South Africa has a long history of living with drought. A drought during the early 1930s that coincided with the great depression made a deep impression on many policy makers. Significant droughts also occurred during the 1960s, 1980s, and early 1990s. Despite this familiarity with drought, policy makers still struggle to quantify it and to develop a stable policy framework. Drought policy falls at the interface among the numerous definitions of drought that require some quantification of intensity, duration, and geographical extent; the demand of human activities for water; and the safeguarding of the natural environment. Therefore drought policy continues to evolve, particularly with the dynamic political environment in South Africa.
South Africa is characterized by east-west degradation in rainfall, from greater than 1000 mm in the east to around 150 mm in the west. Much of the country lies above the escarpment (1000 m) and experiences a combination of frontal and convective rainfall, falling mainly during summer. The southern coast receives rainfall throughout the year and the southwestern corner is dominated by winter rainfall. The 500mm isohyet divides the country into arable land to the east and primarily rangeland farming to the west. South Africa, receiving a little less than 500 mm as a national average, is classified as a dry country where the influence of variable rainfall cannot be underestimated.
Part of the difficulty in addressing drought in South Africa is the large proportion of the population that depends on rainfed subsistence agriculture. This sector relies heavily on the success of the rainy season to maintain adequate stocks of food. Historically, the infrastructure development and records maintenance in these areas have been neglected, which has made it difficult to monitor food status.
Traditionally, the broad definition of drought in South Africa has been seasonal rainfall less than 70% of normal (Bruwer, 1990). Using this criterion, drought has been shown to occur about 1 in 3 years in the western and northwestern regions of the country. Only 30% of the country receives 500
mm per annum or more. To further emphasize this, less than 18% of the country can be classed as arable land, of which 8% has fairly serious limitations to arable production (Schoe-man et al., 2000). This underlines the vulnerability of the country to the vagaries of rainfall.
In one of the early examinations into the causes of drought, the 1923 Drought Investigation Commission concluded:
Whether the character has altered or not, its quantity diminished, drought losses can be fully explained without presuming a deterioration in the rainfall. Your Commissioners had a vast amount of evidence placed before them from which only one conclusion can be drawn, namely, that the severe losses of the 1919 drought were caused principally by faulty veld [rangeland] and stock management. (Union of South Africa, 1923 p.5).
Subsequent investigations into various aspects of agriculture in the 1960s and 1970s reiterated this observation, indicating that no real lessons were learned, and once conditions returned to normal the policy status quo was generally maintained. Some initiatives were implemented to reduce overgrazing: conversion of cropland to grazing land in marginal areas and a revised scheme to limit assistance only to farmers who followed sound agricultural and financial practices.
In the past, state aid required magisterial districts (third-tier government) to be declared "drought disaster" status. This legal requirement necessitated a quantitative index that could be uniformly applied. This index was broadly defined as two consecutive seasons of 70% or less rainfall (Bruwer, 1990). Normal drought, for which a farmer was expected to be self-reliant, was for a period of 1 year or less. Disaster drought was defined as two consecutive seasons of below 70% of normal rainfall. A disaster drought implied that an area would qualify for state relief.
In fact, Bruwer (1990) noted that certain magisterial districts had been declared disaster drought areas for 70% of a 30-year review period, whereas some eastern portions of the country had never been declared. This indicates that the quantitative index is not optimal. In fact, because it depends on a deviation from mean annual rainfall, and the skewed distribution of annual rainfall totals in the drier areas, it favored drought declarations in the lower rainfall regions.
The National Drought Committee (NDC) consisted of representatives of farmers' organizations, the Soil Conservation Advisory Board, the financial sector (i.e., banking, agricultural credit organizations, and Department of Agricultural Economics and Marketing), and the agricultural community. This committee scrutinized applications concerning disaster drought status and then advised the Minister of Agriculture regarding these applications. On the local level, a district drought committee was formed under the chair of the local magistrate. This committee examined all local applications and submitted these to the NDC using the prescribed format. Declaration and revocation of drought-stricken areas were evaluated considering the following five criteria:
1. Rainfall over at least three seasons
2. Veld (rangeland) condition
3. Availability of water (for stock)
4. Stock condition or deaths
5. Availability and volume of fodder to be purchased
Drought assistance schemes were aimed at maintaining a nucleus herd or stock for reestablishment after the drought was over. A phased system of drought assistance was developed. The first level consisted of rebates on transport costs, followed by loans and finally subsidies at increasing rates as the drought continued. The assistance scheme served to protect natural resources and provide for livestock farmers during a disaster drought. The maintenance of a healthy and viable nucleus herd was not to be at the expense of the natural resources or to the detriment of a farmer's financial position. However, by the mid 1980s it was clear the policy had failed to protect natural resources as envisaged.
The government acknowledged that the drought assistance schemes contributed to sustaining selected agricultural production and communities. However, they struggled to define clear onset and shut-down phases of the drought assistance, and at times, assistance was out of phase with environmental conditions.
Many farmers overestimated the condition and potential of their agricultural resources. Farmers who overexploited their resources also benefited from the drought assistance schemes. This was clearly unacceptable. The scheme did not encourage a proactive or risk management approach. During their visit to South Africa, White and O'Meagher (1999) observed that inappropriate policy and incentives had led to inappropriate management, which resulted in nonviable agriculture and degradation of scarce natural resources. As noted above, Australian drought policy places considerable emphasis on encouraging primary producers to adopt self-reliant approaches to cope with drought and farm management.
For stock farmers, the government provided assistance in time of drought for the movement of stock or fodder and availability of loans. The definition of drought resulted in certain areas being under a disaster drought declaration for >50% of the time between 1956 and 1986 (Smith, 1994). Incidental observation of pre-1990 drought policy for stock farmers noted that periods of drought declaration were excessive, relating more to overstocking than to climate. The effects on land degradation were serious, and government assistance, while substantial, could be interpreted as exacerbating the problem rather than reducing it (Smith, 1994). The challenge of determining when intervention should occur and in what form has occupied experts in many countries. Until the 1990s, drought policy in South Africa was directed primarily at stock farmers (Walters, 1993). Stock farming was considered best adapted to the highly variable rainfall conditions in these areas. However, assistance tended to favor the poorer managers and climatically marginal area (Smith, 1993).
Bruwer (1990) pointed out that most drought counter-measures were reactive in nature. The development of new policy in the early 1990s required greater emphasis on a proactive approach. Bruwer also pointed out that drought effects are largely human induced. This signified an important turnaround in the approach of government policy regarding state drought intervention. Rangeland specialists identified farming areas that were overstocked by >50%. Research in the Free State province demonstrated how overstocking increases the length of fodder shortage periods and the probability of such shortages.
The government concluded that if no remedial actions were taken, land degradation could encroach on a significant proportion of the country (Bruwer, 1990). Expenditure on drought and flood relief was increasing significantly, from $150 million during 1984-85 to $330 million in 1992-93 (Mon-nik, 1997). Was this expenditure justified in achieving the government's objectives? With the continued degradation of natural agricultural resources, government was provided with a strong motivation to review its approach in providing financial and other relief. As put by Tyson (1988 p.17), who emphasized the need for a different paradigm: "All future planning must be predicated on the assumption that it is a land of drought rather than a land of plentiful rain."
During the 1980s, drought stakeholders in South Africa were captivated by a sense of anticipation. Research by Tyson (1986) concerning rainfall patterns and cycles of wet and dry spells on a decadal scale led to the successful prediction of the drought during the early 1980s. Van Heerden et al. (1988) and many others internationally investigated the feasibility of providing seasonal rainfall anomaly predictions. The first-ever "official" long-term prediction was attempted for the 1986-87 summer season. Because of possible misunderstanding, the forecast was submitted personally to interested parties and not released to the media (Van Heerden, 1990). Tyson's (1986) research on 33 widely distributed rainfall sites across the summer rainfall region of South Africa showed a clear oscillatory pattern in rainfall, producing 9-year spells of alternating generally dry and wet conditions.
Parallel to the progress in seasonal forecasts were developments in satellite-based operational monitoring systems such as the National Oceanic and Atmospheric Administration's or NOAA's Advanced Very High Resolution Radiometer, or AVHRR. Coupled with the availability of crop and rangeland models and rainfall deciles, there was a sense of anticipation that effective monitoring and forecasting were coming together. Dr. J Serfontein, the then deputy director-general of environmental affairs, speaking at the Southern African Regional Commission for the Conservation and Utilization of the Soil (SARCCUS) Workshop on drought, stressed the benefit such information could be to government for long-range planning (Serfontein, 1990).
Du Pisani (1990) summarized some of the objective techniques used to monitor drought in South Africa. These included utilization of water balance models, rainfall deciles, crop models, and remote sensing techniques. He highlighted the difficulty in determining drought severity because of the interaction of intensity, geographical extent, duration, and water resource requirements.
During the early 1990s, drought policy was changed to place greater obligations on farmers to reciprocate for state aid. Farmers were asked to commit to practices aimed at promoting resource conservation and the long-term sustain-ability of economic production (Walters, 1993). For example, only farmers who submitted their stocking rates (stock units per hectare) quarterly to their local Department of Agriculture office, and who reduced stock on drought warnings, were eligible for state aid. Although this policy led to improvements in the management of drought assistance for stock farmers, it continued the institutionalized neglect for the protection of the rural poor from threats posed by insufficient water, food, and employment (Walters, 1993). In addition, the definition of disaster drought conditions continued to favor the western portions of the country, where the coefficient of variation of rainfall was greater.
One of the keys to the new drought policy (post-1990) was the recognition of grazing capacity zones (Smith, 1993). Five grazing capacity classes were defined across the country. Drought assistance was limited to farmers who remained within the prescribed capacities. Farmers had to maintain records of stock numbers and report on a quarterly basis. Drought declarations were proposed by local drought committees and supervised by the National Drought committee. The director-general made the final decision, basing the declaration on the same five factors used by the NDC that were mentioned previously.
One of the central aims of the system was to protect natural resources, which had been a largely unrealized aspiration of the original policy. By elevating this within the policy framework, the goal was to take better care of natural resources in South Africa. In fact, since the 1960s, there has been little incentive for farmers to refrain from cultivating marginal agricultural land (Vogel, 1994). Inappropriate subsidies and a high level of mechanization and fertilization resulted in deterioration of farmland in many instances.
The approach during the early 1990s attempted to develop a longer term view with a greater emphasis on risk management. Part of the motivation was to address the apparent escalation in the cost of implementing the existing schemes. The scheme as developed in 1990 aimed to:
• Provide financial assistance to farmers
• Ensure agricultural resources are protected
• Encourage farmers to apply optimal resource utilization
• Contribute to the maintenance of a nucleus breeding herd
These aims did not differ much from the original aims, although the scheme did attempt to address some of the concerns that had become evident over the past years.
In 1990, the government argued that, because South Africa is an arid country, the state has a moral obligation and responsibility to assist people during times of hardship and prevent long-term disruption to communities and infrastructure (Bruwer, 1990). Although the systems seemed to address the situation among commercial stock farmers, subsistence farmers, crop farmers, and other significant stakeholders were overlooked. In contrast to this, the then Minister of Agriculture just 7 years later noted that drought aid encouraged bad practice, was inequitable in the past, and created expectations that government would bail out farmers in all disasters; he also noted that a prolonged drought would affect everyone in the country (Hanekom, 1997).
During the early 1990s, the National Consultative Forum on Drought was formed. This forum, comprising members of government and nongovernmental organizations, sought to raise the profile of all communities that were being affected by drought. This allowed for drought assistance to be provided to a broader range of the population (Vogel, 1994).
The White Paper on Agriculture (Ministry of Agriculture, 1995 p. 7) stated that: "Drought will be recognised as a normal phenomenon in the agricultural sector and it will be accommodated as such in farming and agricultural financing systems."
The new democratic government's stand on drought assistance was still to be tested during a "real" drought. The farming community, which had generally benefited in the past from the majority of government assistance, expressed concern. The acknowledgment that drought was part of the normal environment caused them to reevaluate their practices.
Thus as media attention focused on the El Niño/Southern Oscillation (ENSO) phenomenon during 1997, when there was an expectation of a large El Niño event, response from the private sector was noticeable. It was reported that McCarthy Motor Holdings sold no light delivery vehicles from their Hoopstad agency after the first El Niño press release (September 1997). In addition, tractor sales for October 1997 were 20% lower and sales of haymaking equipment increased by 50% over the corresponding period the previous year (Redel-inghuys, 1997).
The impact of this ENSO event on South African rainfall did not materialize as predicted, which caused many people to lose faith in these forecasts. However, economists observed that the financial discipline exerted by farmers resulted in them being in a much healthier state at the end of the season than if they had ignored the forecast from the beginning. This indicated that commercial farming can benefit from a greater appreciation of drought risk. It may also contribute to developing some degree of robustness to drought.
The discussion paper on agricultural policy in South Africa provided more detail concerning the new government's view on drought assistance. The authors recognized that past policies had weakened the farmers' resolve to adopt risk-coping strategies. Expenditure of $1.2 billion to write off and consolidate farmer debt during 1992-93 was identified as unsustainable. The government set itself on a course to provide other options, besides relief, to help farmers cope with drought (Ministry of Agriculture, 1998).
More recently, the government has placed more emphasis on risk management. During 2002, an agricultural risk insurance bill was developed. The purpose of the bill was to enhance the income of those farmers and producers most vulnerable to losses of agricultural crops and livestock due to natural disaster, including drought. In addition, a drought management strategy was under development during 2003 and 2004. This document is eagerly anticipated to provide greater detail in line with the policy guidelines.
Williams (2000) pointed out that the recent advances in long-lead forecasting provide the opportunity to focus more on managing climatic variability instead of being the passive victim of an "unexpected" drought. South Africa needs to maintain its investment in meteorological research and communication to the public, and to encourage links with the global meteorological community.
A challenge remains for the South African government: to maintain a policy balance between encouraging a risk management approach for large agricultural enterprises and providing a safety net for the resource-limited sectors of the population.
Was this article helpful?