Annex 4: Econometric analysis of online and onsite survey data – key assumptions
Visitor expenditure data is then used to estimate GVA using the Cambridge Model which produces the final economic outputs: (i) business turnover, (ii) associated employment supported by expenditure and (iii) GVA estimates.
The estimation of GVA for surveyed sites is based on a number of key assumptions. These assumptions have been made using a combination of tourism-related information and informed judgement where necessary. The following summarises the key points.
A-4.1 Interpreting ‘visits’ to sites
In general, estimates of visitor numbers and changes in visitor numbers should be interpreted as ‘visits to the coastline’ rather than actual beach users. In particular, the level of supporting data required to distinguish beach visitors from broader definitions is not available for the majority of sites. Each surveyed site has been considered individually with supporting information and survey data to produce a ‘best estimate’ for annual visitor numbers. Estimates are based on car park capacities in the vicinity of the sites to which survey data has been applied. Adjustments have been made using Scottish coastal seasonality data (Great Britain Tourism Survey and Great Britain Day Visits Survey) to account for the potential impacts of the timing of the fieldwork on the annual outputs provided. The outputs have been broadly sense checked using population data, accommodation stocks and when available tourism volume and value estimates.
A-4.2 Attributing overnight visits to sites
A default assumption that is made is to attribute the entire length of an overnight visitor’s stay and their expenditure to the local site. This implicitly assumes that the coastal location is the main motivation for visits, and if the quality of the coastal location were to deteriorate (i.e. de-designation of the bathing water), the whole visit would be lost rather than a reduced length of stay. Whilst this approach potentially over-estimates the economic impact of staying visitors - as typically a proportion of the visit and expenditure could be spent away from the site (so-called ‘leakage’) - no information is available to consistently estimate ‘leakage’ of economic activity across sites. The approach taken therefore, to attribute the entire economic impact of an overnight visitor to the local site, ensures that all sites are treated consistently thereby avoiding further site-level ad-hoc assumptions.
The issue of attribution also ties in with the ‘substitutability’ of visits to bathing waters and recreational sites more generally. The substitution effect between different sites does not feature in the scope of the local level analysis. The focus is on the implications of a reduction in visitor numbers at a given site, which constitutes a decline in economic activity at the local level rather than a potential displacement towards alternative bathing water sites (or alternative recreation sites). This implies that the local level results should not be aggregated at larger scales (e.g. national or river basin district scales) because they do not account for substitutability between sites whereby changes in expenditure in one location are likely, for the most part, to be offset by a re-distribution of expenditure to other locations. Potential substitution effects are, however, controlled for in the revealed preference analysis.
There is a problem
Thanks for your feedback