcode level to specific
centres, the following methodology was undertaken:

1. At a postcode level, derive the total turnover achieved by all retailers in that postcode, at a
product category level. This was done by multiplying out the Quantium Share of Electronic
Spend, by category, by the total market as defined in the report.

URBIS.
NOOSA RESPONSE REPORT - RETAIL ECONOMICS JUNE 2015 V2 TRANSACTION DATA 1 3