«Tien Duc Pham, Geoff Bailey and Ray Spurr Acknowledgments The authors would like to thank Professor Philip Adams at Monash University and Dr. Hom ...»
The positive and negative effects
of the mining boom – A technical
Tien Duc Pham, Geoff Bailey and Ray Spurr
The authors would like to thank Professor Philip Adams at Monash University and
Dr. Hom Pant at the Australian Bureau of Agricultural and Resource Economics and Sciences
for their technical advice during this project.
Tien Duc Pham (Tourism Research Australia)
Geoff Bailey (Tourism Research Australia)
Ray Spurr (University of New South Wales) Editing Darlene Silec (Tourism Research Australia) ISBN 978-1-922106-98-8 (PDF) 978-1-922106-99-5 (Word) Tourism Research Australia Department of Resources, Energy and Tourism GPO Box 1564 Canberra ACT 2601 ABN 46 252 861 927 Email: firstname.lastname@example.org Web: www.tra.gov.au Publication date: June 2013 This work is licensed under a Creative Commons Attribution 3.0 Australia licence. To the extent that copyright subsists in third party quotes and diagrams it remains with the original owner and permission may be required to reuse the material.
This work should be attributed as The positive and negative effects of the mining boom – A technical paper, Tourism Research Australia, Canberra.
Inquiries regarding the licence and any use of work by Tourism Research Australia are welcome at email@example.com 2 Introduction This paper supplements the previous TRA mining boom report (Pham, Bailey and Marshall,
2013) by elaborating on the technical aspects of the modelling involved in the report. It is intended to facilitate a clearer understanding of the drivers of the modelling results reflected in those reports. It provides a more detailed explanation of the assumptions used in the modelling and the analysis of results than that provided in the previous reports.
From a policy perspective, the objective of the modelling was to analyse potential impacts, both positive and negative, of the mining boom on the economies and tourism sectors of Australia and its states and territories. The mining boom is defined as comprising the increases in mining exports from Queensland, Western Australia and the Northern Territory.
One special characteristic of the mining boom was an increase in Fly-In/Fly-Out (FIFO)1 employment in the mining areas, as the mining industry brought workers from outside regions to the mining areas. The consequence of FIFO was increased demand for air transport and accommodation, which had impacts on the local tourism sectors. The additional demand generated by FIFO employees adds a significant complication to the effects of the mining boom on tourism.
The modelling was devised to decompose the economic impacts of the mining boom through three stages. In the first stage, the modelling focuses on a base case that captures the mining boom with some realistic conditions of FIFO and a constrained supply of accommodation and air transport services. In the second stage, the effect of FIFO on the economy and tourism sectors is measured more explicitly. In the final stage, the modelling measures the potential effects when supply of both air transport and accommodation sectors increases capacity to the extent necessary to serve the FIFO demand.
The Computable General Equilibrium (CGE) tourism model used in preparing these reports is explained in greater detail in Appendix 1 (see also Pham and Dwyer in Tisdell, 2013). It is important to note that the modelling version adopted, MMRF-Tour, captures only nonbusiness tourism for interstate, intrastate, inbound visitors, and outbound travellers. Business tourism is embedded within the individual cost structures of the industries within the model database. For the purposes of the modelling, however, the demand generated by the mining FIFO employees is separately identified and modelled specifically as increases in usage of air transport and accommodation by the mining sectors.
A base case scenario is used to reflect the impacts of the mining boom and its associated FIFO activity on the economy generally and on the tourism sectors specifically. Two 1 Drive-in/Drive-out (DIDO) is also included but for simplicity FIFO is used for both.
3 additional simulations are used in conjunction with the base case to measure the effects of FIFO and the effects of additional investment injected into accommodation and air transport to satisfy FIFO demand. The first simulation captures the effect of high demand for accommodation and air transport required by the mining sector for FIFO activity under existing supply constraints applying in these two sectors. The accommodation and air transport sectors are assumed to be unable to respond quickly enough to the rapid increase in demand. This reflects the reality of a rapid expansion of FIFO demand crowding out the tourism services (in this case aviation and accommodation) available to service leisure tourism activity more generally in the affected regions.
The second simulation investigates the impacts when accommodation and air transport sectors have responded fully to the additional FIFO demand, expanding their services level by increasing their investment to build up capital stocks. This step was aimed at measuring the benefits from the spill-over effect of increased supply in the accommodation and air transport sectors on leisure tourism.
The mining boom dates back to 2005. There was a subdued period during the global financial crisis, before it picked up again over the period 2010–12. The model database was for 2004– 05, suitable for the impacts of the mining boom to be assessed on an average annual basis over the period 2004–05 to 2011–12.
The mining boom was mainly driven by strong demand for coal, iron ore and other nonferrous ores from overseas countries such as China and India. Because the existing model database contains separate data for only coal, oil and gas—and all other mining outputs are aggregated into a single industry defined as “other mining”—, the modelling explicitly examined coal exports from Queensland and exports of other mining from Western Australia and Northern Territory. The inclusion of Northern Territory is to ensure that the relative importance of mining in Northern Territory economy was factored into the analysis, even though total exports of other mining from Northern Territory are no larger than those from some other states.
Over the period 2004–05 to 2011–12, black coal exports from Queensland were estimated to increase by 5% on average, while other mining is estimated to have increased by 12% from Western Australia2 and 10% from Northern Territory3.
Increases in demand for accommodation and air transport generated by FIFO activity in Queensland and Western Australia were separately estimated for the purpose of the modelling. Drawing on comparison of Input-Output data on input usage by mining sectors between 2004–05 and 2008–09 (ABS, 2008 and 2012), it was assumed that both black coal and other mining in Queensland and Western Australia increased their demand for both 2 We acknowledge the assistance of the Bureau of Resources and Energy Economics in the development of these estimates.
3 The assistance of Tourism Northern Territory in the development of these estimates is acknowledged.
4 accommodation and air transport services by 300%, while other mining in Northern Territory increased its demand for accommodation and air transport by 100%. The lower increase by FIFO in Northern Territory reflects its already well-established reliance on FIFOs to source its mining workforce at the time that the mining boom gained momentum in the mid-2000s.
The modelling steps The MMRF-Tour model was used in a long-run comparative static mode to assess the economic impacts of the mining boom. Results of a comparative static simulation represent the impacts on an annual average basis.
The modelling consists of three simulations which are described in Figure 1 with corresponding shocks for each simulation. The first simulation examines the effects of the mining boom alone. Shocks to exports of coal and other mining in the three mining states determine the requirement of inputs used by the mining sectors, including labour, capital and intermediate inputs. However, the structure of the model could only raise the demand for accommodation and air transport in the same proportion as changes in mining outputs. The demand for accommodation and air transport in the presence of FIFO is not captured by the theory of the model. Therefore, without a direct control of the demand shocks to reflect FIFO in Simulation 1, the demand for accommodation and air transport services in this first simulation represents a case of a mining boom without FIFO. Consequently, the output levels of accommodation and air transport services are used as the benchmark level without the FIFO effect.
Simulation 2 presents the actual mining boom situation in the three states where demands for accommodation and air transport services are also assumed to increase strongly due to FIFO demand. The supply of these services is kept at the benchmark level in Simulation 1 to reflect the reality that these sectors could not expand their capacity quickly enough to satisfy FIFO demand. Results of Simulation 2 are presented as the base case result for the mining boom impacts.
In Simulation 3, the assumption of constrained output levels of accommodation and air transport is relaxed. These two sectors are now assumed to respond fully to the demand imposed by FIFO activity by increasing investment to expand their capacity. While the actual extent to which these sectors expand capacity to meet the increase in FIFO may not be the same as the level depicted by the model, the results in Simulation 3 indicate the potential impacts when the two sectors have responded fully to the new market conditions.
Simulation 1 (1) Mining export shocks This determines output levels of accommodation and air transport services Simulation 2 (1) Mining export shocks (2) Additional FIFO demand shocks for accommodation and air transport (3) Re-use output levels of accommodation and air transport from Simulation 1 Simulation 3 (1) Mining export shocks (2) The same additional FIFO demand shocks for accommodation and air transport as in Simulation 2 While results of Simulation 2 are used to portray the impacts of the mining boom on the regional (state) economies and tourism sectors, the difference in results between Simulation 2 and Simulation 1 is used to show the specific impacts of FIFO activity on the tourism sectors and state economies. Similarly, the difference in results between Simulation 3 and Simulation 2 is used to show the potential impacts of investment in accommodation and air transport (as a result of expansion of their supply capacity) on the state economies and tourism sectors.
Results of Simulation 2 are presented as annual average percentage changes in a typical year in the timeframe 2004–05 to 2011–12. In the following sections, explanation and analysis of Simulation 2 are presented first.
The impacts of FIFO and of capacity expansion of accommodation and air transport services on the regional economies and tourism sectors will be presented as percentage point changes, i.e. the difference in result between two associate simulations.
Simulation closures As is usual in CGE modelling, the results of the modelling are influenced by the choice of closure, in which variables are set either exogenously or endogenously. While results of endogenous variables are determined by the model, exogenous variables are often set at zero to reflect the absence of any changes before and after the shocks.
In all three simulations, the model was set in a long run environment. In a typical long run comparative static simulation, the supply of capital is assumed to be perfectly elastic. This implies that capital is not constrained by the supply from the domestic source only. Capital could also be supplied from the rest of the world at a given rate of return determined outside of the economy.
6 For the labour market, it is assumed that national employment is fixed while the real wage adjusts to clear the labour market. This implies that for a given level of labour supply in the economy, all of those who wish to work will be willing to accept a market real wage rate in order to find a job. This real wage rate is from the consumer point of view, i.e. the wage rate receivable by workers after being adjusted by the consumer price index. Although employment at the national level is fixed, employment at the state level is not. The supply of employment in a state will be higher than the national level of employment if the real wage rate in the state rises above the real national wage rate, implying an upward sloping labour supply curve.
The setting up of employment and capital markets determines real Gross Domestic Product (GDP) through changes in capital stocks.
For the demand side, real total household consumption is determined by the disposable income subject to a constant average propensity to consume. Real investment is assumed to move in line with movements of capital stock for every industry. Nominal government consumption at the state level is driven by nominal GSP and, similarly, nominal government consumption at the Federal level is driven by nominal GDP. Given the fact that GDP is already determined by the supply side, specification of those components in the demand side of GDP leaves the net balance of trade as a residual determined by the model.
Throughout all three simulations, the Consumer Price Index (CPI) is held constant and used as a numeraire.
As results here are from a comparative static model, they are not directly compatible with observation of historical data. They are intended to give annual average changes from the identified shocks and, most importantly, they indicate the patterns and direction of the impacts.