Conclusions and Prospects for Future Work on Asian Developing Economies

The main aspects of our survey of modelling of poverty in a CGE modelling framework can be summarized quickly. The first and second generation models included distributional questions, but did not address poverty explicitly. The third generation models do address the question of poverty reduction impact of SAPs explicitly. The main strength for poverty analysis purposes is the CGE models ability to capture the interdependence (the general equilibrium effects) in the economy. In the dual-dual formulation, the further incorporation of interdependence among the labor markets in the rural and urban sectors leads to a more realistic assessment of the poverty reduction impact of trade liberalization. However, with a few exceptions such as Azis (2002), the CGE models of the third generation still do not capture the structure of the financial markets and analyze the impact of financial liberalization on poverty reduction. There are no models in existence which try--- in the spirit of the dual-dual approach--- to reduce the dimensionality of the standard static CGE models and still retain the causal structure essential for analyzing the poverty reduction impact of financial liberalization. It will be highly desirable to attempt an exercise for financial liberalization analogous to that of Stifel and Thorbecke for trade liberalization for some representative Asian economies.

Broadly speaking, there are at least four categories of Asian economies that could be the subject of such research. First, we have low income countries of South Asia. Here, a model based on an economy such as Bangladesh could offer some insights. The second category would include middle income Asian Developing countries such as Indonesia. Here, both national and regional poverty analysis would be equally important. The third and fourth categories will include the transitional low and middle income economies. As emphasized before, the four categories are not exhaustive, but they do cover a large number including the most populous poor countries with a large number of poor people. Addressing the structural and institutional issues even in each of the four categories is beyond the scope of this paper. However, as a beginning, the case of Bangladesh can be discussed briefly here.

Bangladesh, along with India, Pakistan, PRC and Indonesia is one of the 'big five' in Asia in terms of both population and poverty. These five Asian countries comprise three fifths of the world’s population and two fifths of poor people. Bangladesh is a relatively homogeneous country with fairly reliable national and regional data sources. It has also been a major recipient of both program and project loans from the IFIs.

In Asian economies such as Bangladesh, the low level of income and pervasive poverty present a prima facie case for a policy focus on poverty reduction. However, there has only been mixed success in poverty reduction so far. One reason is that over the past few decades, the growth record has also been somewhat mixed. There have also been macroeconomic imbalances in the form of high fiscal deficit, low domestic savings, and sizable external account deficit. Consequently, both inflation and interest rates were high, making the economic environment unfriendly to adequate and accelerated investment. In the wake of the policy focus on macroeconomic balancing, the growth and poverty impacts of such policies are natural candidates for rigorous investigation.

A parsimonious financial CGE model for Bangladesh with built-in capability for poverty analysis can thus be the goal of the next phase of CGE modelling for poverty analysis. In keeping with the classification scheme developed in the body of this paper, such a model could be seen as part of a new 'fourth generation' of CGE models for developing economies. It will share with the third generation dual-dual models the concern for both incorporating poverty analysis in a general equilibrium framework, and to do so with as parsimonious a structure as possible. Given that there is probably considerable disguised unemployment in the rural sector and also rural-urban migration in the Harris-Todaro fashion, these features of the dual-dual model can be maintained. At the same time, work on Bangladesh can extend the dual-dual model in a minimalist fashion in at least two directions.

First, regarding the traded goods sectors, exports can be treated as both rural and urban based with different characteristics in each case. For example, both production structure and wages may be different. The second significant modification will be in the introduction of a financial sector in order to study the impacts of financial liberalization.

Within such a model, the consequences of the adjustment policies for allocation of resources, household income distribution of income and impact on poverty can be examined in a way similar to the dual-dual approach. Beginning with a solid understanding of the causality of household income distribution, flow-of-funds etc. the model will incorporate the causality of the poverty at both national and regional levels as completely as possible. The incorporation of financial SAM with important financial instruments, for instance, interest rate, required reserve ratios, and credit control will be a necessary part of the modelling process.

A properly constructed financial SAM will serve as the data-base for the financial extended dual-dual CGE model. The introduction of financial markets along with different financial assets, for example, currency, and interest-bearing deposits on loans, would specify within limits the alternative forms of assets available to the portfolios of savers. The main participants in the financial sector are the firms, households (mainly urban and upper income rural), the central bank, the banking system, and other financial institutions. Incorporation of rural-based financial institutions such as the Grameen Bank can also be carried out if data are available. Modelling these markets adequately but economically so that focus is still on poverty and the model does not become too complex will need to be the major emphasis of this work. For example, the behavior of the central bank may be specified in terms of (i) financing of government debt, and (ii) managing changes in foreign reserves, changes in money supply, required reserve ratio of the banking sector, and foreign and domestic borrowing. The domestic banking sector should be treated as fulfilling the task of financial intermediation between the savers and the borrowers. The other financial sectors can serve as alternative sources of financing in addition to the banking sector. There are specific issues with respect to the behavior of firms such as working capital management, debt and equity structure and new investment that require close attention when finance becomes an integral part of the CGE model.

Although the first stage of the modelling process can only aim at comparative statics experiments, an eventual dynamic extension will clearly be desirable.

Download this Discussion Paper [ PDF 469.8KB| 73 pages ].

Comment(s)

There are [0] comment(s) for this entry. Post a comment.