This thesis examines the theory-consistent economic determinants of aggregate and sectoral domestic investment in the context of the oil-rich and -based Iranian economy within the modified neoclassical-accelerator type investment model, into which uncertainty-driven oil measures are incorporated. The thesis thus adds new sets of data and knowledge to these fields particularly based on the empirical analyses conducted in Chapters Five and Six. To the author’s knowledge, long-run economic determinants of domestic aggregate and sector-level investment have not been estimated for Iran and therefore the findings of the thesis may be of interest for scholars in Iranian Studies and policy-makers in Iran. Furthermore, the findings of this study contribute to current debates on the economics of natural resources and on investment as well as the application of the investment literature in the context of partial-market, oil-abundant, and exporting economies like Iran.
The thesis further studies the role of the Iranian state in the process of capital formation and structural shifts, and the means through which these have been affected by the availability of oil income during the pre- and post-revolutionary years. Hence, this thesis provides an analytical basis for studying how economic and political institutions as well as the role of the state have evolved in an oil-based economy context. From a policy-making perspective, therefore, it guides policy-makers in their decision-making processes which can lead to greater long-run and sustainable growth and stability. To be contributory to policy studies that go beyond short-term forecasting requirements, particular attention is given to long-run equilibrium properties and stability of the models.
16
Importantly, the neoclassical-accelerator type theories of investment, such as the ones employed in this thesis, proved relevant and important as a tool to assess the effectiveness of economic determinants of domestic investment in semi-market oil-dependent economies and the underlying reasons for their partial applicability for such economies. This was despite the restrictive assumptions inherent in the standard investment models which are (partly) at variance with the structure of the financial systems and markets of these economies. In addition, a theory-consistent neoclassical-accelerator type identification of a CVAR model, such as the one applied here, was found useful to investigate the theoretically motivated long-run relationships between market-based economic determinants of investment in such economies, its outcome depends on the peculiarities of individual countries under study.
Contributing to the natural resource curse literature, this thesis identified the presence of the
‘Iranian Disease’, as a special case of the Dutch Disease, based on the empirical results related to the sector-level analysis presented in Chapter Six. According to the Dutch Disease theory, it was expected that the relation between investment and oil windfalls in sectors of manufacturing and agriculture to be negative, and that relation in oil and services sectors to be positive.
However, the Iranian-type Dutch Disease was characterized by the expansion of investment and output in the sectors of services and manufacturing, yet by the contraction of output in the oil and gas sectors. The findings suggested that, primarily through state-directed investment for the promotion of industrialization in the country, oil income was continuously invested in the manufacturing sector during the pre- and post-revolutionary era. Therefore, it is plausible to believe that, in a mixed-market economy like Iran, the state’s ambitions to achieve industrial development could dominate the Dutch Disease effects.
One of the limitations of this study concerns the reliability of macroeconomic and oil data. Since most of the world’s natural resources are state-owned, there is considerable evidence that nationally reported statistics may be politically biased (Davis, 2006). Thus, data necessary to examine the described hypotheses are collected from several national and international sources and cross-checking is conducted.
Another constraint is the modest extension of the research sample, since the analysis in this study focuses on Iran because of the unique institutional set-up of its economic and political system. An extension, however, could take into consideration other oil-based and oil-rich
17
countries at a regional level (e.g., the Gulf region or the Caspian region), or a cross-country analysis framework which includes a large number of countries endowed with (various) natural resources (e.g., as classified by Collier and Goderis, 2007) to test the generalizability of the results at regional and cross-country levels. This study does not focus on actual versus optimal investment and saving rates. Such an approach may require a more general model which is beyond the scope of this study. Even though the current study concerns an rich and oil-dependent economy, it does not intend to examine the issue of ‘exhaustible resources’ as the latter belongs to a different area of research.
At sector-level analysis, this study applies individual time-series cointegration tests to separately study domestic investment determinants for each of the four major economic sectors in Iran.
A possible yet important extension of this work could be to account for cross sector dependence by employing panel time-series techniques including panel unit root and panel cointegration tests to investigate whether and the extent to which sectoral dependence affects inferences.6
Another limit of this study concerns the scope of this thesis as it chiefly focuses on the macro determinants, rather than the micro foundations, of investment decision-making in the country.
The latter is beyond the scope of this thesis, given the limited availability of (firm-level) data and detailed studies on this topic which could allow for an in-depth analysis of the nuances of investment decision-making processes at the micro level in the country. An important extension of this study hence would be an investigation of how investments are decided at micro-level in the country which gives a particular emphasis on the role of the political actors and institutional factors in the process of investment decision-making in the context of the Iranian economy.
Lastly, this study is only partially applicable to other resource-rich and -based countries. This is because these countries are not necessarily similar in their type of government, economic policies, international relations and their political institutional economic system. Thus, it is likely that the modified mainstream investment models, including the ones followed in this thesis, may not be sufficient in unravelling the determinants of investment in (oil-based) economies with imperfect financial markets. This calls for new theoretical framework and
6 Panel unit-root tests, among others, include Levin, Lin and Chu test, Im, Pesaran and Shin test and residual-based LM test. Panel cointegration tests, among others, include residual-based DF and ADF tests, residual-based LM tests, Pedroni tests and likelihood-based cointegration test (for details, see Baltagi (2008, pp.273-300)).
18
techniques for investigating investment determinants in economies whose conditions do not fully correspond to the assumptions of the existing investment models in the literature.