Economic effects of GM crops: Reviewing farm-level evidence The case - PowerPoint PPT Presentation

Economic effects of GM crops: Reviewing farm-level evidence The case of Bt crops in South Africa by Marnus Gouse Department of Agricultural Economics, Extension and Rural Development, University of Pretoria Delivering Agricultural Biotechnology to African Farmers: Linking Economic Research to Decision Making 19-21 May 2009 Entebbe Uganda

Outline • Introduction – brief history of GM crops in South Africa • Socio-economic impacts: Farm-level impact • Bt cotton • Bt maize • Conclusion

History • South African Committee for Genetic Experimentation (SAGENE) formed in 1979 monitored and advised NDA on responsible development and use of modern biotechnology. • Approval of Bt cotton and maize under SAGENE’s guidelines and procedures in 1997/98 and 1998/99. • South Africa’s GMO Act no. 15 of 1997. • South African GMO Act 15 / 1997 amended in 2006 Commercial releases to date: • Bt cotton 1997/1998 • Bt yellow maize 1998/1999 • RR cotton 2001/2002 • RR soya-beans 2001/2002 • Bt white maize 2001/2002 • RR maize (white + yellow) 2003/2004 • Stacked cotton (Bt/RR) – 2005/2006 • Stacked maize (Bt/RR) – 2007/2008

SA has a dualistic agricultural sector Approximately <40 000 large-scale commercial farmers Approximately 240 000 small-scale farmers + Approximately 3 million subsistence farmers SA commercial farmers compare well with US farmers Small-scale SA farmers experience of special interest to rest of Africa Socio-economic studies – evidence to date • University of Pretoria (UP) – Rockefeller Foundation – Bt maize and cotton • University of Reading + UP – DFID - Bt cotton – small-scale farmers • French Research institute – CIRAD – Bt cotton • UP – Monsanto – Bt white maize – small-scale farmers • UP + MRC + Rutgers + Imperial College- Rockefeller Foundation – Bt + RR maize • Imperial College +UP + SOAS (Wye college) – DFID/ESRC - Bt + RR + Stacked maize

Bt Cotton Adoption of GM cotton in South Africa GM cotton area – percentage of total cotton area 100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 Total GM plantings % Bt Cotton % RR Cotton % Stacked cotton %

Adoption by small-scale and large scale farmers Small-scale cotton farmers over last 11years produced only 8% of total cotton crop – (1.2 – 20.9 per cent) Makhathini Flats in Northern KwaZulu Natal – one of only two areas where smallholders have produced cotton over last 20 years 1997 – 4 farmers Bt cotton adoption on M. Flats 1998 – 75 farmers 1999 – 411 farmers 2000 – 1184 farmers (about 40%) 2006 – about 90% GM but only 768 farmers Adoption enabled through production credit from local gin

Reasons for adoption / focus or study areas Adoption of Bt crop impacts farm income in mainly 3 ways • Increase in yield (decrease in loss to specific insect) • Decrease in input expenditure through saving on insecticide chemicals and application costs • Increase in input expenditure through higher seed price and/or additional technology fee

Bt Cotton Yield effect due to better bollworm control All SA studies found yield increases with Bt cotton compared to conventional cotton Field trails by Clark Cotton: 16.8% UP - Irrigation large-scale farmers: 18.5% - Dryland large-scale farmers: 14% UP and Reading – Small-scale Makhathini farmers 16-40% CIRAD – 23% (substantial but not statistically significant) Reading found in later study increases of up to 80% Yield advantage depends on: • bollworm pressure (season) • farmer’s pest control practises (chemicals and application) Varies from season to season, between regions and between farmers and fields within regions

Bt cotton Yield effect SA large scale farmers (14-18%) SA smallholders (23->80%) Relatively high yield effect compared to Australia (0%), US (9-11%), China (8-10%) but compares well with India (50+%) and Argentina (25-30+%) Linked to insecticide use – SA smallholder farmers use less than 50% of the optimal insecticide application level (Shankar and Thirtle, 2005). Average dryland smallholder yield over 11 years: only 553 kg seed cotton / ha Commercial dryland farmers 3-4 tons / ha and irrigation farmers > 6 tons/ha (Some countries’ Bt yield benefit more than SA smallholder total production)

Saving on insecticide expenditure and application cost Saving on insecticide chemicals to control bollworms • Large-scale irrigation farmers saved 56% • Large-scale dryland farmers saved 59% (Gouse et al, 2003) Smallholders: • Saved between 27% (Fok et al, 2007) and 53% and 63% (Bennett et al, 2006) • Hofs (2001) estimated a saving of 4 chemical applications - about 3.4 litres of endosulfan (organophosphate) / ha. Farmers still need to spray for sucking insects like jassids and aphids previously killed in the cross-fire aimed at bollworms Application costs: • Large-scale farmers save on diesel, mechanical wear-and-tear, labour and water • Smallholders save labour (water fetching and spraying) • Health benefits of spraying less But in most seasons, saving on insecticides were not enough to cover increased expenditure on Bt seed due to technology fee – importance of increased yield.

Cotton seed prices and technology fees in ZAR / 25 kg bag of seed Seed cost Bt Stacked Seed cost RR technology Year DeltaOpal technology technology NuOpal (Bt) fee (conventional) fee fee 1997/98 120 - - - - 1998/99 150 165 786.68 - - 1999/00 150 165 600 - - 2000/01 170 190 600 - - 2001/02 185 210 600 240 - 2002/03 215 250 600 300 - 2003/04 295 295 700 300 - 2004/05 350 350 700 350 - 2005/06 370 370 785 365 - 2006/07 390 390 785 365 1150 2007/08 410 410 785 365 1150 2008/09 430.50 430.50 785 365 1150

Who benefited? Distribution of additional benefit based on 1999/00 and 2000/01 data Small-scale Large-scale Large-scale Dryland Dryland Irrigation Farmers Farmers Farmers Seed company: D&PL 3% 2% 1% Technology supplier: Monsanto 28% 52% 20% Farmer 69% 45% 79% Primary Consumer: Ginning 0% 0% 0% company Insecticide companies R-90,600 R-777,700 R-1, 086, 400

Cotton production in South Africa 200,000 180,000 160,000 140,000 Hectares 120,000 100,000 80,000 60,000 40,000 20,000 0 2000/01' 2002/03' 2004/05' 2006/07' 1978/79 1980/81 1982/83 1984/85 1986/87 1988/89 1990/91 1992/93 1994/95 1996/97 1998/99

Makhathini Flats Technological triumph but institutional failure: Number of farmers have declined drastically mainly due to lack of credit Vunisa credit and input supplier on Flats – cotton crop is only collateral One buyer – loan recovery rate of 90% But in 2001/2002 a second gin opened on the Flats….. Farmers borrowed from one gin and delivered to other gin Collapse of credit scheme – few farmers able to finance own production 4000 3500 3000 2500 2000 1500 1000 500 0 1996/97 1997/98 1998/99 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 Total number of smallholders Number of smallholders on Makhathini

Maize • Maize is most important field crop in SA • Annually covers about 30% of total arable land • Since 2000 average production of 9.3 million tons on 2.75 million ha. • 60% white maize mainly for human consumption 6.000 4.5 4 5.000 3.5 Area (million ha) Yield (ton/ha) 4.000 3 2.5 3.000 2 2.000 1.5 1 1.000 0.5 0.000 0 2003/2004 1924/1925 1927/1928 1930/1931 1933/1934 1936/1937 1939/1940 1942/1943 1945/1946 1948/1949 1951/1952 1954/1956 1958/1959 1961/1962 1964/1965 1967/1968 1970/1971 1973/1974 1976/1977 1979/1980 1982/1983 1985/1986 1988/1989 1991/1992 1994/1995 1997/1998 2000/2001 2006/2007 Area Yield/ha

Bt maize African maize stem borer ( Busseola fusca ) and the Chilo stem borer (C hilo partellus) are the most harmful pests to maize and grain sorghum in South Africa (Kfir, 1997) Busseola causes between 5 and 75% damage on maize in Southern Africa every year. Busseola annually reduces South African maize crop by 10% on average. In SA: 10% damage = loss of just under a million tons (white and yellow) valued at 200 million US$ Smallholders: Small-scale farmers agree that stalk borers are the main pest But insecticide application levels differ dramatically between farmers and production areas Insecticide application to control stalk borers on maize not as vital as for bollworms on cotton. Especially for smallholders. = low insecticide saving benefit = low labour saving benefit

Percentage GM maize area in South Africa 60% 50% 40% 30% 20% 10% 0% 1999/00 2000/01 2001/02 2002/03 2003/04 2004/05 2005/06 2006/07 2007/08 Bt Yellow % Bt White % RR Yellow % RR White % Stacked yellow % Stacked white % Total white GM area % Total yellow GM area %

Smallholder yield effect Measure in kg grain yield / kg seed planted 350 21% 32% 30% 300 62% kg yield / kg seed plan 250 62% 200 34% 150 26% 100 50 0 All sites Northern Southern Hlabisa Venda Mqanduli Flagstaff Highveld Highveld Ow n Conventional isoline Bt