Real convergence: empirical evidence for Latin America

Abstract

This article investigates the real convergence of 17 Latin American countries to the US economy for the period 1950 to 2011. Time series methods are used to test stochastic and β-convergence. These methods include the possibility of one or two structural changes. The results show that when endogenous structural changes are considered several Latin American countries exhibit stochastic convergence. Nevertheless, real convergence to the US is found only for three Latin American countries: Chile, Costa Rica and Trinidad and Tobago, with these countries also presenting evidence of stochastic and β-convergence.

Keywords:

• real convergence
• structural breaks
• Latin American countries
• empirical convergence tests

JEL Classification:

• C32
• O40
• O54

Acknowledgements

Comments from two anonymous referees are gratefully acknowledged. Juncal Cunado gratefully acknowledges financial support provided by the Ministerio de Ciencia y Tecnologia (ECO2008-02458-E).

Notes

¹ The US economy is chosen as the benchmark country, since this developed economy is the main investor and trade partner of Latin America. Therefore, it is interesting to test if the Latin American economies have improved their economic performance with respect to the US.

² See Levine and Renelt (1992), Mankiw et al. (1992), Barro and Sala-i-Martin (1995) and Fung (2009).

³ See Quah (1993) and Evans and Karras (1996). For a review of the differences between cross-sectional convergence and time series forecast convergence, see Durlauf and Quah (1998).

⁴ σ-convergence is satisfied when the SD of real per capita output across a group of economies tends to decrease over time.

⁵ A more precise definition of stochastic convergence is provided in Section III.

⁶ See Greasley and Oxley (1997), Cellini and Scorcu (2000) and Strazicich et al. (2004).

⁷ See Sowell (1992), Michelacci and Zaffaroni (2000), Dolado et al. (2002), Cunado et al. (2006) and Mayoral (2006).

⁸ Organization for Economic Co-operation and Development (OECD).

⁹ See the definitions of I(0), I(1) and I(d) processes in Hamilton (1994).

¹⁰ The specification of δ′Z _t in Equation 1 in this test is δ₀ + δ₁ t.

¹¹ See Carlino and Mills (1993), Tomljanovich and Vogelsang (2002) and Nieswiadomy and Strazicich (2004).

¹² The specification of δ′Z _t in Equation 1 for this test is δ₀ + δ₁ t + δ₂ D _t  + δ₃ DT _t . See LS (2004) for the selection of the number of augmentation terms, k.

¹³ The specification of δ′Z _t in Equation 1 for this test is δ₀ + δ₁ t + δ₂ D _1t  + δ₃ D _2t  + δ₄ DT _1t  + δ₅ DT _2t .

¹⁴ The IGK is a hypothetical unit of currency that has the same purchasing power as the US dollar had at a given point in time in the US.

¹⁵ Available at: http://www.conference-board.org/data/economydatabase/ (accessed on 14 February 2012).

¹⁶ The unit root tests of LS (2003, 2004) were performed using the Gauss codes downloaded from the website of Junsoo Lee. Available at http://www.cba.ua.edu/~jlee/gauss.

¹⁷ The information on Latin American economies is obtained from Reinhart and Rogoff (2009).

¹⁸ Gross National Product (GNP).