Using electronic card transac�on data to measure and monitor regional tourism in New Zealand Rebecca Burson, Senior Analyst, MBIE Peter Ellis, Sector Performance Team Manager, MBIE
Tourism Data Domain Plan Comprehensive review of tourism data and insight in New Zealand Five key topics of interest: 1. The value of tourism to New Zealand 2. The growth, innova�on, produc�vity and efficiency of tourism businesses in New Zealand 3. The value of government interven�ons 4. The global compe��veness of New Zealand’s tourism industry 5. The sustainability of New Zealand tourism Par�cular drive for strong regional insight.
Regional Tourism Indicators: Method • Electronic card transac�on data from Paymark (interna�onal sample) and Bank of New Zealand (domes�c sample). • Card-present spending only. No cash, internet purchases, or other methods of payment. • Monthly aggregated spend and transac�on counts by cardholder origin, merchant loca�on, and industry classifica�on (ANZSIC-06)
Regional Tourism Indicators: Method • Extract Transform Load (ETL) processes classify data into local, domes�c, and interna�onal data sets • Domes�c / local split of data based on geographic rela�onship between cardholder and merchant • Spending within a territorial authority (TA) or across a ‘fluid’ TA boundary is considered local. All other types of spending between TAs is considered domes�c tourism. • ETL also provides addi�onal geographic and industry classifica�ons (e.g., Regional Tourism Organisa�ons, country groupings) • Spend data are converted to indexes (100 = average month in 2008) before analysis or publica�on
Regional Tourism Indicators: Products • MBIE publish updates to Domestic 150 RTIs every month 125 • Na�onal and subna�onal Index (100 = average spend in 2008) 100 insights into state of tourism expenditure 75 • Products include 50 International commentary, graphs and 150 tables 125 100 75 50 2008 2010 2012 2014 Figure 1: Domes�c and interna�onal indexes (dashed lines) and 12-month rolling indexes (solid line). Source: MBIE, Regional Tourism Indicators
Regional Tourism Indicators: Products • Variety of products suitable for basic and advanced users • Pivot tables provide detailed insights for specific des�na�ons and markets/industry groups
Method: Regional Tourism Es�mates Leverage the RTIs to produce es�mates of total spending (regardless of payment method) using Itera�ve Propor�onal Fi�ng (IPF) • Uses Tourism Satellite Account (TSA, Sta�s�cs New Zealand) and the Interna�onal Visitor Survey (IVS, MBIE) as source of total spending by product and country respec�vely • RTI data are ‘raked’ so that their marginal totals equal totals from IVS (interna�onal only) and TSA
Method: Regional Tourism Es�mates A basic example of IPF (‘raking’) 7 5 12 15 3 11 14 8 10 16 26 10 13 23 Table 1: Seed data Table 2: Target totals • Goal is to adjust seed cells so that marginal totals align with target totals • Itera�ve process involving weigh�ng to row totals and column totals
tourism-research-data/regional-tourism-estimates/about-the-rtes http://www.med.govt.nz/sectors-industries/tourism/ Method: Regional Tourism Es�mates Method has some assump�ons and limita�ons: • Excludes air passenger transport and educa�on spending • Method contols for different propensi�es to use cards for different products and by people from different countries (but not interac�ons of these propensi�es). • Assumes no within-New Zealand regional differences in propensity to use cards Detailed methods can be found here:
Regional Tourism Es�mates: Products • MBIE publish updates to RTEs around November each year, a�er the release of the Tourism Satellite Account • Products include pivot table, regional summaries, and interac�ve graphs
Regional Tourism Es�mates: Products Auckland Otago Canterbury Wellington Bay of Plenty Accommodation Waikato Food and beverage Northland serving services Other passenger West Coast transport Southland Other tourism Hawke's Bay products Retail sales - fuel Marlborough and other automotive Manawatu-Wanganui Retail sales - other Tasman Nelson Domestic International Taranaki Gisborne 2 1 0 1 2 Spend for year ending March 2013 ($ billion) Figure 2: Es�mates of interna�onal and domes�c visitor spending by Region and product. Source: MBIE, Regional Tourism Es�mates
So�ware and Capabili�es • Transi�oned from Microso� Excel and SPSS to R for analy�cal work • Established database for centalised storage and analysis of data sets • Developed custom programmes and func�ons to automate produc�on of standard products with less human interven�on
Policy and Business Applica�on • Performance indicators and lobbying tools • Regional Economic Ac�vity Report, interna�onal connectedness • Key performance indicators and evidence for tourism agencies • Assessing market shocks • Christchurch Earthquakes • Rugby World Cup 2011 • Gaining market insights • Emerging and declining markets • Distribu�on of market spending
Market Shocks: Christchurch Earthquakes 100 Seasonally adjusted spend index (100 = average spend in 2008) 80 Local 60 Domestic International 2010 2012 2014 Figure 3: Domes�c, interna�onal, and local spend in Christchurch City. Source: MBIE, Regional Tourism Indicators
Market Shocks: Rugby World Cup Accommodation Cultural and recreational services 160 140 200 Index (100 = average spend in 2008) 120 150 100 100 80 Food and beverage services Other retailing 140 140 120 120 100 100 80 2010 2011 2012 2013 2010 2011 2012 2013 Figure 4: Interna�onal seasonally adjusted indexes (solid black line) and raw indexes (points) for selected industries in Auckland Region. Source: MBIE, Regional Tourism Indicators
Market Insights United Kingdom United States Average growth (Per cent per year) 15% 10% 5% 0% -5% -10% -15% Figure 5: Growth in electronic card spending for UK and US visitors by region. Source: MBIE, Regional Tourism Indicators
Future Direc�ons • Review of RTIs in 2015 • Methodology improvement • Defini�on of domes�c tourism in RTIs • Imputa�on of non-paymark merchants for interna�onal RTIs • Product expansion • Seasonal adjustment of RTIs • Infla�on adjustment of RTIs
Ques�ons?
Recommend
More recommend
