The Long-run Effects of Structural Change and the Treatment of International Capital Accumulation, Mobility and Ownership

Abstract

Taking a commonly-used and commonly-available trade policy model as our starting point, we examine the long-run effects of large-scale structural change with and without international capital accumulation, mobility and ownership. We demonstrate the relative merits and limitations of different treatments of international capital accumulation, mobility and ownership. In doing so, we present a treatment of international capital accumulation, mobility and ownership that gives policy analysts an approach to analysing the effects of large-scale structural policies that is not too heavy in its theoretical and data demands. Our findings support the work of Baldwin and others who have demonstrated that ignoring capital accumulation, mobility and ownership underestimates net output and welfare effects of large-scale structural change.

Keywords:

• Computable general equilibrium models
• capital accumulation
• international capital mobility and ownership
• structural change

JEL Classification:

• C68
• F15
• F21

Acknowledgements

Most of the work reported here was undertaken while both authors were at the Productivity Commission, Australia. The views expressed here are ours and do not necessarily represent those of Monash University or the Centre for International Economics. We thank Philippa Dee for helpful comments on this work.

Notes

1. Others who have also explored this issue are Smith (1976, 1977) and Srinivasan and Bhagwati (1980).

2. Another commonly-used global model is G-Cubed (McKibbin & Wilcoxen, 1999). A recent search on Google Scholar for the term “GTAP” resulted in 8110 matches whereas for the term “G-Cubed” resulted in 1390 matches.

3. For some examples, see Adams (2005), Anderson et al. (2006), Anderson and Pangestu (1998), Brockmeier and Pelikan (2008), Domingues et al. (2008), Elbehri et al. (2000), Fernandez de Cordoba et al. (2005), Fugazza and Vanzetti (2008), Islam (2003) and Nijkamp et al. (2005).

4. Here we are referring to the standard GTAP model as documented in Hertel and Tsigas (1997), which is comparative-static. The Dynamic GTAP model (Ianchovichina McDougall, 2000) is a recursively-dynamic model that addresses the capital accumulation and mobility limitations of the standard GTAP model.

5. The SALTER treatment of international capital mobility has never been presented in an academic journal publication. Thus, one objective of this paper is to introduce the SALTER treatment to applied trade policy practitioners and make it more widely known.

6. The UR can be regarded as a large-scale structural policy: it was a significant first step towards capping the level of protection for agriculture, manufacturing, services and intellectual property to the status quo of the early 1990s, while also committing contracting parties to significant liberalization in manufacturing and agriculture.

7. This is the approach taken in Adams (2005).

8. This treatment allows regions to fund growth in their capital stock using domestic saving and foreign borrowing, and allows full arbitrage of regional rates of return. It is similar to the “long-run closure with the steady-state database” in Walmsley (2002).

9. This treatment forces all capital stock growth in a given region to be funded by saving only from within the region itself. This is similar to the “endogenous capital, fixed savings rate” treatment in Francois et al. (1995) but without international capital mobility and cross-border ownership of capital.

10. In this respect, the work of Francois et al. (1995) makes a similar contribution but only addresses the issue of capital accumulation; it does not address the issues of capital mobility and ownership. It also does not apply a widely-used and widely-available trade policy model.

11. This is version 4.1, which is available at https://www.gtap.agecon.purdue.edu/resources/res_display.asp?RecordID=415 and is extensively documented in Hertel and Tsigas (1997).

12. The models applied here are implemented and solved using the multistep algorithms available in the GEMPACK economic modelling software (Harrison & Pearson, 1996).

13. This section contains only a brief description of the ICM extension. Those seeking further detail on the nature of the ICM extension should consult McDougall (1993) and Hanslow et al. (1999), upon which this section is based. Hanslow et al. (1999) documents all the modifications required to the theory (Chapter 3) and data (Appendix E) of GTAP to implement the ICM extension.

14. This new investment allocation mechanism replaces the GTAP investment allocation mechanism (equation (3)).

15. This is a relatively simple method of calibrating the capital stock for long-run equilibrium. A more complex method is that used in Walmsley (2002), where shocks to output are applied to the model so as to achieve equalization of regional rates of return on capital.

16. With negative net saving, the allocation of income between consumption and savings can no longer be regarded as the result of a utility maximizing decision. Left unchanged, this would present problems for the utility maximizing theory in the model.

17. The UR was concluded in 1994 and allowed for yearly staged reductions in tariffs, with most final offer rates due to come into effect by 1 January 1999 (McDougall, 1997).

18. As per the UR agreement, Australia, New Zealand, Japan, Canada, the US and the EU are treated here as industrial economies; all other regions are treated as developing (GATT Secretariat, 1994).

19. Thailand experiences a negative change in allocative efficiency as the biggest increases in domestic output are in the most highly protected industries in the base data, e.g. wheat, sugar cane and sugar beet, dairy products, sugar, and beverages and tobacco products.

20. This assumes that for all regions. This is a good approximation of the results in this scenario, as the capital accumulation relationship (equation (5)) makes KE_r mainly a function of KB_r (which is fixed) because investment-capital ratios are around 0.1 or less for all regions.