Issue 3

It is widely believed that fluctuations in regional employment and economic growth can be reduced by increasing the diversity of a region's economic base. Several approaches have been developed to measure regional economic diversity. Most of these techniques while providing a measure of diversity do not link this measure to changes in the economic performance of the reg ion. An exception to this is provided by the portfolio selection framework which explicitly links the concept of industrial diversity to a measure of regional efficiency. This framework can be used to study the implications of diversification policies. In this paper a portfolio selection model is applied to employment data for Queensland. The framework is used to provide an insight into how changes in the economic base of the State are subject to growth and instability trade-offs. Interestingly, the results suggest that greater stability may be achieved through greater regional specialisation.

BERNARD TRENDLE

Page Number - 263

A structural VAR model is used to identify the sources and effects of shocks on the Victorian economy. The two identifying restrictions in the model are that demand shocks are assumed to have no long-run effect on the level of output and unemployment, and that supply shocks are assumed to have no long-run effect on unemployment, but may have a long-run effect on output. Three types of shocks are identified: a terms of trade shock; a permanent 'supply' shock, of which output is the proxy variable; and a temporary ' demand ' shock, of which unemployment is the proxy variable. The three-variable model is applied to the State level in Australia, using Victorian quarterly data on Gross State Product and unemployment rate, and the Australian terms of trade over the period 1984-1997. The econometric framework is designed to simultaneously identify the effects of these shocks on output, demand and supply. Terms of trade shocks were found to produce major fluctuations in the Victorian economy. Results from the model suggest that terms of trade shocks can account for 41 per cent of variation in the change in unemployment and 62 per cent of the variation in growth in real output.

ROWAN J. O'HAGAN

Page Number - 279

Industrial regions facing economic restructuring commonly generate collaboration processes between organisations from the public sector, the private sector and the civic society. Economic theory offers little explanation of the collaboration processes underpinning coordination of projects and strategies for addressing regional economic imbalance. These processes can be analysed by focusing on the 'network capital' of regions, but to do that a specific methodology of network analysis has to be applied. This paper addresses the question of what is the best approach to analysing regional networks for economic development. The paper argues that the best approach is to work at two levels, the organic and the functional. Organic analysis focuses on the structure of communication and collaborative links amongst network me mb ers and on the position of organisations within the network. Functional analysis focuses on network impacts on regional development and the factors behind the network performance. The purpose of this paper is to argue that an organic analysis is a necessary step of the methodology for analysing the network capital of a region. An organic analysis of the Hunter regional network for economic development is presented as an example.

M. CRISTINA MARTÍNEZ-FERNÁNDEZ

Page Number - 297

This paper applies duality theory to a generalised Von Thünen model of location. Generalised Von Thünen models are defined as those location models where there is economic attraction in location to some central place but competition for central location leads to decentralising forces in location through the price of goods fixed in location. The paper extends duality analysis within the location paradigm to new dual problems of the direct choice of location to minimise or maximise location relative to central place. The paper finds that the optimal value functions associated with these problems are potentially useful avenues for comparative static analysis in location. Further, the dual approach enables the identification of alternative conditional demand functions for location fixed goods which leads to a potentially useful decomposition in the examination of the influence of location on the demand for such goods.

R. HAM

Page Number - 313

This paper examines the role, growth and functions of Australia's regional cities at the 1996 Census. Previous research (Beer, Bolam and Maude , 1994a and b; Beer and Maude , 1995) highlighted a significant development within the evolution of Australia's urban system, with Australia's regional cities growing much more rapidly in the period 1976 to 1991 than the nation as a whole. These centres - defined as having populations greater than 10,000 persons - increased in number, population and workforce. They also became more specialised in their economic activities, with the fastest growth recorded amongst centres specialised in the tourism, recreation and finance sectors. This paper presents an update of that earlier work through an examination of relevant Census data for the period 1991 -96. It shows that regional cities have continued to grow at a rate above the national average both with respect to population and workforce. Importantly there has been a change in the direction of growth. When compared with earlier inter-censal periods, cities with economies based around tourism, recreation and leisure industries have secured a far greater share of growth within this group of centres.

ANDREW BEER

Page Number - 329

Eligibility for the AUSTUDY scheme is tested on the basis of a household ' s income and asset levels. Under the asset test in place in 1995 farm assets could be discounted by 50 per cent. The asset test has been criticised as it excludes 'asset rich but income poor' farm households from assistance, and so reduces the educational opportunities of some farm children. Rural and farm family lobby groups have called for the abolition of the asset test. In the 1996 Federal election campaign the incoming Australian Government promised to lift the discount rate allowed on farm assets from 50 per cent to 75 per cent, a promise they did not deliver on. The purpose in this study is to consider the effect this promise would have had, had it been implemented, in terms of the number of households that would have become eligible for AUSTUDY . Head count ratios are used to assess the number of households affected by different levels of assets and income used for AUSTUDY eligibility. It was found that 22 per cent of the surveyed farm households would have been denied AUSTUDY in 199 4/95 solely on the basis of their asset levels. Raising the discount level to 75 per cent would have made 65 per cent of those surveyed households previously denied assistance, eligible for AUSTUDY.

JOHN A. GARDINER, BRIAN DAVIDSON

Page Number - 349

A multi-regional computable general equilibrium model is used to investigate both the economy-wide and regional consequences of the fringe benefits tax in the long-run. The model is a hybrid “top-down" / "bottom-up" model: economic activity is modelled in two regions (Western Australia and the Rest of Australia) with results at the national level being an aggregation of the results for these two regions. Top-down modelling is then undertaken of sub-state regions within Western Australia. However, a significant share of the economic activity of one sub-state region - the PiIbara - is modelled within the bottom-up component of the mod el. The paper focuses on results at three regional levels: national, state, and sub-state. It is found that this Commonwealth policy, despite having relatively uninteresting consequences at the macroeconomic level, nevertheless has a significant impact on the distribution of economic activity across the country. Two features of the modelling are important in elucidating the regional consequences of the Commonwealth policy: the modelling of the policy responses at the state government level, and detailed modelling of regional cost and sales differences.

JAMES A. D. GIESECKE

Page Number - 365

The location of new infrastructure is often a matter of interregional competition. In a multiregional context where there exist linkages between regions a comprehensive resolution of the specific regional location of a new facility may not clearly favour the most directly obvious region. In this paper, recent results on inverse matrix decomposition and structural path analysis are employed to examine this issue. The analysis demonstrates that the ability of an infrastructural development to strengthen a region's internal intersectoral multiplier is not necessarily related to the proximity of the infrastructure but depends crucially on links between the infrastructure and the region in which it is located as well as on regional trading linkages. An example which looks at the effects of alternative locations of new infrastructure is used, firstly, to demonstrate that the ultimate effects of locational development are complex and may produce surprising results and, secondly, to highlight the types of relationships which, in a multiregional input-output modelling context, hold the key to provision of a quantitative solution to the effects-of­ location question.

RUSSEL J. COOPER

Page Number - 377

Balance relations, as inferred from the 'supply and demand pool' assumptions of Leontief and Strout, were used by Wilson in the 70 's as technological constraints on a probabilistic multi-regional model of the flow of commodities. The mode l, based on entropy maximization, was also constrained to reproduce the observed average travel cost per unit of commodity shipped, integrating the technology and gravity effects in the flow determination. In this paper, the Leontief-Strout approach is interpreted in terms of expected data avail ab ili ty. Although such arguments mirror their supply and demand pool assumptions, their own balance relations turn out to be partially inconsistent with these assumptions. The first task of the paper is the development of a consistent set of balance relations. Secondly, these corrected balance relations are used in a new entropy maximization framework, which also allows for regional interdependencies by introducing logistic constraints on regional input capacity. Further extensions of the approach lead to the generation of probabilistic supply functions as tools within a potential CGE analysis. This option requires the introduction of prices, permitting a profit constraint to replace the simple transport cost constraint of the earlier model.

JOHN R. ROY

Page Number - 393