Market Integration and Uncertainty: The Impact of Domestic and International Commodity Price Variability on Rural Household Income and Welfare in Ghana and Peru

Abstract

We estimate rural household income uncertainties and welfare changes due to commodity price and production variability in Ghana and Peru under different scenarios for international and domestic market shocks. Uncertainties significantly affect the variability of household income, especially for households that are specialised in the production of few commodities. Wider exposure to international markets would increase the income variability for producers of commodities that are subjected to stabilisation policies in Ghana but would reduce the variability that rural households in Peru face. In terms of welfare, rural households in both countries are expected to gain if fully exposed to international markets.

Acknowledgements

No senior authorship is assigned. The views expressed in this article are those of the authors and do not necessarily reflect the views of the Food and Agriculture Organisation of the United Nations. The authors would like to thank the editor and two anonymous referees for their constructive comments, as well as Carlo Azzari, Agostina Zanoli and Panayiotis Karfakis for excellent research assistance.

Notes

1. The use of a cointegration-VECM framework has been questioned in terms of discontinuous trade flows and nonstationary transfer costs within the context of a narrower definition of market integration (Barret and Li, 2002). We maintain that the application of cointegration tests to a pair of prices is an ostensibly empirical exercise.

2. We use the Augmented Dickey-Fuller (Dickey and Fuller, 1979) and the tests by Phillips and Perron (1988) and Kwiatkowski-Phillips-Schmidt-Shin (Kwiatkowski et al., 1992). The tests are performed by including up to 12 lagged terms of the differenced terms in the regression and we use the Akaike information criterion to choose the appropriate lag length, thus trading off parsimony against reduction in the sum of squares. The results are not presented due to space limitations but are available from the authors.

3. The estimation of VECMs, rather than unrestricted VARs, for nonstationary prices results in forecast error variance estimates that are consistent and predictions that are asymptotically optimal (Phillips, 1998). In addition, at the recommendation of a referee, we also test for structural breaks in the price series, as well as for threshold versus linear cointegration applying the method by Hansen and Seo (2002). In all cases threshold cointegration was rejected. Again, the results are not presented in the paper but are available from the authors.

4. Tables presenting fully detailed household classifications are included in the Online Appendix which is available with the online edition of this issue.

5. For more details on the institutional context of commodity markets in Ghana, see Asuming-Brempong (2003).

6. Exposure of household to world prices expressed in US$ excludes the portion of variability or stability that is induced through exchange rate variations.

7. Welfare effects have been calculated as the difference between the term in the left-hand side and the first term on the right-hand side of expression (5). If households were not exposed to market and non-market variability, all these terms would amount to zero. In order to calculate these changes, we have utilised a risk aversion coefficient of 2.0, whilst own and cross price demand elasticities are taken from the coefficients of the FAO global partial equilibrium projections model COSIMO.

8. Heuft and Buchenrieder (2003) provide a comprehensive review of the agricultural policies implemented in Peru.