|
|
Tourism Statistics | news |
 |
| |
AUSTRALIAN TOURISM SATELLITE ACCOUNTS (ATSA)
Briefing by the Australian Bureau of Statistics
____________________________________________________
Introduction
The most recent release of the Australian National Accounts: Tourism Satellite Account (ABS cat no. 5249.0) has highlighted some concerns about the nature and purpose of the Australian Tourism Satellite Account (ATSA), and in turn the limitations of the ATSA.
2 The objective of this paper is to provide information about the history, methodology, data limitations and revisions process for ATSA, and also notes likely future changes to ATSA.
3 The paper forms part of the ABS strategy for improving communication with users of ATSA.
Background
4 In March 1995, following representations from tourism industry representatives and government bodies, a committee of the House of Representatives recommended in a report into the impact of Australia's taxation regime on the tourism industry that:
"the Australian Bureau of Statistics .. be provided with the resources to develop a statistical system for quantifying the significance of tourism in Australia as a matter of priority"
5 In response to this request, the ABS worked with key stakeholders and concluded that the need could be best met by producing a tourism satellite account. Within the System of National Accounts (SNA), the notion of a satellite account arises where there is a need for information regarding views of the economy that are not adequately covered by standard statistical classifications or concepts.
6 Satellite accounts complement the core national accounts for selected areas of interest, by using relevant concepts and structures of the SNA. A tourism satellite account involves the identification and classification of tourism activities in a way that can be interpreted within the national accounting framework resulting in a comprehensive set of economic data in respect of tourism. In the Australian context the results of this process are characterised as being the value of economic activity and employment relating to the tourism industry.
7 Following a comprehensive consultation process the ABS, with the support of key stakeholders, decided to develop and release an ATSA. Australian National Account: Tourism Satellite Account 1997-98 (cat.no.5249.0), was the first ATSA and was released in October 2000. The concepts and methods used in the ATSA are based on international standards developed by a working group comprising the OECD, UN Statistical Division, EuroStat and the World Tourism Organisation. These standards are in turn largely compatible with the standards that underlie the national accounts.
8 Subsequently ATSA has been released annually with the most recent release on 11 May 2007, in respect of the 2005-2006 reference period. That release included significant revisions to the data, including significant downward revisions to the headline employment figure and upward revisions to tourism’s share of gross domestic product (GDP) and value added. The remainder of this paper explains the reasons for these revisions and notes that similar revisions may occur in the future.
9 The 2005-06 Australian ATSA outlines the tourism contribution to GVA and GDP and is broken up by tourism connected and tourism characteristic industries. It presents the % changes in contribution over the 2000-01 and 2003-04 which have been the two previous benchmark periods.
Table five in the ATSA shows that over the two benchmark periods, which were 2000-01 and 2003-04 there was a higher % change in contribution for tourism characteristic industries with a -2.4% change.
Tourism connected industries had a -0.7% change since 2000-01 period.
ATSA Methodology
10 In the main an industry is defined as consisting of all businesses that undertake similar or like processes of production. This means that the relationship between the outputs (goods and services) that they produce and the inputs (capital, labour and intermediate goods and services) they use are similar for each business in the industry. For example, the Agriculture industry might be distinguished from other industries by its extensive requirement for land.
11 Those businesses that are characterised as being part of the tourism industry on the basis that visitors consume the goods and services that they produce are involved in a range of diverse production processes. For example an airline requires a different mix of inputs (capital, labour etc) than is required by a hotel (their respective outputs are also dissimilar).
12 The Australian TSA distinguishes between products and industries that are related to tourism, and those which are not. Tourism related products and industries are further classified into tourism characteristic and tourism connected.
13 Tourism connected industries are those, other than tourism characteristic industries, for which a tourism related product is directly identifiable (primary) to, and where the products are consumed by visitors in volumes which are significant for the visitor and/or the producer. All other industries are classified as non-tourism industries, though some of their products may be consumed by visitors and are included in the calculation of tourism gross value added and tourism GDP.
14 Tourism characteristic industries are defined as those industries that would either cease to exist in their present form, or could be significantly affected if tourism were to cease. In the Australian TSA, for an industry to be 'characteristic', at least 5 per cent of its output must be consumed by visitors. Whether or not an industry is classified as characteristic has no effect on total value added resulting from tourism, as the TSA measures the gross value added resulting from the production of products directly consumed by visitors, not the total gross value added generated by tourism related industries.
15 Simply put, the estimates for the tourism industry reflect the contribution of businesses, from across a range of conventional industries that serve visitors. As well as satisfying the requirements of visitors, these businesses also serve their local (non-visitor) clientele and thus there is a need to determine what proportion of their activity relates to the servicing of visitors. Put another way, tourism is defined from the demand side rather than the supply side, ie by type of customer rather than by type of product produced.
16 The primary objective of the ATSA is to measure the value added and contribution to gross domestic product (GDP) that occurs as a result of expenditures on goods and services by visitors. These measures are described as tourism value added and tourism GDP respectively. It is the activity of visitors that is of interest, and a visitor is defined as ’any person travelling to a place other than that of his/her usual environment for less than twelve months and whose main purpose of trip is other than the exercise of an activity remunerated from within the place visited’.
17 Importantly for the purposes of the ATSA, "tourism" is defined to include visitors whose main purpose of visiting is for business, in addition to those visitors for the purpose of leisure and personal reasons. Overseas students who are deemed to be short term in Australia are also included.
18 The first step in the compilation of the ATSA is to estimate the value of tourism consumption by product, using mainly information sourced from the National Visitor Survey and the International Visitor Survey conducted by Tourism Research Australia. Figure 1 shows the flow of tourism consumption through the Australian economy. Attachment A details the main data sources used in the compilation of tourism consumption estimates.
Figure 1
19 These estimates of visitors’ demand for each product are compared with the known aggregate (tourism and non-tourism) supply of these products. The aggregate numbers are sourced, in the main, from the supply and use tables that underlie the national accounts. In a number of cases it is necessary
to supplement the information from these supply and use tables in order to provide extra detail around those products and industries that are particularly significant in the tourism context.
20 This process establishes the share of supply for each product that is attributable to tourism. These share (of supply) estimates are used to estimate the value of output and value added attributable to each of the industries that produce the products which the visitors have consumed. Using these industry values, an estimate of the tourism share of the output and value added for each of the constituent industries is estimated. Figure 2 highlights this process, while figure 3 provides a numerical example to describe these relationships.
Figure 2 - TSA benchmark processes
Figure 3
21 It is important to recognise that this process is only undertaken once every three years and the resultant estimates are described as benchmarks. The three yearly process is done to ensure that the ATSA operates within the context of a recent assessment of the structure of the Australian economy.
22 In years for which these benchmark estimates are not compiled it is assumed that the industry share of value added etc established in the benchmark year remains constant. In update years (years for which the benchmarks are not compiled) the tourism consumption estimates available annually from NVS and IVS are converted to an industry basis, using the relationships established in the benchmark year, and assuming that the production relationships observed in the supply use tables hold constant, estimates of value added etc are derived in respect of the update year.
23 The detailed national accounts supply-use tables which are used in a benchmark year are a critical input into the benchmark, and any revisions to these tables generally have a significant impact on the tourism value added ratios, the main output of the benchmarking process. As such, movements and revisions to the National Accounts data will flow through and impact the ATSA estimates.
24 For the 2005-06 edition of the ATSA new benchmarks were established in respect of 2003-04. The supply-use tables used to derive this benchmark incorporated upward revisions to the national accounts estimates with the release on 7 November 2005 of Australian System of National Accounts 2004-05 (cat no. 5204.0). These revisions were due to the incorporation of improved data sources flowing from The New Tax System.
25 The revisions to the supply and use tables that flowed from the revisions to the national accounts meant that the tourism shares established for previous editions of the ATSA were replaced by a more contemporary set of information. Generally, the latest analysis indicated a reduced tourism share in most tourism industries. For example, for the tourism industry accommodation, the tourism share of this industry fell from 88.1% at the 2000-01 benchmark to 84.4% at the 2003-04 benchmark. Despite this, when revisions to the national accounts were fully incorporated into ATSA estimates this resulted in a small upward revision to the tourism share of gross value added and GDP estimates in the 2005-2006 edition of ATSA. For example, tourism gross value added for 2003-04 was published as $26,016m for the 2004-05 edition of the ATSA. When compared with total gross value added for the economy of $762,850m, this equated to a tourism share of 3.4%. However, for the 2005-06 edition of the ATSA tourism gross value added for 2003-04 was revised upward to $29,365m. When compared to the revised total gross value added for the economy of $764,791m, this corresponded to a revised tourism share of 3.8%.
26 While there were minimal revisions to employment estimates from the Labour Force Survey, these estimates are combined with Service Industries Surveys (SIS) data for some tourism industries to generate total employment for each tourism industry. Downward revisions to SIS employment estimates for some industries (in particular accommodation), combined with the application of the revised lower tourism industry shares translated to a significant downward revision to the tourism employment estimates. This emphasises an aspect of the methodology for modelling of the employment estimates that needs to be clearly understood in the use of these estimates.
27 Given the current strategy it can be expected that there will be revisions in future releases of the ATSA and the table at Attachment B presents details of planned future releases of the ATSA. As significant revisions are most likely to manifest in years when new benchmarks are introduced, it is apparent that the 2008-09 edition of the ATSA to be released in mid 2010 will be the next release where significant revisions may occur. Figures 4 and 5 show an example of how revisions are seen in published ATSA aggregates.
Figure 4
Figure 5
Using ATSA - Level and Movement Estimates
28 The nature of the ’tourism industry’ (ie it is not a standard industry but rather is determined by the status of the purchaser) means that the ABS is required to make assumptions and model a number of the key estimates as directly relevant data are not available. In brief the benchmark years represent
the best estimate at that point in time of the levels and the tourism ’share’ of industries. Estimates for the following two years are compiled on a comparable basis, and to the extent that the demand estimates are accurate and there has been no significant structural change in the ’tourism industry’, the estimates of movement will be accurate.
29 In the example below, the update years B and C were compiled on a comparable basis with the benchmark in Year A. In the next benchmark there is evidence of structural change with a fall in the accommodation industry tourism gross value added.
| Output Year D | Year A Benchmark | Year B Update | Year C Update | Year D Benchmark |
| Tourism consumption of accommodation (converted to basic prices) $m | 10000 | 10400 | 10600 | 11000 |
| Accommodation industry intermediate consumption $m | 3000 | 3120 | 3180 | 3135 |
| Accommodation industry gross value added $m | 7000 | 7280 | 7420 | 7865 |
| Tourism contribution to accommodation industry gross value added % | 85% | 85% | 85% | 83% |
| Accommodation industry Tourism GVA | 5950 | 6188.00 | 6307.00 | 6527.95 |
| | | | |
| Year on year change $ | | 238.00 | 119.00 | 220.95 |
| Year on year % change | | 4% | 2% | 4% |
30 In the following year, an update year, the supply benchmarks of Year D are applied to Year E. There were also upward revisions to the consumption estimates and while the levels have been revised the percentage movements between the years are comparable to those published in Year D.
| Output Year E | | | | | |
| If upward revisions to consumption estimates occur | Year A Benchmark | Year B Update | Year C Update | Year D Benchmark | Year E Update |
| Tourism consumption of accommodation (converted to basic prices) $m | 10050 | 10500 | 10800 | 11250 | 11400 |
| Accommodation industry intermediate consumption $m | 3000 | 3150 | 3240 | 3206.25 | 3249 |
| Accommodation industry gross value added $m | 7050 | 7350 | 7560 | 8043.75 | 8151 |
| Tourism contribution to accommodation industry gross value added % | 85% | 85% | 85% | 83% | 83% |
| Accommodation industry Tourism GVA | 5992.5 | 6247.50 | 6426.00 | 6676.31 | 6765.33 |
| | | | | |
| Year on year change $ | | 255.00 | 178.50 | 250.31 | 89.02 |
| Year on year % change | | 4% | 3% | 4% | 1% |
31 The issue for the statistician arises when the following benchmark produces relationships at significant variance to the previous benchmarks. The problem is whether to accept that there has been significant structural change in the intervening period, or whether one of the benchmark estimates is deficient.
32 If one accepts that the new benchmark reflects structural change since the last benchmark then it would be appropriate to interpolate the differences by revising the estimates for the intervening periods between the respective benchmarks. However, if it is considered that the differences do not reflect structural change in the intervening years then the decision has to be made as to which benchmark is the most appropriate and the levels of the time series adjusted accordingly.
33 To continue the example, in Year G, there are further upward revisions to the consumption estimates and evidence in the benchmark of structural change from Year E. The accommodation industry tourism gross value added are all revised due to the changes in demand and supply but the year on year percentage movements are consistent with those published in earlier years.
| Output Year G | | | | | | |
| Revisions to consumption estimates and evidence of structural change occurring from Year E | Year A Benchmark | Year B Update | Year C Update | Year D Benchmark | Year E Update | Year G Benchmark |
| Tourism consumption of accommodation (converted to basic prices) $m | 10050 | 10550 | 10900 | 11400 | 11500 | 11450 |
| Accommodation industry intermediate consumption $m | 3000 | 3165 | 3270 | 3249 | 3162.5 | 3148.75 |
| Accommodation industry gross value added $m | 7050 | 7385 | 7630 | 8151 | 8337.5 | 8301.25 |
| Tourism contribution to accommodation industry gross value added % | 85% | 85% | 85% | 83% | 82% | 82% |
| Accommodation industry Tourism GVA | 5992.5 | 6277.25 | 6485.50 | 6765.33 | 6836.75 | 6807.03 |
| | | | | | |
| Year on year change $ | | 284.75 | 208.25 | 279.83 | 71.42 | -29.73 |
| Year on year % change | | 5% | 3% | 4% | 1% | 0% |
34 Short term movements in the headline economic indicators key estimates will be accurate to the extent that the demand estimates are accurate and there has been no significant structural change in the tourism industry. Where there is little variance between the different benchmarks then there will be only be minor revisions to the movement estimates. However if there is a significant variance in the benchmarks, and if there is no observable structural change, then there is likely to be little or no revision to the movement estimates. Rather there will be a level shift to ensure that previous levels align with the current benchmark estimates.
35 This is further explained in the following table of data extracted from Table 4 as published in ATSA 2004-05 and ATSA 2005-06. The level of tourism gross value added across all years shifted upwards as a result of the 2003-04 benchmark to incorporate the latest supply use estimates. Hence, users focussed on the level of tourism gross value added at the total and commodity level saw significant revisions between the 2 publications eg total tourism gross value added for 2004-05 was revised from $26479m to $29693m. By comparison the year on year percentage movements at the total and commodity levels between 2003-04 and 2004-05 are relatively unchanged between the 2 publications.
| ATSA 2004-05 Table 4 Tourism Industry Gross Value Added | | | | | |
| 2000-01 | 2001-02 | 2002-03 | 2003-04 | 2004-05 |
| Travel agency and tour operator services | 992 | 966 | 975 | 962 | 961 |
| Air and water transport | 3727 | 3592 | 3557 | 3521 | 3617 |
| Accommodation | 2775 | 2855 | 2917 | 2941 | 3115 |
| Cafes, restaurants and food outlets | 2501 | 2601 | 2689 | 2599 | 2644 |
| Total | 25044 | 25250 | 25939 | 26016 | 26479 |
| | | | | |
| Year on year movements | | | | | |
| Travel agency and tour operator services | | -26 | 9 | -13 | -1 |
| Air and water transport | | -135 | -35 | -36 | 96 |
| Accommodation | | 80 | 62 | 24 | 174 |
| Cafes, restaurants and food outlets | | 100 | 88 | -90 | 45 |
| Total | | 206 | 689 | 77 | 463 |
| | | | | |
| % year on year movements | | | | | |
| Travel agency and tour operator services | | -0.03 | 0.01 | -0.01 | 0.00 |
| Air and water transport | | -0.04 | -0.01 | -0.01 | 0.03 |
| Accommodation | | 0.03 | 0.02 | 0.01 | 0.06 |
| Cafes, restaurants and food outlets | | 0.04 | 0.03 | -0.03 | 0.02 |
| Total | | 0.01 | 0.03 | 0.00 | 0.02 |
| ATSA 2005-06 Table 4 Tourism Industry Gross Value Added (incorporating 2003-04 benchmark) | | | | | | |
| 2000-01 | 2001-02 | 2002-03 | 2003-04 | 2004-05 | 2005-06 |
| Travel agency and tour operator services | 1144 | 1110 | 1121 | 1267 | 1284 | 1508 |
| Air and water transport | 3473 | 3646 | 4064 | 4089 | 4229 | 4394 |
| Accommodation | 3421 | 3517 | 3641 | 3676 | 3838 | 4247 |
| Cafes, restaurants and food outlets | 2743 | 2845 | 2947 | 2942 | 2974 | 3080 |
| Total | 26920 | 27427 | 28552 | 29365 | 29693 | 31293 |
| | | | | | |
| Year on year movements | | | | | | |
| Travel agency and tour operator services | | -34 | 11 | 146 | 17 | 224 |
| Air and water transport | | 173 | 418 | 25 | 140 | 165 |
| Accommodation | | 96 | 124 | 35 | 162 | 409 |
| Cafes, restaurants and food outlets | | 102 | 102 | -5 | 32 | 106 |
| Total | | 507 | 1125 | 813 | 328 | 1600 |
| | | | | | |
| % year on year movements | | | | | | |
| Travel agency and tour operator services | | -0.03 | 0.01 | 0.13 | 0.01 | 0.17 |
| Air and water transport | | 0.05 | 0.11 | 0.01 | 0.03 | 0.04 |
| Accommodation | | 0.03 | 0.04 | 0.01 | 0.04 | 0.11 |
| Cafes, restaurants and food outlets | | 0.04 | 0.04 | 0.00 | 0.01 | 0.04 |
| Total | | 0.02 | 0.04 | 0.03 | 0.01 | 0.05 |
36 In the context of the above users are encouraged to focus on the short term year to year movements rather than the absolute level estimates.
International Standards - Developments and Implications
37 As noted earlier the concepts and methods used in the ATSA are based on international standards developed by a working group comprising the OECD, UN Statistical Division, EuroStat and the World Tourism Organisation (WTO) and in turn are largely compatible with the standards that underlie the national accounts.
38 When Australia produced its first tourism satellite account in 2000 for the reference period 1997-98 an agreed set of principles, practices and protocols for tourism satellite accounts had not been promulgated by any of the relevant international organisations. However, it was understood that guidelines to be released during 2000 by WTO would recommend that a direct relationship would
be deemed to exist between the visitor and all of the firms involved in the production and delivery of goods through a retail outlet.
39 To be consistent with what was expected to be the proposed international standard, ABS compiled the first Australian TSA on that basis, even though ABS input to the proposed guidelines recommended the methodology now being proposed by the WTO.
40 Subsequent to this decision by the ABS in early 2000, the first set of international recommendations on tourism satellite accounts was released. The international recommendations were changed late in the development cycle after the intervention of the OECD to allow two approaches to the interpretation and measurement of the concept of a direct relationship between the visitor and a firms output.
41 The first, which was the approach adopted by ABS to compile the first Australian TSA, is broad in nature in that it effectively deems a direct relationship exists between the visitor and all the firms involved in the production and delivery of goods through a retail outlet.
42 The second and narrower approach deems the direct relationship to only exist between the visitor and the retailer and only the output (ie the ’margin’) and value added of the retailer should be attributed to the tourism industry.
43 The UN/WTO released in May 2007, draft revised international TSA standards which recommended adoption of a single approach ie the narrower approach where only the output (ie the ’margin’) and value added of the retailer is attributed to the tourism industry.
44 The ABS agrees with and has always supported this approach. The impact on tourism gross value added is shown in the following table:
| Table 5 ATSA 2005-06 | Tourism Gross Value Added
2000-01 | Tourism Gross Value Added
2003-04 |
| Food manufacturing $m | 878 | 864 |
| Beverage manufacturing $m | 576 | 597 |
| Transport equipment manufacturing $m | 170 | 181 |
| Other manufacturing $m | 1,864 | 1,374 |
| Total contribution to TGVA of manufacturing industries $m | 3,488 | 3,016 |
| Total TGVA (all industries) as published $m | 26,920 | 29,365 |
| Revised Total if manufacturing industries are removed $m | 23,432 | 26,349 |
| Total industry GVA $m | 628,124 | 764,791 |
| Revised tourism contribution to industry GVA if manufacturing industries are removed % | 3.7 | 3.4 |
| % change to TGVA if manufacturing industries are removed | 13% | 10% |
The table assumes that none of the manufacturing industries have a direct relationship with visitors, instead visitors purchase manufactured goods from retailers and the retail margin is also included in tourism gross value added. The removal of the manufacturing industries from tourism gross value added using the results from Table 5 of ATSA 2005-06, reduces tourism gross value added by 13% in 2000-01 and 10% in 2003-04. A more precise analysis is only possible by closely examining available data and compiling the results through the ATSA compilation system.
45 The proposed approach is preferred in terms of consistency with the national accounts, and particularly for the purpose of comparison to other industries and understanding the economic impacts flowing from the direct interaction between visitors and businesses.
46 A key issue for users of ATSA will be the timing of the introduction of the new methodology if the recommendation is accepted; and the period for which ABS would present data on both bases. If these new standards are adopted internationally the ABS will implement the new methodology in respect of
2008-09 - to be published in May 2010, and will present data on both bases for a period of time. Note this will also be a benchmark year for ATSA and will also mark the introduction of the updated industry classification ie ANZSIC 2006.
47 In terms of impacts, it is likely that the adoption of these new standards will have the effect of reducing the direct value added, GDP and estimates of employment of tourism by about 15%. The ABS will ensure that users are well informed regarding these changes and appropriate communication plans will be put in place to ensure users’ needs are considered when this change is adopted.
Conclusion
48 Clearly there were communication issues between the ABS and users of the ATSA at the time of the 2005-06 TSA release. ABS has instituted a range of channels to improve communication in future, including:
· seminars and other forums will be the main avenue for discussion of issues
· a Newsletter will be introduced (first edition has already been issued and circulated)
· information sessions will be held at appropriate points in the release cycle
· public announcements will be made to the industry at appropriate times.
49 Enquiries about any issues raised in this paper may be made by contacting Melanie Taylor by email at melanie.taylor@abs.gov.au or by telephone on 02 6252 7495.
Australian Bureau of Statistics
January 2008
Attachment A
MAIN DATA SOURCES USED IN THE COMPILATION OF ATSA ESTIMATES
Attachment B
SCHEDULE OF ATSA RELEASES
| ATSA Edition | Timing of release | Comments |
| 2005-06 | May 2007 | Benchmark in respect of 2003-04 |
| 2006-07 | April 2008 | Update only (to include 2006-07 estimates) |
| 2007-08 | April 2009 | Update only (to include 2007-08 estimates) |
| 2008-09 | May/June 2010 | Update to include 2008-09 estimates
i. Benchmark in respect of 2006-07
ii. First time ATSA produced on ANZSIC06 basis
iii. Any changes in International Standards
are likely to be included in this edition also. |
| 2009-10 | April 2011 | Update only (to include
2009-10 estimates) |
| |
| |
|
|
|
|
|
|
Print Page
Print All