The Political Economy of Maize Production and Poverty Reduction in Zambia: Analysis of the Last 50 Years

Abstract

Poverty and food security are endemic issues in much of sub-Saharan Africa. To eradicate extreme poverty and hunger in the region remains a key Millennium Development Goal. Many African governments have pursued economic reforms and agricultural policy interventions in order to accelerate economic growth that reduces poverty faster. Agricultural policy regimes in Zambia in the last 50 years (1964–2008) are examined here to better understand their likely impact on food security and poverty, with an emphasis on the political economy of maize subsidy policies. The empirical work draws on secondary sources and an evaluation of farm household data from three villages in the Kasama District of Zambia from 1986/87 and 1992/93 to estimate a two-period econometric model to examine the impact on household welfare in a pre- and post-reform period. The analysis shows that past interventions had mixed effects on enhancing the production of food crops such as maize. While such reforms were politically popular, it did not necessarily translate into household-level productivity or welfare gains in the short term. The political economy of reforms needs to respond to the inherent diversity among the poor rural and urban households. The potential of agriculture to generate a more pro-poor growth process depends on the creation of new market opportunities that most benefit the rural poor. The state should encourage private sector investments for addressing infrastructure constraints to improve market access and accelerate more pro-poor growth through renewed investments in agriculture, rural infrastructure, gender inclusion, smarter subsidies and regional food trade. However, the financing of such investments poses significant challenges. There is a need to address impediments to the effective participation of public private investors to generate more effective poverty reduction and hunger eradication programmes. This article also explores the opportunities for new public–private investments through South–South cooperation and Asia-driven growth for reducing poverty in Zambia.

Keywords

Subsidies fertilizers food security smallholders economic reforms public–private investments

Introduction

Climate change and turbulence in world food markets are posing great challenges to world food security. In 2008, world food security was at its lowest ebb in half a century. Grain carryover stocks in mid 2007 were the lowest since records began in 1960 (FAO, 2008). This was complicated by even more dramatic rises in energy and fertilizer prices and the onset of the global credit crisis in 2008. High energy prices and the lack of credit also raised food prices; food production dwindled, making food unavailable and unaffordable for many around the globe (Hanjra and Qureshi, 2010). Global grain and fertilizer prices have subsequently fallen, but high food prices persist in many countries and the future food outlook (FAO, 2008) shows that food prices will continue to come under pressure until at least 2015 due to population growth, income growth, nutritional transition and climate change (Khan and Hanjra, 2009).

This situation has brought agriculture firmly back on the global development agenda. The World Development Report 2008 was dedicated to (as reflected in its subtitle) Agriculture for Development for the first time in the last three decades. The Report (World Bank, 2008) has placed agriculture afresh at the centre of development. The intricate world food security situation has also been accompanied by a renewed interest in large-scale input subsidies, specifically fertilizer subsidies, and related economic reforms in agriculture for development and poverty reduction policies in Asia and Sub-Saharan Africa (Bwalya et al., 2009). However, the policy objectives of food and fertilizer subsidies remain paradoxical and are subject to major political-economic and implementation challenges. Subsidies must largely target those producers whose food production and input use is constrained due to market access, affordability and poverty issues (Poulton et al., 2006). Reaching those producers whose productivity did not rise significantly and who did not benefit directly during the Green Revolution is the key policy challenge (Namara et al., 2010). Often there is a divergence between the political economy of input subsidies and its impact on food production, food prices and household welfare in terms of consumer versus producer welfare. A review of the historical experience and impact of input subsidy programmes can enhance the contribution of subsidies to agriculture for development, food production, food security and poverty reduction programmes. However, the lack of analysis of subsidy programmes in sub-Saharan Africa highlights a major need for country studies that report historical experience with and impact of subsidies on household welfare (Dorward, 2009).

This article therefore examines the impact of agricultural policy reforms in Zambia since the mid 1960s, with an emphasis on the maize and fertilizer subsidy policy and its politics. There is quite dated literature showing that the impact of economic reforms range from mixed outcomes (Nenci and Pietrobelli, 2008) to ‘a devastating failure’ of structural adjustment in Africa (Gibson, 2004). Zambia’s economic policies and attempts at structural adjustment since the 1980s have also been previously studied (Dorosh et al., 2009; Pletcher, 2000), and some analyses have examined the impact on food security. An exceptional contribution is a recent study (Xu et al., 2009) examining the factors influencing the profitability of fertilizer use on maize in Zambia. Yet the political economy of maize policies and its impact on food security and poverty have not been examined to the same extent. This article builds on our previous work (Culas and Hanjra, 2011). It examines the political economy of maize policies in Zambia during the last 50 years. Empirical evidence is presented from secondary sources as well as a case study of the impact of agricultural reform policies on maize production in the Northern Province of Zambia. The article also contributes to our understanding of the relationship between state and international policies and local agricultural production and poverty reduction.

The Poverty Scorecard for Zambia

The eradication of extreme poverty and hunger is a key priority in sub-Saharan Africa, a continent that will fall short of achieving all of the Millennium Development Goals by 2015. Zambia ranks sixth among the sub-Saharan African countries in terms of purchasing-power adjusted GDP per capita (World Bank, 2008). But Zambia ranks 164 of the 182 countries based on the United Nations (UNDP, 2010) Human Development Index (HDI) – a multidimensional indicator for human poverty showing the overall level of human well-being, capturing education, life expectancy and GDP per capita. It ranks (167) even lower on the Gender Development Index (GDI), implying marked inequalities in human development among men and women. Inequality fell dramatically during the 1990s but remains high. The Gini coefficient (an indicator of inequality) was 0.48 in 1959, 0.59 in 1974 and 1991, and 0.49 in 1998 (World Bank, 2009). Rural areas have higher levels of inequality than urban areas. The poverty level was high in the initial period (1974), rose rapidly by 1996 and then declined somewhat. Poverty incidence, depth and severity have all remained higher in the rural than in the urban areas. For instance, the poverty incidence is 74 per cent in rural areas against 52 per cent in urban areas (World Bank, 2009). The country has slipped in its international ranking on the UN’s HDI: out of 177 countries, Zambia was ranked at number 153 in 2002 before slipping further down to 165 in 2006. Between 1990 and 2007, Zambia’s HDI fell by -0.17 per cent annually from 0.495 to 0.481 (UNDP, 2010).

Recently the reforms in Zambia have emphasized poverty reduction (Gibson, 2004) and the trade regime has been further liberalized. Macroeconomic performance has not been good, and the boom and bust in copper production has led to the fall in investments and a slow economic growth of around 1 per cent per annum during the 1990s (World Bank, 2009). Mining exports centred on copper still account for 70 per cent of the total value of exports and around 6 per cent of real GDP (ZDA, 2011). Agriculture accounts for about 17 per cent of real GDP and 70 per cent of employment. Manufacturing adds 18 per cent to real GDP, and services about 60 per cent (World Bank, 2009). Health issues such as malaria and HIV/AIDS have also compounded Zambia’s development challenges.

Zambia’s agriculture remains highly vulnerable to climate change and its adverse impact on food security and prices (Jayne et al., 2006). Maize is imported according to weather-related needs and the expected impact on maize prices, which has serious implications for the political economy in terms of rural and urban food security. For instance, the 2002 drought-related food crisis fuelled inflation largely driven by food prices (Jayne et al., 2006). A poor road infrastructure, limited access to farm credit for smallholders, high interest rates, fewer export crops and inadequate access to input and output markets are the main constraints to a more pro-poor agricultural growth process (WTO, 2002).

The government has recently paid a closer attention to smallholder producers of maize and other food crops. This process identified shortcomings in the agricultural liberalization programmes in operation since 1989, removed quantitative restrictions on imports, eliminated subsidies for agricultural inputs and outputs and deregulated prices. However, little was done to encourage the private sector to fill the gap in service delivery left by these reforms. The main outstanding issues that constrain investment include (van Rooyen and Sigwele, 1998): poor infrastructure such as roads, bridges and transport fleet; inadequate storage facilities and marketing institutions (Pinckney, 1993); poor access to microcredit; low funding and priority given to the agriculture sector; inadequate support for new drought-tolerant crop varieties and for integrating the crop sector with livestock (Love et al., 2006); high input prices and low maize productivity (Jayne et al., 2006); and macroeconomic instability and high inflation (WTO, 2002). These issues also impede the progress towards effective poverty reduction in Zambia.

Economic Reforms in Zambia

As to the political economy of policy reforms, Zambia underwent four policy regimes: market interventions and failed reforms, 1965–1990; comprehensive structural adjustment programme, 1991–1998; renewed growth reforms 1999–2002, with varied implications for economic growth (Table 1) (World Bank et al., 2004; World Bank et al., 2005; World Bank, 2009) and social development (UNDP, 2010); and pro-poor growth era, 2003-to date.

Table 1.

Policy interventions and socioeconomic outcomes in Zambia, 1964–2002

Indicators	Policy intervention			Structural adjustment programme (SAP)	Renewed growth reforms
	1964–1972 Market economy	1973–1984 State control	1985–1990 Economic transition	1991–1998 Structural adjustment	1999–2002 Renewed growth
Macroeconomic indicators	Average annual change (%)
GDP per capita (1995 local currency)	2.1	−2.6	−1.4	−2.4	1.9
Exports (1995 $US)	3.4	−1.8	−3.4	4.3	6.5
Imports (1995 $US)	8.0	−8.6	2.4	1.3	2.9
Fixed capital formation (1995 local currency)	—	−8.7	−1.0	6.3	11.0
External debt (1995 $US)	—	10.8	14.8	0.9	−6.5
Inflation (deflator)	6.9	11.4	67.8	71.7	23.9
Exchange rate (to $US)	0.0	9.0	69.4	76.2	24.9
Real interest rate (%)	9.5	−1.1	−25.4	0.9	13.4
Social indicators	Average value for the period
Life expectancy (years)	45.8	49.9	49.7	45.8	38.5
Infant mortality (per 1000 births)	111.8	96.3	103.7	111.3	112.0
Adult literacy (%)	48.7	56.7	65.7	72.7	77.3

Based on data from UNDP (2010); World Bank (2009).

During the first regime, Zambia opted for a market economy (1964–1972), favouring mining and urban areas, and few people benefited in the rural areas and inequality worsened. Market economy policies were abandoned in the early 1970s and replaced by state control (1973–1984), with the nationalization of copper mines and the expansion of state-owned enterprises. Regulatory policies, producer subsidies and fixed consumer prices were introduced to protect the urban consumers, which further entrenched the maize bias (Dorosh et al., 2009) and urban biase (Copestake, 2002). This strategy achieved little growth and was widely regarded as a failure. Hence, Zambia implemented its first structural adjustment programme (SAP) in 1978. Producer and consumer subsidies declined to 20 per cent of the fiscal budget by the mid 1980s, and Zambia entered into a period of economic transition (1985–1990) and attempted the second structural adjustment programme in order to correct the price distortions and to diversify into agriculture (World Bank et al., 2004; 2005). Following unrest in urban areas, the reforms were backtracked and a new set of reforms were implemented, involving a partial return to a command-and-control economy. As the economy went into recession, the third structural adjustment programme was introduced in 1989, in which the prices of all goods were decontrolled except for maize and fertilizers. Urban bias again dominated the policy agenda and the programme was backtracked; maize and fertilizer reforms were stopped (Kydd, 1998). Agricultural policies favoured urban consumers and rural development policies involved price controls, thereby undermining agricultural diversification (World Bank et al., 2005; World Bank et al., 2004). Incomes fell, poverty and inequality rose and gains in social outcomes were reversed by 1991 (World Bank, 2008).

Zambia started the fourth structural adjustment programme in 1991, which involved macroeconomic stabilization, privatization of state assets, public sector reform, agricultural reforms and regulation of banking sector (Maimbo, 2002).The liberalization of agricultural markets was a key component of the reform package. Previously Zambia favoured maize production through pan-territorial pricing (Masters and Nuppenau, 1993), input and output marketing subsidies and food subsidies to urban consumers. The new package attempted to eliminate food subsidies and reduce state intervention in the maize and fertilizers markets. Consequently food prices increased substantially and led to a considerable resistance in urban areas.

There was an initially adverse impact of SAP on poverty: poverty rose sharply during 1991–1996, and fell marginally subsequently (Table 2) (World Bank et al., 2004; World Bank, 2008, 2009). Also, the incidence of poverty rose, while the depth and severity of poverty fell. Inequality fell initially and later stabilized.

Table 2.

Poverty and inequality in Zambia

	Structural adjustment programme (SAP)
	1991	1996	1998
Poverty (upper poverty line)
Incidence	68.9	79.4	75.4
Depth	41.7	45.4	40.0
Severity	30.6	30.7	25.6
Extreme poverty (lower poverty line)
Incidence	56.5	65.8	59.8
Depth	32.4	33.2	27.6
Severity	23.2	20.8	16.2
Gender and poverty incidence
Male (household head)	67.9	78.9	74.9
Female (household head)	74.9	81.6	77.2
Inequality (Gini coefficient)
National	0.59	0.50	0.49
Urban	0.62	0.48	0.48
Rural	0.47	0.44	0.43

Based on data from World Bank (2008, 2009).

During its renewed growth reforms of 1999–2002, Zambia witnessed a positive and stable annual GDP growth of 3.4 per cent. The new government indicated a return to pre-reform policies and agricultural subsidization (maize and fertilizer subsidies). It is unclear whether the recent surge in growth was a result of structural adjustment, and whether the post-structural growth process will generate a broad-based poverty reduction (Dorosh et al., 2009). Our case study in the following section extends our previous work (Culas and Hanjra, 2011) to examine the impact of SAP on maize production and its consequences for household welfare and poverty. It shows that, in general, the political economy of maize policy reforms entrenched urban bias and had mixed effects on maize production and poverty during various reform periods.

Agricultural Policies and Maize Production

The urban bias that dominated the economic reforms and food policies in Zambia since the 1980s shifted investments away from agriculture to mining and manufacturing – the sectors that were politically more important to the urban population. Agricultural policies did not promote diversification in the agriculture sector. Maize production and fertilizers were heavily subsidized so as to ensure low food prices for the urban population with the political promise of maintaining food security, as maize remains the main staple food crop in Zambia. Prices of producer and consumer goods were determined centrally and implemented uniformly across the country. Maize production was promoted through further marketing support and the public supply of fertilizers and other inputs. These policy interventions shifted the production away from the areas of comparative advantage, despite the agro-climatic diversity of Zambia. The export potential of the agricultural sector was undermined; farmers cultivated maize in areas that were not normally suited to growing this drought-prone crop. The costs of subsidies amounted to 14 per cent of the total government budget and many farmers became monocroppers by 1991 (McCulloch, et al., 2000). Poverty was chronic and severe among the households isolated from markets and with poor access to infrastructure, particularly those living in the more remote rural areas. Most households in these more remote and isolated areas are subsistence farmers. Yet, despite their isolation and basically due to the maize bias in the reform policy, more remote households produced and sold maize albeit to a lesser extent than less remote households. Urban bias and maize bias created a perpetual trap of low investments and high poverty in these areas (Jayne et al., 2006).

The Impact on Maize Production

Like many African countries, the economic crisis in Zambia became acute in the 1980s. It was characterized by mounting external debt, deteriorating standards of living and stagnant productivity. The crisis was the outcome of external shocks, including the collapse of copper prices in 1975, an overvalued currency, trade protectionism, the public ownership of enterprises, and price controls and policy interventions in agricultural and food markets (Pletcher, 2000).

Historically, Zambia has pursued an interventionist policy in maize markets. Maize and maize meal were subject to price controls. Maize production was also subsidized through state provisions of smallholder credit, fertilizers and seed. Interventions in the maize market involved subsidies in both the input and output markets. For instance, in an attempt to provide cheap food for urban consumers in the 1980s, the government of Zambia created monopoly institutions in order to enable it to intervene in and control the market through the use of subsidies for maize. The excessive promotion of maize was at the expense of other crops, through heavier subsidies for fertilizers and credit to produce maize (Copestake, 2002), subsidized maize marketing, consumer price subsidies and also through the disproportionate allocations of resources to maize research, extension, storage and processing. The government also enforced pan-territorial pricing (and equity pricing) for maize throughout the country. As a result, the production of maize spread to areas with no comparative advantage, in areas with low rainfall, and contributed to soil acidification and degradation without proper soil management practices. When the production of maize moved away from the main consuming centres, transport costs also increased and the profitability for farmers was adversely affected. Maize production and pricing policies were not Pareto-Optimal nationally (Masters and Nuppenau, 1993), both in terms of total production and the impact on food security.

The FAO (2003) data on the production of maize and other staple crops during the various reform periods show that the SAP implemented in 1991 caused substantial changes in crop production and its composition. For instance, maize production fell sharply both in absolute terms and relative to other staple crops. Likewise, pan-territorial maize-pricing policies were unsustainable. During the 1992 and 1995 droughts, maize producers were worst affected as maize is more susceptible to drought than are other staple crops such as cassava and groundnuts (Figure 1) (FAO, 2009). The impacts on poverty were diverse. Maize producers concentrated closer to urban areas and railway lines were most impacted as marketing and input support for maize production provided by the government was higher in these areas previously; more remote and isolated areas did better despite the drought, including more cassava producers in Northern Province as they responded to agricultural liberalization by shifting production away from maize, and crop prices also responded to these shifts.

The contraction in maize production meant insufficient food supplies for the urban population. Zambia therefore had to import large quantities of maize. There was a switch from producing maize to other more profitable crops and livestock. The withdrawal of fertilizer and credit subsidies also resulted in an increase in the practice of the slash-and-burn farming system (chitemene) by poor farmers due to the lack of money for purchasing fertilizers (Kakeya et al., 2006). The farmers are forced to expand the cropped area as productivity falls; in fact an inverse relationship has been suggested between productivity and area expansion. However, the extent of the role that natural population growth plays in this area expansion remains unclear due to data limitations. The ability of farmers to derive subsistence and income from productively and sustainably managed natural resources is part of a much larger question of alleviating rural poverty in Zambia.

The Impact on Poverty

The agriculture sector employs most of the rural households in Zambia and provides food for more than half of the country’s population. In Zambia, about 85 per cent of the farmer are small-scale (cultivating 2 to 5 ha of land) (Table 3) (World Bank, 2003; World Bank et al., 2005), while the rest are medium- and large-scale farmers (cultivating >5 ha of land), whose operations tend to be more mechanized and who make use of well-organized farmer networks. The smallholders have low agricultural productivity and live below the poverty line. For instance, a comparison of yield, inputs, labour and returns of different smallholder crops for 1994 and 2002 shows that gross returns per day of labour are below $1 a day (the international poverty line), typically for maize compared to cash crops (Figure 2) (World Bank et al., 2005). Despite some improvement in gross returns over time, cotton and burley tobacco farmers are typically those earning above $1 a day. Subsidized inputs clearly increase the income and welfare of poor farmers. Removing these subsidies may lead to reductions in maize production and their welfare (reducing returns and worsening food security and poverty for smallholders) but also to rapid increases in the cultivation of cotton, cassava and other secondary staples (by medium- and large-scale farmers). Therefore, the analysis of the impact of the policy reforms on total farmer income, poverty and welfare need to take account of the alternative cropping patterns and the diverse impact on small- and large-scale farmers. The political economy of reforms needs to respond to the inherent diversity among the small- and large-scale farmers as well as to poor rural and urban households.

Table 3.

Overview of agricultural producers in Zambia

Farm category	Approximate number	Farm size	Technology, cultivation practices	Market participation	Location	Key constraints
Small-scale farmers	800,000 households	2–5 hectares, majority cultivating <2 ha of rainfed land	Hand hoe, minimal fertilizer and other inputs, household labour	Maize, other staple foods, primarily home consumption	Whole of Zambia	Remoteness, seasonal labour shortages, lack of cash on hand, lack of input and output markets
Emergent farmers	50,000 households	5–20 ha	Oxen, hybrid seeds, fertilizers, few with irrigation, mostly household labour	Staple foods and cash crops, primarily for market	Mostly along line-of-rail (Central, Lusaka, Southern) and some Eastern, Western Provinces	Seasonal labour shortages, lack of credit, weak market information
Large-scale commercial farms	700 farms	50–150 ha	Tractors, hybrid seeds, fertilizers, some irrigation, modern management, hired labour	Maize and cash crops	Mostly Central, Lusaka, Southern Provinces	High costs of credit, repayment of debt
Large corporate farms	10 farms	1000 ha and above	Highly mechanized, irrigation, modern management, hired labour	Maize, cash crops, integration with markets	Mostly Central, Lusaka, Southern Provinces	Uncertain policy environment, global market competitiveness

Source: Modified from World Bank, 2003; World Bank et al., 2005.

The Impact on the Environment

There are five major farming systems being practised in Zambia. One of those systems is the traditional slash-and-burn farming system called chitemene or shifting cultivation (Kakeya et al., 2006). This method involves a short cropping period of 2–6 years, which is then followed by long fallow periods of 10–20 years. Maize, cassava, millet, groundnuts and beans are the dominant crops grown in this system. This type of farming system is, however, unsuitable in the long run because of the inadequacy of land to allow for long fallow periods. Although this system is regarded as a method of low-input production technology, it is a major cause of deforestation in Zambia, and a poverty trap has long persisted in the system.

The chitemene is one of the dominant traditional land-use systems practised in the Miombo woodlands of Northern Zambia. It is estimated that the Northern Province alone has lost almost 35 per cent of its biomass, which is about 43,000 km² of the forestlands over the past 40 years due mainly to this system (Kakeya et al., 2006). The system therefore contributes to deforestation and other adverse effects such as the loss of biodiversity within the forests, water runoff and soil erosion and the depletion of soil fertility due to the leaching of nutrients and energy loss due to burning. However, the slash-and-burn biomass contributes to soil fertility in the form of ash. The ash is very efficient in terms of crop-yield improvement. Living closer to the woodlands can also result in gains in household consumption and welfare (Bandyopadhyay and Tembo, 2010).

The chitemene in Northern Zambia is an outcome of several factors such as population factors as population density growth, insecure land tenure, inadequate infrastructure development and the insufficient provision of necessary supporting services, as well as of the characteristics of farming families related to household size, age, gender, education and land availablability (Kakeya et al., 2006). These causes and characteristics are often the common determinants of poverty in Zambia. Therefore, the same factors that contribute to environmental degradation also keep the smallholders trapped in poverty.

The practice of chitemene has been increasing, particularly after the government embarked on the SAP. The policy measures taken to implement the programme, such as price deregulation, the liberalization of agricultural marketing and cuts in government expenditure through the removal of subsidies on fertilizers and credit, have all had their effect on the practice of chitemene (Holden and Shanmugaratnam, 1995). Reliance on subsidies and the lack of money and access to loans to acquire necessary inputs contributed to the decision of smallholders to return to slash-and-burn since it does not require fertilizers (Kakeya et al., 2006). The political economy of policy reforms must therefore respond to the heterogeneity and diversity among the farmers.

Empirical Evidence from a Case Study

The following case study examines the impact of policy interventions on maize production and the profitability of smallholders and their welfare. The case study evidence is based on primary data collected from the densely populated Kasama District in Northern Zambia (Culas, 1995), a typical district practising the slash-and-burn system. Socioeconomic data from three villages in the district, namely New Chambeshi, Old Chambeshi and Yunge, were analysed. The data were collected using random sampling at village and farm household levels for the cropping seasons 1986/87 and 1992/93. This dataset therefore facilitated for before and after comparisons of the impact of policy interventions with respect to the SAP implemented by late 1989. A two-period econometric model with a partial approach was estimated, using these data for the cropping season of 1986/87 – before intervention, and the cropping season of 1992/93 – after intervention.

We examined the profit maximizing behaviour of a sample of peasant farmers in the Northern Province, following the hypothesis of Schultz (1961) that farm families in developing countries are efficient but poor in the allocation of resources in traditional agriculture. With a profit maximizing goal, the economically efficient level of an input occurs when the marginal value product (MVP) of the input equals its marginal factor cost (MFC); that is, when the marginal physical product of the input (MPP) equals the input–output price ratio. This model is shown in Figure 3 for our case study of maize and fertilizer(s).

The Impact of Maize Prices and Input Market Liberalization

We examine the impact of changes in the fertilizer policy in three ways: first, we examine how increased fertilizer prices affect production and farm profits. Fertilizer price increases are likely to be more significant in the Northern Province because of its remoteness. In our model, we examined the impact of raising the price of fertilizers. Second, market liberalization might make fertilizers more readily available to farmers; therefore, we constrained the model so as not to allow adjustments (availability of fertilizers) and compared the relative impact of this on technical efficiency. Third, we used multiple regression models to determine the factors influencing fertilizer use in maize production, and to understand the impact of policy changes on these factors and vis-à-vis the welfare of smallholders.

Profitability in Maize Production

Withdrawal of input subsidies for fertilizers, combined with price ceiling of maize (Masters and Nuppenau, 1993), reduced the producer profit in maize production (Table 4). This estimation was done for a quadratic production function, using the Ordinary Least Square (OLS) method, for the before (1986) and after (1992) intervention period. Subsidized inputs clearly increased farmer incomes during the pre-reform period. Removing these subsidies led to reductions in maize production.

Table 4.

Profitability in maize production (per ha basis) in Kasama District, Zambia

	Before	After
Estimated production function (quadratic form)^†	Y = −6 + 5.95 X − 0.00197 X²(Y = Maize in kg/ha; X = Fertilizer in kg/ha)
Marginal physical product	∂Y/∂X = 5.95 − 0.00394 X
Fertilizer–maize price ratio	1.5738	2.3516
Factor cost on fertilizer at profit maximization	1066.224 kwacha	1894.334 kwacha
Gross return on maize at profit maximization	2548.728 kwacha	3336.216 kwacha
Gross margin in terms of fertilizer use	1482.504 kwacha	1441.88 kwacha
Reduction in profitability	40.62 kwacha

Note: ^† X and X² are significant at 1% and 10%, respectively. R² = 0.518. df = 38. Prices adjusted to constant term (1975 = 100).

It was found that there was a reduction in profitability in terms of fertilizer use by about 41 kwacha per hectare for the 1992 cropping season compared to that for the 1986 cropping season. The reported reduction in profit was mainly due to an unfavourable price ratio (relative price) between fertilizers and maize that resulted from the SAP.

The OLS regression analysis models are estimated for predicting the average value of the dependent variable in terms of the known (or fixed) values of the explanatory variables. An estimated coefficient of an explanatory variable can be interpreted as its effect on the dependent variable while considering the effects of other explanatory variables as fixed in the model. For an empirical illustration, consider Table 4 where the dependent variable is maize produced in kg/ha and the explanatory variable is fertilizers used in kg/ha. The relationship between the maize produced (Y) and the fertilizers used (X) is modelled by a non-linear model such as quadratic form (with fertilizer use as X and X²) as this form is fairly common for describing the responses of a crop output to fertilizer use. In the estimated model, both X and X² are statistically significant at 1 per cent and 10 per cent probability levels, respectively. Likewise, the results in Table 5 show that with one unit increase in fertilizer use, maize output increases by 0.77 kg at a 1 per cent statistically significant level, while the results for the effects of labour and land inputs on maize output are statistically insignificant. Further, as the coefficient for the dummy variable is small and negative and significant only at a 10 per cent level, there was weaker or no clear evidence for the effect of technological efficiency on maize production after the SAP.

Table 5.

Technical efficiency dimension of maize production in the Kasama District, Zambia

Predictor	Coef	Stdev	t-ratio	p-value
Constant	2.8321	0.4733	5.98	0.000***
In L (workforce in man labour)	−0.1693	0.1318	−1.28	0.205
In A (land area in ha)	0.1340	0.1354	0.99	0.327
In F (fertilizer in kg)	0.7748	0.0798	9.71	0.000***
Dummy (before = 1, after = 0)	−0.3126	0.1856	−1.68	0.098*
Dependent variable: In Y (Maize production in kg). R² = 0.733. F-value = 37.10***. df = 58

Note: * Significant at 10% level. *** Significant at 1% level.

Technical Efficiency in Maize Production

The technical efficiency of maize production was tested in a setting of production function with a dummy variable of 1 for the cropping season 1986 and 0 for 1992 season. The dependent variable was maize production in kilogrammes and the independent variables were fertilizers in kilogrammes, workforce in man hours and area of land cultivated in hectares. The Cobb-Douglas production function was finally chosen, based on the goodness of fit to the data as evidenced by the high R² value of 0.73, which means that the model variables explained 73 per cent of the variability in maize output (Table 5). Using the cross-section data, the estimation was carried out on the pooled dataset. Other empirical studies provide evidence on use of the Cobb-Douglas production function for the cross-sectional analysis of agricultural production (Hayami, 1970) and selecting alternative functional form in production function analysis (Griffin et al., 1987).

The estimated model shows that there was an upward shift in production for the 1992 cropping season compared to that for the 1986 season, as shown by the coefficient of the dummy variable. However, the value of the estimated coefficient was small and significant only at a 10 per cent level, which indicated that there was weaker or no clear evidence to hypothesize that the technical efficiency of maize production had improved ‘after’ the SAP. It was not clear whether this shift in production was due to a rise in the return to production inputs, an increase in factor endowments or both. Other factors, such as the provision of roads and infrastructure and support services for research and development were also important in boosting production in the province (Kakeya et al., 2006).

Economic Efficiency of the Use of Fertilizers in Maize Production

The impact of the SAP on the economic efficiency of maize inputs was estimated with a Cobb-Douglas production function using the OLS method (Table 6a and Table 6b). The ratio of MVP of fertilizers to MFC of fertilizers is given in Table 6c. This ratio was statistically tested for its departure from one for both cropping seasons. The test evidenced that farmers were economically inefficient in using fertilizers for the 1986 cropping season but made an efficient allocation of fertilizers during the 1992 season. This finding supported the hypothesis that the elimination of price distortions – in particular, the removal of subsidies for fertilizers – led to an efficient allocation of fertilizer.

Table 6a.

Estimates of regression coefficients of inputs in the Kasama District, Zambia

Production function (maize)	Constant (In β₀)	Land area (β1)	Fertilizer use (β₂)	Labour use (β₃)	R²	df	F–value	SE_v
Before	1.5644** (0.6431)	0.2797 (0.2966)	0.9887*** (0.1154)	0.0027 (0.2721)	0.776	25	25.35***	0.6728
After	3.7061*** (0.7015)	0.2226 (0.1641)	0.6222*** (0.1237)	−0.1845 (0.1512)	0.757	32	30.06***	0.5784

Note: Parentheses contain standard errors. ** Significant at 5% level. *** Significant at 1% level.

Table 6b.

Geometric means of variables and marginal products in the Kasama District, Zambia

Inputs	Before			After
	Mean	MPP	MVP	Mean	MPP	MVP
Labour use (man days)	0.8508	1.0445	0.6372	1.8084	−79.429	−70.037
Fertilizer use (kg)	127.7148	2.5483	1.5545***	214.9918	2.2531	1.988***
Land area (ha)	0.3359	274.0975	167.199	0.45168	383.6733	338.333
Output (kg)	1718			5241

Note: The standard error of MVP was calculated by the following formula: SE x = (Y/X) ^* S ^* Py, where: Y is expected production, X is the input, S is standard error of regression coefficient of the input and Py is the price of output.

Table 6c.

Ratio of marginal value product to marginal factor cost of fertilizer^a in the Kasama District, Zambia

Input	Before	After
Fertilizer	1.92912***	0.95811^ns

Note: ^at-distribution was used to test the statistical significance of the ratio of MVP/MFC equals one.^ns Not significantly different from 1 even at a probability level ≥80%. *** Significant at 1% level.

Factors Influencing the Use of Fertilizers in Maize Production

The influence of other socioeconomic factors on the amount of fertilizers applied was tested. Variables of interest were, family labour, area of land under the slash-and-burn system (ha), the previous year’s farm income (from maize), access to credit (dummy variable), and the gender of the household head (dummy variable). A multiple regression model using the OLS method was estimated for 1987, the before intervention period and 1992, the after intervention period (Table 7a and Table 7b).

Table 7a.

Factors influencing fertilizer use before intervention in the Kasama District, Zambia

Predictor	Coef	Stdev	t-ratio	p-value
Constant	14.01	61.27	0.23	0.820
Workforce	−6.336	7.627	−0.83	0.410
Chitemene	−0.003356	0.006527	−0.51	0.609
Farm income, before	178.72	25.59	6.98	0.000***
credit (0,1, yes is 1)	173.96	28.19	6.17	0.000***
Sex of household (0,1, 1 is female)	−15.86	29.08	−0.55	0.588
Dependent variable: Fertilizer use (kg). R² = 0.707. df = 57. F- value = 25.12***

Note: *** Significant at 1% level.

Table 7b.

Factors influencing fertilizer use after intervention in the Kasama District, Zambia

Predictor	Coef	Stdev	t-ratio	p-value
Constant	349.0	121.6	2.87	0.007***
Workforce	1.07	17.68	0.06	0.952
Chitemene	−0.00280	0.02022	−0.14	0.891
Farm income, after	96.50	32.93	2.93	0.006***
credit (0,1, yes is 1)	316.73	57.94	5.47	0.000***
Sex of household (0,1, 1 is female)	−191.75	61.12	−3.14	0.003***
Dependent variable: Fertilizer use (kg). R² = 0.675. df = 45. F- value = 16.63***

Note: *** Significant at 1% level.

The modelling results show that, access to credit significantly influenced the use of fertilizers for the both periods. Its influence was, however, higher for the 1992 cropping season than for the 1986 season. The previous year’s farm income from maize production had an influence on fertilizer use for both the seasons; however, its influence was higher for the 1986 season than for the 1992 season. Thus maize was a more profitable crop before the SAP. The gender of the household head had influence only for the 1992/93 cropping season where, on average, female-headed households used 191.75 kg less fertilizer than male-headed households. This implies that the negative effect of the high costs of fertilizers, due to the removal of subsidies, impacted more on female-headed households than on the (labour-rich) male-headed households.

Main Findings

The results show that after the SAP there was a fall in maize production; the household-level food security declined; and the fall in maize production encouraged chitemene in the study villages with consequences for the environment. Land area, access to credit, the previous year’s farm income and female-headed households were significant factors influencing fertilizer use in maize production such that future policy must leverage these factors in order to reduce poverty, as discussed below.

Future Policies for Poverty Reduction

At the macro level, SAP reforms in Zambia cannot easily be regarded as pro-poor, but were often on–off policies and were implemented as an immediate political response to a pending economic or food security crisis. These reforms, however, were a pre-requisite for economic growth as evidenced by an increase in investment and private-sector participation in the 1990s (World Bank et al., 2004; 2005). Given the falling per capita income during the previous two decades and rising income during the renewed growth reform period (1997–2002), it can be argued that overall reforms were not adequate for poverty reduction.

At the micro level, the reforms had an adverse impact on food-crop production and hence on poverty, as evidenced by the empirical results of our case study. A clear and decisive impact was lower maize yield and production – which must reduce household welfare as maize is a food staple and main crop. This is supported by other studies. Empirical evidence for a sample of 40 countries shows that a 1 per cent improvement in crop productivity decreased $1 a day poverty by about 1 per cent and improved HDI by 0.01 per cent (Hanjra and Gichuki, 2008). In Uganda, good agricultural growth led to a fall in poverty while a downturn in agricultural growth was accompanied by a rise in poverty (Kappel et al., 2005).

More Pro-Poor Agricultural-Led Growth

Agricultural-led growth offers a good pathway to poverty reduction in Zambia (World Bank, 2008), but it had to be more pro-poor. Findings from the CGE model of Zambia show that the reductions in poverty under a copper- or non-agricultural-led growth path are substantially lower than under the agricultural-led growth scenario but the goal of halving poverty may not be reached until 2040, in the current growth path (Figure 4) (World Bank et al., 2004; World Bank, 2009). A more pro-poor agricultural growth process can help achieve the hunger- and poverty-related MDG by 2015 and beyond (Hanjra and Gichuki, 2008).

Staple-Led Growth

Staple crops are grown throughout Zambia and can directly impact food security and poverty. The expansion in staples alone may lead to an increase in poverty (World Bank et al., 2003; World Bank, 2004), because a rapid rise in staple production may lead to lower prices. While lower food prices always benefit urban poor and rural non-farm households, they may reduce producer profits. Increased staple production may improve food security in the short run but may not result in sustainable poverty reduction in the long run due to market constraints (Diao et al., 2003).

Cash-Crop-Led Growth

Cash-crop-led growth can complement but would achieve a lower reduction in poverty than broad-based agricultural growth because of a lower consumption share of cash crops by the poor, higher benefits to the cash-crop producers (minority), the lack of capital available to smallholders to engage in cash-crop production, and the possible spending of the cash-crop revenue on higher imports that would largely benefit the non-poor (World Bank et al., 2004; World Bank, 2003). A lack of participation by a majority of the smallholders who constitute the bulk of the poor would mean a lower reduction in poverty.

There is too much expectation from and emphasis on agriculture in the fight against poverty in Zambia. Even agriculture alone cannot and will not spur growth and reduce poverty. In the light of this, the policy should be geared towards maximizing the synergies between farming and non-farming sectors through a comprehensive policy package for reducing poverty in the country.

Improved Access to the Markets

The potential of agriculture to generate a more pro-poor growth process depends on the creation of new market opportunities for smallholders. Market access must consider various agents such as smallholders, rural and urban consumers, processor, distributors, wholesalers and rural–urban market linkages as well as interlinkages with international markets and macroeconomic policies. Despite trade liberalization opening up new markets for agriculture, the implementation of appropriate policies and the provision of an institutional infrastructure have proven equally difficult in Zambia (Jayne et al., 2006). The analysis of past policies and their failures suggests that the state should limit interventions in markets, especially in ways that directly distort input and output prices and protect specific crops or consumer groups (Culas and Hanjra, 2011). Rather, the state should focus on addressing infrastructure constraints to help improve market access. Increased government and donor support for public good investments to drive down the cost of production and marketing as well as institutional arrangements will be critical for poverty reduction in Zambia (Jayne et al., 2006).

Rural Infrastructure

The role of rural infrastructure in promoting pro-poor agricultural growth has been well accepted (Narayanamoorthy and Hanjra, 2006). Simultaneous investments in rural water, roads, education, markets and related policy support measures can reduce poverty faster (Hanjra et al., 2009a). The rural infrastructure has suffered from market interventions through to macroeconomic stabilization policies in Zambia. For instance, under the SAP a ‘cash budget’ system was adopted, which limited expenditures to available revenues. Furthermore, available funds were diverted towards general public services with higher public-sector outlays but away from the vital economic services such as rural infrastructure with social services suffering as a result (Dinh et al., 2002). For example, just 18 per cent of rural households lived within 5 km of input markets in 1998 and 30 per cent of the paved roads were in need of repair. Beyond addressing issues of the existing infrastructure, the road network must be extended to remote rural areas.

Gender Inclusion

Pro-poor spending must address the gender differential in the access to markets and education. Zambia’s GDI ranking has remained low at 167, due to poor indicators for male–female school enrolment ratios, literacy rates, educational achievements and earned incomes (UNDP, 2006). The gender gap is a significant issue (World Bank, 2008) since the share of women employed in low-productivity agriculture is much higher than that of men, especially for food-crop production. Women also face considerably higher barriers in access to farm credit and informal credit. Addressing the gender gap is therefore a key pathway to poverty reduction and environmental sustainability.

Regional Food Trade

The urban population may double in the next 15 years (UNDP, 2006), and urbanization will lead to a growing demand for purchased and prepared foods in Zambia. Rural specialization will culminate in growing rural food markets. Despite diversification, staple crops such as maize will remain important to the poor. One study modelled the price impact of both drought and bumper maize harvests under closed border and free trade options in Zambia, using 15 years of historical data for five household groups in two geographical zones. The results showed that trade restrictions increased maize-price volatility, and raised transaction costs due to lower farm prices and higher consumer prices and thus diminished farm prices and producer incentives (Table 8) (Govereh, 2007; Haggblade et al., 2008). Inter-zone trade liberalization that links surplus with deficit zones is shown to reduce price volatility, improve farm prices and producer incentive, lower consumer prices and improve calorie consumption in poor households (Govereh, 2007). Empirical evidence (Dorosh et al., 2009) also shows that maize prices are more unstable in countries that restrict trade. For instance, maize-price instability is highest in Malawi and Zambia (with strict trade controls), moderate in Mozambique and Mali, and lowest in Kenya (no controls). Future agricultural policy reforms must address these regional food trade issues.

Table 8.

Impact of trade policies on maize prices and household consumption in Zambia

	Trade policy
	Closed border	Open border
Maize prices (% change from normal)
Bumper harvest (30% above normal)	−50%	−26%
Drought (30% below normal)	163%	36%
Calorie consumption of poor households
North	−2%	−3%
South	−19%	−9%
Production
Cassava production	24%	6%
Maize area, next season	34%	10%

Source: Based on Govereh (2007) and Haggblade et al. (2008).

Smarter Subsidies

Future policies must rethink the role of agricultural input subsidies, and must reconsider the current fertilizer and maize prices, uncertainty regarding future food prices, the impact on subsidy programmes and the role of subsidy programmes on food security under different price regimes and the emerging contexts in the future (Dorward, 2009). There is a need for smarter subsidies that focus on design and implementation issues and improve programme effectiveness and efficiency through improved targeting to the smallholders and poor zones.

Future agricultural policies must also support rural income diversification. Most smallholder crops such as maize, groundnut and cotton have low input costs, but lower returns to labour – below the internationally accepted minimum of $1 a day per person, to keep out of poverty (Figure 2) (World Bank et al., 2005). Entry into high-value crops such as burley tobacco and paprika requires substantial investments in new technology, inputs, extension and access to markets. The estimates of investment costs needed to enter high-value, commercial agricultural production are shown in Table 9 (Keyser, 2002; World Bank et al., 2005). High initial investments costs, technical skills required, risk involved and the need to finance indivisible investments in tractors and farm machinery are significant barriers to smallholders’ entry into high-return agriculture. The provision of agricultural credit at subsidized interest rates must be one of the primary policy actions (Graham and Mark, 2001).

Table 9.

Estimated investment costs of entry into commercial farming for selected enterprises in Zambia

Enterprises	Total investment cost ($)	Annualized investment cost ($)	Assumed area (ha)	Annualized investment cost ($/ha)
Centre pivot irrigation	187,163	11,840	80	148
Drip irrigation	72,791	1,820	40	163
Maize, commercial	565,000	77,370	600	129
Greenhouse	287,500	25,832	2	12,916

Source: Based on Keyser (2002) and World Bank et al. (2005).

The Financing Model

Much of the investment for reducing poverty in Zambia must come from the public sector. Private-sector investment along with donor assistance will continue to be important for poverty reduction in Africa (Hanjra et al., 2009b). Asian drivers have recently had an important impact on renewed growth and poverty reduction in Zambia; new European and South African investment is also of substantial importance in the current economic recovery (Carmody, 2009). South–South cooperation combined with more pro-poor agricultural growth offer substantial potential for poverty reduction. Commitment to public–private investments in agricultural and pragmatic economic reform is essential for sustained reductions in poverty in Zambia.

Conclusion and Policy Implications

The political economy of agricultural policy reforms and food security and poverty in Zambia was examined historically for the period 1964–2008. The empirical work was based on secondary sources and an evaluation of farm household data from three villages in the Kasama District from 1986/87 and 1992/93 (before and after intervention period, respectively). The study examined the impact of policy reforms undertaken within the framework of the SAP on Zambian agriculture with an emphasis on maize crop. The article identified a range of sociopolitical factors that might have changed and driven the specific policy changes implemented by the Zambian state. The analysis resulted in the conclusion that Zambia’s agricultural policy reforms induced diversification out of maize by reducing the profitability of maize production and of fertilizer use. Subsidized inputs increased farmer incomes in the pre-reform period. Removing these subsidies led to reductions in maize production and household welfare in the post-reform period. An analysis of the impact of these reforms on total farmer income, and on poverty, would need to take account of the indirect impact of the policy interventions in future research on this subject.

Based on the findings of secondary sources and an empirical evaluation of farm household data from three villages in the Kasama District, it is suggested that the policy package for reducing poverty in Zambia must realize synergies from: the staple-led growth; the cash-crop-led growth; improved access to the markets; the rural infrastructure; gender inclusion; regional food trade; and smarter subsidies. Emphasis on agriculture or any single policy alone is unlikely to reduce poverty, rather the policy package as a whole is needed to reduce poverty. Above all, the political state must maintain a principled pragmatism towards economic reforms in the future and develop public–private models for financing the investments.

References

Bandyopadhyay

Tembo

(2010) Household consumption and natural resource management around national parks in Zambia. Journal of Natural Resources Policy Research 2(1): 39–55.

Bwalya

Diallo

Phiri

Hamadoun

(2009) Sustainable Land and Water Management: The CAADP Pillar 1 Framework. African Union, NEPAD, University of Zambia. Johannesburg: New Partnership for Africa’s Development.

Carmody

(2009) An Asian-driven economic recovery in Africa?: The Zambian case. World Development 37(7): 1197–1207.

Copestake

(2002) Inequality and the polarizing impact of microcredit: Evidence from Zambia’s copperbelt. Journal of International Development 14(6): 743–755.

Culas

(1995) A case study on the impact of structural adjustment program on maize production in Northern Province of Zambia. Unpublished MSc thesis, Agricultural University of Norway.

Culas

Hanjra

(2011 forthcoming) Some explorations into Zambia’s post-independent policies for food security and poverty reduction. In: Contreras

(ed.) Agricultural Policies: New Developments. Hauppauge, NY: Nova Science Publishers, Inc., ch. 4.

Diao

Fan

Zhang

(2003) China’s WTO accession: Impacts on regional agricultural income – multi-region, general equilibrium analysis. Journal of Comparative Economics 31(2): 332–351.

Dinh

Adugna

Myers

(2002) The Impact of Cash Budgets on Poverty Reduction in Zambia. Africa Region Working Paper Series No. 39. Washington, DC: World Bank.

Dorosh

Dradri

Haggblade

(2009) Regional trade, government policy and food security: Recent evidence from Zambia. Food Policy 34(4): 350–366.

10.

Dorward

(2009) Rethinking agricultural input subsidy programmes in a changing world. Paper prepared for the Trade and Markets Division, Food and Agriculture Organization of the United Nations, Rome.

11.

Food and Agriculture Organization (FAO) (2003) World Agriculture: Towards 2015/2030: An FAO Perspective. Rome: Food and Agriculture Organization of the United Nations/Earthscan.

12.

Food and Agriculture Organization (FAO) (2008) Food Outlook: Global Market Analysis. November. Rome: FAO.

13.

Food and Agriculture Organization (FAO) (2009) FAO STAT. Rome: FAO. Available at: http://faostat.fao.org (accessed 15 July 2010).

14.

Gibson

(2004) Africa and globalization: Marginalization and resistence. Journal of Asian and African Studies 39(1–2), 1–28.

15.

Govereh

(2007) Regional trade in food staples: Using trade policy to improve farmer incentives and food security. Paper presented at Conference on Strengthen and Widening Markets and Overcoming Supply-Side Constraints for African Agriculture, Michigan State University, 3–5 June 2007.

16.

Graham

Mark

AGD

(2001) Relationship between the mode of land distribution, tenure security and agricultural credit use in Kwazulu-Natal. Development Southern Africa 18(3): 295–308.

17.

Griffin

Montgomery

Rister

(1987) Selecting functional form in production function analysis. Western Journal of Agricultural Economics 12(2): 216–227.

18.

Haggblade

Jayne

Tschirley

Longabaugh

Weber

(2008) Potential for intra-regional maize trade in Southern Africa: An analysis for Zambia at the sub-national level. Paper presented at the Michigan State University Food Security Group, South African Development Community (SADC) Southern Africa Regional Conference on Agriculture, Gaborone, Botswana, 8–9 December 2008.

19.

Hanjra

Gichuki

(2008) Investments in agricultural water management for poverty reduction in Africa: Case studies of Limpopo, Nile, and Volta river basins. Natural Resources Forum 32(3): 185–202.

20.

Hanjra

Qureshi

(2010) Global water crisis and future food security in an era of climate change. Food Policy 35(5): 365–377.

21.

Hanjra

Ferede

Gutta

(2009a) Pathways to breaking the poverty trap in Ethiopia: Investments in agricultural water, education, and markets. Agricultural Water Management 96(11): 1596–1604.

22.

Hanjra

Ferede

Gutta

(2009b) Reducing poverty in Sub-Saharan Africa through investments in water and other priorities. Agricultural Water Management 96(7): 1062–1070.

23.

Hayami

(1970) On the use of the Cobb-Douglas Production Function on the crosscountry analysis of agricultural production. American Journal of Agricultural Economics 52(2): 327–329.

24.

Holden

Shanmugaratnam

(1995) Structural adjustment, production subsidies and sustainable land use. Forum for Development Studies 2: 247–266.

25.

Jayne

Zulu

Nijhoff

(2006) Stabilizing food markets in eastern and southern Africa. Food Policy (Managing Food Price Risks and Instability in a Liberalizing Market Environment) 31(4): 328–341.

26.

Kakeya

Sugiyama

Oyama

(2006) The Citemene system, social leveling mechanism, and agrarian changes in the Bemba villages of northern Zambia: An overview of 23 years of ‘fixed-point’ research. African Study Monographs 27: 27–38.

27.

Kappel

Lay

Steiner

(2005) Uganda: No more pro-poor growth? Development Policy Review 23(1): 27–53.

28.

Keyser

(2002) Zambia Financial Crop Models: Comparison of 1994–5 and 2001–2 Seasons and Zimbabwe 2000–2 Seasons. Report prepared for the World Bank. Washington, DC: World Bank.

29.

Khan

Hanjra

(2009) Footprints of water and energy inputs in food production – Global perspectives. Food Policy 34(2): 130–140.

30.

Kydd

(1998) Coffee after copper? Structural adjustment, liberalisation, and agriculture in Zambia. Journal of Modern African Studies 26(2): 227–251.

31.

Love

Twomlow

Mupangwa

van der Zaag

Gumbo

(2006) Implementing the millennium development food security goals – Challenges of the southern African context. Physics and Chemistry of the Earth Parts A/B/C 31(15–16): 731–737.

32.

McCulloch

Baulch

Cherel-Robson

(2000) Poverty, inequality and growth in Zambia during the 1990s. Working Paper, Institute for Development Studies, University of Sussex.

33.

Maimbo

(2002) Explaining bank regulatory failure in Zambia. Journal of International Development 14(2): 229–248.

34.

Masters

Nuppenau

E-A

(1993) Panterritorial versus regional pricing for maize in Zimbabwe. World Development 21(10): 1647–1658.

35.

Namara

Hanjra

Castillo

Ravnborg

Smith

van Koppen

(2010) Agricultural water management and poverty linkages. Agricultural Water Management 97(4): 520–527.

36.

Narayanamoorthy

Hanjra

(2006) Rural infrastructure and agricultural output linkages: A study of 256 Indian districts. Indian Journal of Agicultural Economics 61(3): 444–459.

37.

Nenci

Pietrobelli

(2008) Does tariff liberalization promote trade? Latin American countries in the long run (1900–2000). Global Economy Journal 8(4): Article 2.

38.

Pinckney

(1993) Is market liberalization compatible with food security? Storage, trade and price policies for maize in Southern Africa. Food Policy 18(4): 325–333.

39.

Pletcher

(2000) The politics of liberalizing Zambia’s maize markets. World Development 28(1): 129–142.

40.

Poulton

Kydd

Dorward

(2006) Overcoming market constraints on pro-poor agricultural growth in Sub-Saharan Africa. Development Policy Review 24(3): 243–277.

41.

Schultz

(1961) Investment in Human Capital. American Economic Review 51(1): 1–17.

42.

United Nations Development Programme (UNDP) (2006) Human Development Report 2006: Beyond Scarcity: Power, Poverty and the Global Water Crisis. New York: UNDP.

43.

United Nations Development Programme (UNDP) (2010) Human Development Index (HDI). Human Development Reports 2009. New York: Zambia HDI Rank, UNDP. Available at: http://hdrstats.undp.org/en/countries/profiles/ZMB.html (accessed 23 July 2010).

44.

van Rooyen

Sigwele

(1998) Towards regional food security in southern Africa: A (new) policy framework for the agricultural sector. Food Policy 23(6): 491–504.

45.

World Bank (2003) Zambia: The Challenge of Competitiveness and Diversification. Report No. 25388–ZA. Washington, DC: World Bank.

46.

World Bank (2008) World Development Report 2008: Agriculture for Development. Washington, DC: World Bank.

47.

World Bank (2009) World Development Indicators. Washington, DC: World Bank. Available at: http://data.worldbank.org/indicator (accessed 27 July 2010).

48.

World Bank, Siegel

Alwang

(2005) Poverty-Reducing Potential of Smallholder Agriculture in Zambia: Opportunities and Constraints. Africa Region Working Paper Series No. 85, Washington, DC: World Bank.

49.

World Bank, Thurlow

Wobst

(2004) Operationalising Pro-Poor Growth – A Country Case Study on Zambia. UK: London: Department for International Development (DFID) and Washington, DC: International Food Policy Research Institute, and World Bank.

50.

World Trade Organization (WTO) (2002) Trade Policy Review, Zambia. Report by the Government of Zambia, Trade Policy Review Body, World Trade Oranization. Geneva: World Trade Organization.

51.

Guan

Jayne

Black

(2009) Factors influencing the profitability of fertilizer use on maize in Zambia. Agricultural Economics 40(4): 437–446.

52.

Zambia Development Agency (ZDA) (2011) Export Sectors. Available at: http://www.zda.org.zm/68-export-sectors (accessed 22 May 2011).