Modeling Gender Effects of Pakistan's Trade Liberalization

Abstract

This study uses a computable general equilibrium (CGE) model specially constructed for investigating gender dimensions of the effects of trade liberalization in Pakistan in both production and consumption. The model employs various indicators to measure the gendered impacts, including income poverty (Foster-Greer-Thorbecke [FGT] Indices), time poverty (leisure), capability poverty (literacy and infant mortality), and welfare (Equivalent Variation [EV]). The simulation results show that revenue-neutral trade liberalization in Pakistan increased women's employment in unskilled jobs and increased women's real wage income more than men's for all types of labor, but kept the division of labor biased against women. The study finds that Pakistan's trade liberalization adversely affected women in relatively poor households by increasing their workload, deteriorating capabilities, and increasing relative income poverty. However, the effects remained gender neutral or favored women in the richest group of households.

Keywords:

• Capabilities
• gender
• poverty
• trade liberalization
• JEL Codes: C68, J16, O24

Acknowledgments

This study was carried out with the aid of a grant awarded through the Gender Challenge Fund of the Poverty and Economic Policy (PEP) Research Network (www.pep-net.org), a joint venture of the Gender Network (Phase III) and the PEP Network, both financed by the International Development Research Centre (IDRC). I am thankful for the comments from discussants and participants in PEP Conferences in Colombo, Sri Lanka; Dakar, Senegal; and Addis Ababa, Ethiopia. Thanks are also due to Professors Bernard Decaluwé and John Cockburn from Université Laval, Canada for their comments. I would also like to acknowledge the valuable comments and suggestions by Dr. Marzia Fontana of the Institute of Development Studies at the University of Sussex. My special gratitude is due to Dr. Howard White, the executive director of International Initiative for Impact Evaluation (3ie), for his constant help commenting on different versions of this piece. Without his extensive help and encouragement, I would not have been able to prepare the study in this format. I am also grateful to anonymous referees at Feminist Economics for their very detailed comments on earlier versions of this piece. Last but not least, I am very grateful to the guest editors for their invaluable comments on many versions of this piece.

Notes

¹ Rationalization of the tariff structure involves the simplification of the tariff structure, which includes reducing tariffs and reducing the number of duty rates (these were reduced to four in the 1990s). However, on some items such as petroleum, tariffs have increased.

² Women earned 43 percent less than men in 1993–4. A 55 percent wage differential between men and women is due to discrimination in the labor market (Rehana Siddiqui and Rizwana Siddiqui 1998).

³ For details, see Erik Thorbecke (1992); Bernard Decaluwé, Jean-Yves Christophe Dusmond, Luc Savard (1999); and Siddiqui et al. (2008).

⁴ In 1989–90, the GOP introduced a general sales tax (GST) on both imports and domestic products to compensate for the loss in tariff revenue due to tariff reduction imports. Reduction in tariffs on imports along with tax rate adjustment on imported and domestic goods is called revenue-neutral trade liberalization.

⁵ The satisfaction of basic needs reflects (in aggregates) individual capabilities such as a long and healthy life, acquisition of knowledge, and having enough resources to buy food and other necessities. Empirical studies measure individuals' capabilities by various indicators such as IMR, life expectancy (LE), and literacy rate (LR) (Ravi Kanbur 1987; Siddiqui 2008b; Sudhir Anand and Martin Ravallion 1993). Infant mortality rate (IMR) is the best indicator to measure aggregates of capabilities and welfare, among other outcomes, because it measures the availability of several basic needs. It is an outcome variable of inputs such as health, nutrition, clean water, and sanitation facilities. The second capability development indicator is the literacy rate (LR). It indicates accumulation of knowledge. An increase/decrease in IMR implies an increase/decrease in capability poverty.

⁶ “Social Reproduction” describes household activities as classified by the system of national accounts such as cooking, cleaning, and looking after children and the elderly. However, my study includes fetching water as a household rather than market activity, and I use “household work” and “social reproduction” interchangeably.

⁷ The major difference between the Fontana and Fofana approaches is in the modeling of leisure time. Fontana and Wood (2000) assume men and women's leisure can be substituted for each other. In this approach, the model is calibrated assuming fourteen available hours for work and leisure and ten hours for personal care. Fofana, Cockburn, and Decaluwé (2003, 2005) use an explicit labor supply function, calculating maximum time available for work or leisure by using elasticities of labor supply with respect to income. The problem with this approach is that elasticities are not generally available. However, in the absence of time-use data, this approach would be preferred.

⁸ Trade liberalization such as tariff reduction on imports reduces domestic price of imports, and consumers increase the demand for imported goods and reduce demand for domestically produced goods. This reduces production in the domestic economy, and factors of production move toward relatively competitive sectors. The change in the structure of production changes the prices of domestic goods. As a result the prices of commodities produced in the domestic economy and the prices of factors of production change.

⁹ I calculated number of adults in a household using adult equivalent scale (see Deaton 1997 for details).

¹⁰ Other commodities such as housing, sanitation facilities, and utilities like water, electricity, and gas are public goods that men and women consume equally.

¹¹ Table 8 reports poverty based on head count ratios (Po). In the urban area, households are categorized on the basis of the education of the head of the household – no education, low education (less than five years), medium education (five to nine years), and high education (more than nine years). In the rural area, households are first categorized by gender as female- and male-headed households, and male-headed households are further distinguished by the employment status of the head of the household – employee, employer, self-employed, and “others,” which contains miscellaneous households.These households are arranged on the basis of the incidence of poverty in ascending order as follows. Urban households: no education – poorest; low education –poor; medium education – poor; high education – richest. Rural households: employees – poorest; self-employed – poorer; female-headed – rich; others – richer; employer – richest. The ranking is based at the regional level.

¹² SAM consists of twenty market sectors in agriculture (crop, livestock, forestry, fisheries, and others); industry (mining, food and beverages, textile, wood and paper, chemical, non-metallic, metallic, machinery, and handicrafts); and services (utilities, wholesale, retail trade, education and health, other, sanitation and other, public administration, defense, and construction).

¹³ For details of the social accounting matrix, see Appendix A.

¹⁴ In practice, women frequently multi-task, in particular combining childcare with other tasks.

¹⁵ Detailed results are available on request from the author.

¹⁶ For a detailed discussion of neoclassical models and their assumptions, see Bernhard G. Gunter, Lance Taylor, and Eric Yeldan (2005) and Bernhard G. Gunter, Marc J. Cohen, and Hans Lofgren (2005).

¹⁷ In the short run, we assume that capital cannot move from one sector to the other. If capital is also mobile (which is a long-run phenomenon), the country moves toward complete specialization.

¹⁸ The real exchange rate is implicit in the model and is calculated in the following way: er = e∗ (P^w /P_d ), where er, e, P^W , and P_d , respectively, are the real exchange rate, the nominal exchange rate, world prices, and domestic prices.

¹⁹ The satisfaction of basic needs reflects (in aggregates) individual capabilities such as the ability to have a long and healthy life, to acquire knowledge, and to have enough resources to buy food and other necessities. Empirical studies measure individual's capabilities by various indicators such as infant mortality rate (IMR), life expectancy (LE), and literacy rate (LR) (Kanbur 1987; Anand and Ravallion 1993; Siddiqui 2008b). Infant mortality rate (IMR) is the best indicator to measure aggregates of capabilities and welfare, because it measures the availability of several basic needs. It is an outcome variable of inputs, including health, nutrition, clean water, and sanitation facilities. Increase/decrease in IMR implies increase/decrease in capability poverty. The second capability development indicator is the literacy rate (LR). It indicates accumulation of knowledge.

²⁰ Japan achieved an IMR of 5 deaths per 1,000 live births in 1990 (World Bank 2005).

²¹ For details of income poverty analysis in CGE, see Rizwana Siddiqui and Abdur-Razzaque Kemal (2006).

²² Detailed results are available from the author.

²³ I have used skill and education interchangeably as follows: (1) no education =  unskilled; (2) low education = low skill, medium education = medium skill, and high education = high skill.

²⁴ The same pattern is found in the consumption of basic needs items.

²⁵ Other forms may include mobility constraint, industrial and occupational choices, decision making, etc.