Introduction
It should not be a surprise that poverty reduction is the main development goal in Mozambique. For almost three decades (and counting), the United Nations Development Programme (UNDP) and the World Bank have consistently ranked Mozambique as one of the poorest countries in the world. A sharp decline in the official poverty incidence statistics, from 69% in 1996–97 to 54%, is indicative of significant development strides. Indeed, many donor agencies take Mozambique as an example to follow in sub-Saharan Africa. Mozambique is a ‘donor darling’, writes Hanlon (2009), and international aid has been on the rise. However, the food riots that erupted in 2008 and 2010, and the official poverty statistics showing an increase in the incidence, depth, and severity of poverty between 2002–03 and 2008–09 serve as a sombre reminder that many challenges remain.
The national strategy to reduce poverty, locally known as PARPA1, governs the efforts to address development challenges such as poverty and food insecurity. PARPA hinges on the neoliberal theory, which posits that the benefits of economic growth would trickle down to the poor (Romer 1990). However, a wide range of empirical evidence suggests that this is not happening. About 55% of the population lived below the national poverty line in 2008–9 (Ministry of Planning and Development [MPD] 2010), well above the target of 45% established in PARPA II (Government of Mozambique 2006). Additionally, there has been ‘little substantive progress’ with regard to chronic child malnutrition in recent years (UNICEF 2006, p. 12). Yet, between 2002 and 2008 Mozambique had an annual economic growth rate of about 8%.
The apparent paradox of rising poverty, food insecurity, and chronic child malnutrition despite economic growth is because growth is concentrated in few industries, mainly in ‘mega projects’ in the energy and mineral sector (such as natural gas, aluminium, heavy sands, and charcoal). For example, the Mozambican Aluminium Smelter (Mozal) alone accounts for 60% of the country's total exports. These mega projects create few local jobs, have few local linkages, benefit from huge tax exemptions and rely heavily on imported goods, and only a small fraction of their production is consumed locally. Meanwhile, both the local government and the international community have so far chosen a non-interventionist policy for the agricultural sector. Nevertheless the importance of agriculture both as a source of employment and government revenue is considerable. The underperformance of the agricultural sector is among the main reasons for persistent poverty and food insecurity (Ministry of Planning and Development 2010). This calls for a change in development policies, and therefore a search for an alternative development model that might be more effective in reducing poverty and food insecurity.
In medical sciences, the search for an effective antidote requires a profound knowledge of the poison. Similarly, a search for effective development policies must begin with an understanding of what successful policies require: avoiding failures. Therefore, this paper analyses some of the failures in development policies in Mozambique. The purpose of this paper is both to identify development failures and to discuss alternatives to overcome them.
The remainder of the paper proceeds as follows. The next section discusses the failures that resulted in a decline in agricultural productivity between 1996 and 2008. This is followed by a discussion on disproportional aid for the agricultural sector, as well as the importance of ensuring timely availability of funds. Then the paper discusses the need for the use of accurate data in estimating the gross domestic profit (GDP). Next, the paper elaborates on the effect of inadequate support to farmers' participation in the output market, and how it can foster the adoption of improved technologies. The impact of distrust between government officials is also covered. Finally, the paper concludes with some policy remarks.
Decline in agricultural productivity
In Africa, agriculture contributes a large percentage of total revenues. In Mozambique, about 80% of the population is employed in agriculture (including a segment of the urban population), which contributes a quarter of the GDP (Instituto Nacional de Estatística 2004). Agriculture is therefore an important engine of poverty reduction. Indeed, massive poverty reduction between 1996 and 2002 was possible mainly due to growth in agricultural production (Simler et al. 2004, Fox, Bardasi and van den Broeck 2005, Arndt, James and Simler 2006). Data from the national agricultural survey (locally known as TIA2) show an increase in the use of modern technologies (Figure 1), which might have created the illusion of growth in agricultural productivity. Empirical evidence, however, points to a decline in production of many crops per hectare and per adult household member during that period (Boughton et al. 2006).
Percentage use of selected technologies by year. Source: Trabalho de Inquérito Agrícola 2008.
The already low level of provision of extension services show a downward trend since 2005 because extension workers did not have either vehicles or fuel to reach farmers. The supply of price information to farmers also declined, despite evidence indicating its positive and significant impact on farmers' incomes (Mather 2009). The already low use of chemical fertilisers, animal traction, and access to credit also show a decline in recent years. It is also revealing that animal traction is not even mentioned in PARPA II, and little is being done to overcome the occurrence of tsetse flies on cattle, causing endemic disease in regions of high agricultural potential in central and northern provinces. The central provinces show a dramatic increase in poverty levels, from 46% in 2002–03 to 60% in 2008–09 (Ministry of Planning and Development 2010).
Within the Ministry of Agriculture, budget cuts may have negatively affected some employees. Recently, there were cuts in fuel and telephone allowances and other subsidies. Many employees of the Ministry of Agriculture became discontented and unmotivated. In the headquarters in Maputo, the Ministry of Agriculture lost critical human resources. For instance, within the Directorate of Economics, several heads of department resigned. Furthermore, the Department of Policy Analysis, once a functional and active unit in terms of producing policy briefs and research outreach, was reduced at one point to only two employees from its previous team of more than 11 policy analysts, a decline of more than 80%.
The increase in poverty levels was caused in part by a decline in agricultural productivity. In 2008, production of food per capita and per hectare were lower than that in 2002 (Ministry of Agriculture [MINAG] 2010), suggesting that agricultural productivity shows a negative trend since 1996. Therefore, the increase in agricultural production observed between 1996 and 2002 was caused by an expansion of cropped area (Figure 2) and not by growth in agricultural productivity.
Poverty reduction versus total landholding size. Source: Trabalho de Inquérito Agrícola 2008. See also Inquérito aos Orçamentos Familiares, the consumption expenditure survey conducted in 2008–9 from which the official poverty estimates are derived.
Meanwhile, poverty levels increased between 2002–03 and 2008–09 mainly due to a decline in both landholding size and agricultural productivity. Expansion of cropped area might have been possible in the 1990s because farmers had been cultivating significantly smaller fields (due to war), land was relatively more abundant, and it was still possible to expand land using a hand-hoe. However, at present many smallholders might have reached the threshold where further expansion of cropped area would require other tilling methods (such as animal traction) or the use of hired labour, but the poor cannot afford either.
It is hard to sufficiently emphasise the importance of growth in agricultural production to reduce poverty, whether through expansion of cropped area or growth in agricultural productivity. The third poverty assessment report on Mozambique concurs with other studies in arguing that additional success in reducing poverty depends on growth in agricultural productivity (MPD 2010). Yet the government chooses not to reward and retain qualified employees that would provide great latitude for policy change. Furthermore, the local government has not given a high priority to agriculture, and employees in the agricultural sector are apparently unable to lobby the state and donors for resources. This has resulted in a decline of resources allocated to agriculture, which in turn has resulted in lower agricultural productivity levels.
Disproportional aid for the agricultural sector
Mozambique has experienced one of fastest growth rates worldwide over the two decades up to 2010 (World Bank 2006). Growth in absolute terms, however, might not be as large because of the incredibly low base from which progress began. According to the World Bank, in 1992 Mozambique was ranked the poorest country in the world (Plank 2003); 17 years later Mozambique had a GDP rank of 169 out of 177 countries (UNDP 2009). In addition, about 51% of the government budget is allocated to salaries and subsidies, which represents nearly 10% of its low GDP (MediaFax 2010). It is indisputable that Mozambique needs international aid at the current stage of the development process. Indeed, foreign aid to Mozambique corresponds to 15% of GDP, compared to the African average of 4% of GDP, making it one of the top receivers of aid in the world (World Bank 2006).
Meanwhile, there is also agreement on the role of agriculture as one of the main engines of poverty reduction in Mozambique (MPD 2010). Yet, and unfortunately for the poor whose livelihoods depend markedly on crop and livestock production, only a small proportion of foreign aid is allocated to the agricultural sector (Figure 3). The neoliberal policies imposed by the international community, and pursued by Mozambique, imply less intervention in the agricultural sector. Hence, PARPA stresses Millennium Development Goal (MDG) numbers two through eight, while systematically ignoring agricultural production which arguably would halve hunger and poverty by the year 2015.
Percentage of total foreign aid allocated to agriculture by year. Source: OECD Statistics Database.
This is not to say that other sectors are not as important as agriculture. As a matter of fact, Mozambicans would not be able to discuss poverty reduction issues had it not been for international aid allocated to education and other social areas. However, solving the problem of hunger and poverty first might help to solve other problems more rapidly. How can children in rural areas go to school and excel if their stomachs are growling with hunger? How can gender equality be ensured when hunger threatens the whole family's survival and relatives decide to anticipate the dowry of their 14-year-old girl?
Due to the importance of agriculture to the economy, especially in rural areas, the proportion of aid to agriculture should be increased dramatically. But most importantly, funds should be made available in accordance with the agricultural season. For instance, between mid December 2009 (the period of first weeding in agriculture) and the second quarter of 2010 (main harvests), donors were on strike over electoral reform, corruption, and the growing role of the Frelimo party inside the state apparatus (Mozambique Political Process Bulletin 2010). As much as donors' demands were legitimate, the strike coincided with a critical period in the agricultural season. When funds are made available this late, they can only help a few poor employees through per diem payments when they visit farmers, but not help the poor farmers themselves.
Conflicting statistics
It has happened before. A senior government official once decided to visit a research station and assess how the Mozambican ‘green revolution’ was unfolding. But the demonstration plots had no crops standing due to lack of funds to purchase the inputs and irrigate the plots so research station employees purchased lettuce, kale, and other leafy vegetables in a fresh market and these ready-to-eat crops were ‘planted’ in the demonstration plot the day of the visit, and plots were irrigated. The senior government official was impressed by what he saw, without knowing that those vegetables were actually grown in neighbouring South Africa. Perhaps the employees were inspired by a similar episode that took place in Maputo during the African summit in 2003. Mature palm trees, about seven metres or taller, were ‘planted’ in Julius Nyerere Avenue and other places in order to create an illusion of a green city. As with the demonstration plot, this illusion makes senior government officials ‘believe’ that the ‘green revolution’ plan is bearing fruit. However, such fruits understandably never reach maturity. But what is the relation with contradicting GDP figures?
The ‘impressive’ economic growth figures might be marred by problems with the national accounts, the timing and availability of solid information, and the use of preliminary and unreliable data provided by district authorities such as the research station mentioned earlier and other secondary sources. For instance, due to a widespread drought that occurred in 2005, it is reasonable to expect a decline in crop production. Indeed, TIA data indicate that total maize production declined by 17%, relative to 2003. However, the official figure used in the computation of GDP shows an increase of 7% in the volume of the agricultural sector in 2005, relative to the previous year. Although maize production represents a small share of the total volume of the agricultural sector, it is fair to expect comparable trends.
This is critical since agriculture represents a quarter of the GDP, and therefore an overestimation of growth in agriculture has a far-reaching effect on GDP figures. The Ministry of Planning and Development attributes failure to reduce poverty in the period 2002–03 to 2008–09 in part to the underperformance of the agricultural sector. This is corroborated by data from TIA and several other studies. Yet, GDP figures from this very ministry show growth in agricultural production. Consistent? No!
Systems for the national accounts need to be reviewed, with the possibility of ex-post reviews when data from the agricultural surveys and other solid non-agricultural sources become available. Otherwise, GDP figures should be used with moderation and caution. Negative results in the realm of development may be unexpected and even controversial. In fact, a negative result is undesirable in general, with few exceptions such as the HIV test where it is fair to say nobody wants to be positive. However, negative results are equally or sometimes more important than the positive outcomes. They help avoiding similar mistakes in the future, while at the same time helping the planning process. Decision makers will otherwise think that development policies in place are bearing positive results if negative results are concealed and not reported. After all, why change a winning formula?
Inadequate support to agricultural marketing
Only a small proportion of smallholders are able to participate in the output market, and profitable participation has been linked to participation in input markets and to the ability to store the produce. For instance, Howard et al. (2003) show that, in Mozambique, farmers using improved maize seeds earn 24% more than non-users if they are able to delay their maize sales until November, about five months after harvest.
Recently, the focus of the local government in boosting the agricultural sector has been the search for alternatives to increase participation in the input markets, such as the promotion of input fairs and by reducing the size of fertiliser bags so that they are more accessible to the poor. However, participation in the input market remains extremely low and shows a negative trend in the period 2005–08. Less than 3% of smallholders use fertilisers, mainly for cash crops, and about 5% use improved seeds on maize, the most important staple crop.
Low use of improved inputs is associated with the lack of output markets. Perhaps an analogy with the hypothetical fishermen paradox might be helpful in stressing the importance of output market participation. For one week, a labour economist observed a group of fishermen return home every day at 9 a.m., each carrying 0.5 kg of fish to feed their household members. The fishermen allocated the remaining time till dusk to leisure and other social activities. The economist started to wonder how much more fish could have been caught had the fishermen been persuaded to cut leisure and allocate more time for fishing. In the following week, the fishermen increased their catch by a factor of 10, but most of the fish rotted because they had no cooling system and no buyers either. Although it is risky to catch only half a kilogram each time as the weather may be hostile the next time or the fishermen may fall ill to malaria, and therefore their family members would face hunger, they decided to resume their previous practice and allocate more time to leisure.
Similarly, in 2009–10 the government provided 610 tonnes of improved wheat seed in Tsangano, Tete Province. It was sown on 7550 hectares, and 10,600 tonnes of wheat were produced. But there was no market: the Beira milling company, Mobeira, went to Tsangano and bought only 30 tonnes of wheat, but never returned. So the crop was sold over the border in Malawi, for prices regarded as derisory. Ironically, in 2010 violent riots broke out in part over the rise in the price of bread, due to the increase in the import price of wheat. Will these farmers decide to produce more wheat in the following season, or maintain their status quo of staple food production using traditional seed varieties? Would rational farmers decide to buy chemical fertilisers using a significant share of their incomes while uncertain of the weather and without any output marketing prospect?
A recent study on the economic impact of the use of improved technologies in Mozambique shows no statistical significance on farmers' incomes, unless farmers have access to the output market. Low use of improved technologies is a result of lack of output markets. Viewed alternatively, in order to tackle the low use of improved inputs, the local government should tackle the problem of agricultural marketing first. A guaranteed output market would induce the demand for improved inputs. In 2008–09, the state provided improved seeds of various crops on credit (including wheat) to farmers, but less than 1% of the total amount of credit was recovered because farmers were unable to sell. In the meantime, tobacco farmers who also receive inputs on credit have higher rates of return than those who received improved seeds of staple crops, also on credit. Why?
Although the state decided to close the marketing boards in the 1980s, tobacco and sugar-cane companies, which are owned by transnational firms, are allowed to have marketing boards. This ensures the access to both input and output markets, and hence smallholders are able to pay the inputs back. In contrast, marketing boards are not allowed for maize and other staples, which are crops grown by an overwhelming majority of smallholders, and therefore many farmers are unable to return the inputs received on credit. Nevertheless, only a small fraction of smallholders have such access to inputs on credit.
Rural development programmes should also include investments in road infrastructure and missing markets (output, input and financial markets). However, the relationship between market infrastructure and the use of improved technologies is barely explored in PARPA. Instead, the government expects the private sector to co-participate in the investment of market infrastructure through the development of ‘public–private partnerships’ (Government of Mozambique 2006, p. 126). But this is not happening.
As mentioned earlier in the introductory section, there have been significant improvements in road infrastructure. However, most of the improvement is concentrated in southern provinces, with the exception of a landmark bridge connecting central and northern provinces and few other roads. The decision on the location of infrastructure investments apparently does not take into account regional differences in terms of agricultural potential in Mozambique. More than half of all tractors, used by less than 3% of farmers in Mozambique, are located in the southern provinces, because road infrastructure is better. But central and northern provinces are the regions of higher crop potential. As a result, high transaction costs create an insurmountable hurdle for farmers in central and northern provinces to sell their surplus in the deficit areas in the south. Instead, the southern provinces usually import food, mainly from South Africa, while the central and northern provinces export food to neighbouring countries. This is also a result of the lack of dynamic agroprocessing and value-added industries, which, in addition to improved road infrastructure, would make it more cost-effective for the southern provinces to purchase maize and other agricultural products from the central and northern provinces.
Distrust between the government and the ordinary people
Mozambique has experienced a series of violent protests in recent years. In February 2008 violent riots broke out in several urban areas over the rise in prices of food and fuel. At least five people died, hundreds were injured, several stores were looted, tyres were burnt in the streets, and physical infrastructure was attacked, including a secondary school in Maputo. Moreover, several businesses were paralysed for a couple of days.
A senior government official, who was later convicted in one of the country's biggest corruption cases, had previously announced that changes in prices were irreversible. Following the riots, however, the government introduced fuel and food subsidies which potentially benefited some urban citizens. These subsidies were supposed to alleviate poverty. However, they did not prevent a repetition of the tragedy seen in 2008. In September 2010 a series of violent protests erupted in some urban centres. Again, the government spokesperson announced that changes in price were irreversible.
Protesters were obviously hopeful that the local government would find silver bullets to address food insecurity, hunger, and poverty. Yet to their surprise, live bullets were fired instead. Hélio Muianga, an 11-year-old boy, was returning from school when a bullet took his life accidentally. Another 12 people also lost their lives, and hundreds were injured. The violent reaction from the government, however, was sadly predictable since this time the protesters were labelled ‘outlaws and bandits’. Once again, the government reintroduced food and fuel subsidies in order to quell the violence. Some austerity measures were also announced.
Food riots have had severe social and economic consequences. There have been other, apparently non-food related, riots too. In the northern provinces some people believe that health workers are responsible for the spread of cholera. In 2009, health workers were treating water with bleach (chlorine) when locals attacked them. A health centre that served about 24,000 people was destroyed and dozens of people died, including 12 arrested protesters who died of suffocation in a crowded cell in the police station and at least two Red Cross volunteers. Protesters also destroyed at least 13 houses of Red Cross volunteers.
But why would ‘bums and vandals’, as they were labelled, attack a school? And why would protesters destroy a health centre? Arguably, it is related to the atmosphere of distrust among the poor in Mozambique. The school that was attacked bore the name of the current President Armando Guebuza, suggesting that the protesters did not believe in the economic policies in place because of the soaring cost of living. In addition, it might be hard for the poor to trust the government since its senior officials call them ‘bandits and outlaws’, or ‘bums and vandals’. Moreover, as Hanlon (2009, p. 129) notes:
If a nurse or health worker normally demands a bribe to provide proper treatment, why should they be trusted when they say they are treating the water for free? If government actions have only led to increasing poverty and loss of jobs, why trust the government? The poor have every reason to ask if the sincere health workers sent into rural areas are not just an attempt to build up trust so that the poor can be better exploited. And they have every reason to distrust the local leaders who ally themselves with the new outside exploiters.
The Mozambican government needs to crack down on corruption so that trust can be rebuilt again. Government officials should also be more responsible when referring to the poor, no matter how stressed they might be. Apparently small things can in fact reduce poverty, or at least help by not increasing it. The school would have been spared by protesters had they trusted the name behind it. The health centre would also have been spared if the government first provided subsidies that would allow them to increase their incomes (such as fertiliser subsidies) and treat their water themselves rather than providing free chlorine. Distrust impairs development.
Concluding remarks
Mozambique is not on track to halving the proportion of people living in extreme poverty and/or suffering from hunger. Several factors have a bearing on the current poverty levels. This paper has discussed some of these factors with the purpose of identifying a few of the ingredients for a more effective poverty antidote.
First, the local government should reward and retain qualified employees that can provide timely policy briefs. In addition, there is a need for a change in agricultural policy, demanding a rather interventionist government strategy in the agricultural sector. Second, donors should increase the share of agricultural aid, as well as ensure that availability of such funds is synchronised with the agricultural seasons. Third, some caution is needed when using the ‘impressive’ GDP figures, while at the same time there is a need to review the national accounts with the possibility of ex-post reviews when data from the agricultural surveys and other solid non-agricultural sources become available. Fourth, marketing boards should be created for staple crops, as already exist for cash crops such as sugar cane and tobacco. This would in turn foster the demand for improved inputs. Finally, the government needs to rebuild the trust of the poor. This would involve, for instance, cracking down on corruption and an increased commitment to enhancing household incomes.