IN

VIII. INFORMAL SEED SECTOR

VIII.11. CONTRASTING INFORMAL WHEAT SEED SECTOR IN PAKISTAN AND AFGHANISTAN

For perhaps 20 years, a certain amount of modern seed and improved agricultural techniques has been transferred from Pakistan to Afghanistan by farmers, merchants, and development agencies. However, adoption of modern varieties appears to have been slow, perhaps in part due to lower availability of fertilizer, pesticides, extension services, and market services and demand. While the area planted to modern wheat varieties may be 15-20 years behind Pakistan, lack of government regulatory oversight has meant that NGOs and FAO were free to release modern varieties as soon as they deemed them to be suitable.

VIII.11.1. CAUTIONS ON EXTRAPOLATING EXTERNAL CONDITIONS

It is reasonable to suppose that many factors influencing these dynamics in Pakistan in the 1980's may only now begin to exert an impact on Afghanistan in similar ways. On the other hand, it may be that most Afghan wheat-growing areas are smaller and more compartmentalized topographically than in Pakistan; thus, "breakdown in variety resistance" may be less of an issue in the wheat-growing valleys and plains of Afghanistan.

It is worthwhile to review the factors affecting wheat variety replacement in Pakistan. However, it must be cautioned that a functioning formal seed sector system has been in operation for many years, and provides support and services for the informal sector. While study of many conditions which existed in Pakistan may help expedite improvement in Afghanistan, it must be noted that, by contrast, no formal seed sector exists.

VIII.11.2. PAKISTAN'S WHEAT SEED SUPPLY SITUATION

CIMMYT Research Report No. 1 compiles papers about the dynamics of wheat variety replacement in Pakistan in the 1980's.

In Pakistan, wheat varieties are classified as new recommended, old recommended, not recommended, and banned. In 1985-86, over 40% of Pakistani farmers were still growing banned varieties (SCA, 1990, p.18). These farmers gave four reasons for growing banned varieties:

Presumably, plant pathologists concluded that the banned varieties were more susceptible to rust, and therefore would consistently produce lower yields, and would also be sources of rust inocula that would tend to drag down the productivity of all wheat varieties planted in the area. In the late 1980s, the Pakistan government launched a campaign to reduce the area planned to banned varieties.

VIII.11.3. FACTORS AFFECTING RATE OF VARIETY REPLACEMENT

Byerlee and Heisey (1990) suggest that the rate of wheat variety replacement should depend on three factors:

"To make it worthwhile for a farmer to change varieties, some threshold yield increase-aside from the attractions of improved disease resistance and other superior characteristics that a new variety might have-is usually needed" (Byerlee & Heisey, 1990,p.5). They also noted that (1) the longevity of new resistant wheat varieties also decreases if they are grown alongside susceptible varieties"; (2) the average field life of a modern wheat variety in Pakistan was 6.3 years; (3) the variety Mexipak, released in 1965, had a field life of 7 years.

Verbal reports that Mexipak continues to be grown in some parts of Afghanistan in 2002-if verified-suggest that pressures for loss of resistance may not be as intense in parts of Afghanistan.

VIII.11.4. VARIETAL DIVERSIFICATION OVER SPACE

Byerlee and Heisey(1990) raise the issue of "appropriate varietal diversification over space." "To actually reduce the risk of an epidemic [of rust on wheat], diversification must be practiced at the field level, perhaps by planting multi-lines or varietal mixtures in a given field. There is some difference of opinion over whether planting adjacent fields with genetically different material slows the spread of the disease, but plant pathologists generally agree that diversification above the field level* will probably not influence the rate at which a rust epidemic spreads(Byerlee and Heisey, 1990, p.9).

Table 39 : Indices of Wheat Variety Replacement

Region	Weighted average age of varieties (years)	Areas planted to new varieties (%)
Afghanistan	?	?
Punjab, Pakistan	11.0	11
Punjab, India	6.5	30
Yaqui Valley, Mexico	4.1	85
(adapted from Byerlee and Heisey, 1990, p.17)

VIII.11.5. PENETRATION OF MODERN VARIETIES

In 2000, surveys in selected districts of three Afghan provinces on the main transport corridor between Pakistan and Kabul found the percentage of villages growing modern wheat and maize varieties to be only 32% and 9%, respectively, compared to 84% of these same villages reporting use of chemical fertilizer. Although modern wheat varieties Ataya 85, Bakhthawar, Inqilab 91, Pamir 94, and Pirsabak 85, and modern maize varieties, Sarhad and Shaheen, had been in circulation in these provinces for some years, actual adoption of these varieties was reportedly virtually non-existent in more than two-thirds of the villages sampled (UNOPS, 2000).

The study findings appear to suggest that farmer non-adoption of modern wheat and maize was considerably more widespread than, say, modern wheat being limited to 32% of wheat area and modern maize being limited to 9% of maize area. On the other hand, there may have been some villages where modern wheat and maize covered much or even the entire wheat and maize area. When asked about the relative non-adoption of modern wheat and maize in the three provinces samples, survey director M. Omar Anwarzay replied that villagers didn't like the eating quality of modern maize from in Pakistan (Conversation with J. Dennis, March 2002). For wheat, he suggested that Afghan farmers were unreliable informants when it came to wheat varieties and might report long-established modern varieties such as Mexipak as "local varieties" and may have given them local names. As an example of the Afghan penchant for renaming things, Jahan Dawlaty, an Afghan graduate student at Cornell University, cited an Afghan village where the soft drink, Pepsi Cola, is now called "BBC." Apparently, villagers who had listened to BBC while working in a Gulf State decided that it was a catchier name.

This, and the resultant effect on productivity, is probably due to conditions created by the extended period of conflict, in which there was practically no extension education and support to farmers, only limited variety development and/or introduction, and very limited "formal" seed production conducted by the FAO emergency intervention seed program.

In a small windshield survey in Logar and Kabul provinces carried out by the mission in May 2002, of 22 farmers sampled who planted wheat in 2001-2002, 15 or 68% only planted modern varieties, 4 or 18% only planted local varieties, and three or 14% planted both modern and local varieties. In other words, with 82% of the farmers growing some or all their wheat area to modern varieties, penetration of modern varieties to these irrigated areas along the Kabul-Logar Highway can be considered to be very high.

Varieties mentioned by these farmers are summarized in the following two tables.

Table 40 (PDF File 52Kb)
Wheat Varieties Found Growing in Kabul and Logar Provinces in May 2002

Table 41 (PDF File 51Kb)
Other Wheat Varieties Also Grown During the Past 10 Years in Kabul and Logar Provinces

VIII.11.6. SEED-TO-GRAIN PRICE RATIO

Byerlee and Heisey (1990) note that seed-to-grain price ratio has two effects. On the one hand, a high seed-to-grain price ratio is a disincentive to farmers to buy new seed frequently. On the other hand, a high seed-to-grain price ratio is an incentive to private sector seed companies to produce seed of new varieties and multiply seed of varieties that are in demand. During the late 1980s, the seed-to-grain price ratio in Pakistan was about 1.5. This contrasts with a seed-to-price ratio of 1.1 that NGO's working in Afghanistan have implemented over the past ten years.

VIII.11.7. SOURCE OF SEED

Telay et al. (1999. p. 63) found that most frequently, a Pakistani wheat farmer was his own source of seed. The second most important source of wheat seed was other farmers. Less than 10% of farmers surveyed obtained seed at a government outlet in a given year.

Fewer than half of the farmers who retained their own seed sought to ensure its purity by selecting or threshing it separately.

When farmers change wheat varieties in Pakistan, other farmers are the most important source of seed.

A small windshield survey of 28 rural households in Logar and Kabul provinces found that the sample farmers relied more-or-less equally on themselves, merchants, and NGO's for wheat seed. Other farmers ranked a distant fourth. None of the sample farmers admitted to selecting actual panicles for seed, but a number did admit to growing seed plots that got more frequent irrigations or otherwise enjoyed better production conditions.

VIII.11.8. FARM SIZE

In Pakistan, farm size appears important in several ways. First, wheat farmers with larger holdings are more likely to source their new variety seed from a government outlet than are smaller farmers (Telay et al, 1990). Second, larger farmers are more likely to adopt new varieties when the overall rate of diffusion is relatively slow. Thirdly, larger farmers are more likely to be partial adopters, growing old varieties along with new ones. Furthermore, large farmers are apt to be literate and have better extension contacts. When diffusion begins to proceed more rapidly, the adoption behavior of larger and smaller farmers differs little except for larger farmers' greater tendency to diversify. Smaller farmers lag behind, not because they are reluctant to change, but because they are less likely to have full information about the new varieties in the initial stages of adoption because of lower literacy and few extension contacts (Heisey, Telay, et al., 1990, p.83).

This further supports the view that farmers' financial means dictates or strongly influences their choice of seed. More well-to-do farmers seek to produce the highest economic yields, while poorer farmers seek to "produce at the lowest input cost".

In a sample of 15 farms in Logar and Kabul, classification of wheat varieties grown by small farms was tri-modally distributed between all traditional (4), half traditional/half modern (2), and all modern (5). The remaining four farms that were larger than 2 hectares (>10 jeribs) used all modern varieties (3) and two out of three varieties modern (1). At this small sample size, there appeared to be no correlation between size of farm and number of varieties used.

VIII.11.9. "AMBIDEXTROUS" FARMERS

This term is used here to refer to farmers who grow both local and modern varieties of wheat. This is potentially one of the most interesting groups of farmers to interview in Afghanistan. In theory, future research would want to distinguish between "incremental adopters"-farmers who test newly-obtained modern varieties on a part of their land prior to becoming 100% users of modern varieties-and true "ambidextrous farmers" who have used modern varieties for a number of years and yet continue to think that (1) local varieties are better-suited to a portion of their land; or (2) growing modern and local varieties in the same season offers other advantages when compared to growing all modern or all local varieties.

In Pakistan, "When farmers of old and new varieties were asked why they planted both types, the most common response was that they wanted to compare the performance of each type. In addition, 30-40% of the farmers planted both types for reasons related to diversification: they wanted one variety for late planting or for home consumption, or simply wanted to avoid risk"(Heisey, Telay, et al. 1990. p. 68).

As mentioned above, only three farmers interviewed in May 2002 in Kabul province and Logar provinces were found to be ambidextrous farmers. All three farmed relatively close to Kabul and two rented in all their land. Their profiles are summarized in Table 42 below.

Table 42 (PDF File 51Kb)
Profiles of three ambidextrous wheat farmers in Kabul province

VIII.11.10. FACTORS AFFECTING RATE OF FARMER ADOPTION OF MODERN WHEAT VARIETIES

Education. In Pakistan, Education or literacy appears to be the most important characteristic related to information about new varieties, seed depot location, and so on. Extension contact is more ambiguous (Heisey, Telay, et al. 1990. p. 83). However, while education seemed to play a role in earliness of adoption, the effect of education on proportion of modern and local varieties grown dropped out over time.

In Logar and Kabul provinces, education did not appear to have any clear impact on adoption behavior. Of 27 farmers interviewed, 19 were illiterate and 8 had education ranging from 4 years to 12 years. Seven out of eight or 88% of the educated farmers and sixteen out of 19 or 84% of the illiterate farmers planted some or all of their wheat area to modern varieties.

One farmer who grew three local varieties of wheat and no modern wheat was a teacher and was the only one of 28 persons interviewed in Logar and Kabul provinces who admitted to the use of family planning in his household.
In Pakistan, "[Farmers] did not identify variety as a major component [for increasing wheat productivity]. Better land preparation ranked highest and more fertilizer second; variety ranked between third and fifth"(Heisey, Telay, et al. 1990, p.66).

VIII.11.11. TENURE AND OTHER FACTORS INVOLVED IN VARIETY DIFFUSION

In Pakistan, "Tenure, extension contact, and village effects appeared to play no role in the actual adoption of new varieties. …farmer-to-farmer transfer of both information and seed plays a major role in seed diffusion. …Other farmers are an important source of seed; other farmers are the most important source of information about new varieties; other farmers are a major source of seed of new varieties; and the proportion of wheat under new varieties in a village influences farmers' awareness of new varieties"(Heisey, Telay, et al. 1990, p.82).

In Logar and Kabul provinces, only 4 of 27 farmers interviewed, or 15%, rented land and there was no indication from this tiny sample that tenure status affected choice of varieties. In Kulangar Waz Khan village in Logar province, a teacher (mentioned in the previous section) with a high school education and 10 children rented one jerib to supplement his owned 3.5 jeribs of land. He planted White Leaf, Red Leaf and Kalak varieties of local wheat and no modern varieties. He had formerly owned a cooking gas retail shop, but it had failed when not enough of his customers were able to pay for gas that had been sold on credit. Another farmer, with a 4th-grade education, rented 40 jeribs (8 ha) on the outskirts of Kabul and planted his land with one local variety, Red Leaf, and with Roshan and Darluma 4 modern varieties. The two remaining farmers who rented land, rented 12 and 13 jeribs, and were both illiterate. One planted only the modern variety Gul 96, while the other planted Red Leaf local variety and Ataya, a modern variety.

VIII.11.12. THE "MIXTURE EFFECT" LITERATURE

In reviewing Pakistan's wheat breeding program in the late 1980s, Byerlee and Heisey (1990) concluded that release of new varieties would have been better able to keep pace with sector requirements if modern varieties had not tended to lose resistance to rust so quickly. The authors accepted the idea that the planting of banned varieties hastened the emergence of new virulent strains of rust by enabling the presence of larger amounts of rust inoculum on the landscape, but they also suggested that a multi-line approach to rust control might be effective.

A top-down approach to limit farmer use of rust-susceptible wheat varieties would likely be more difficult to implement in Afghanistan due to greater problems of access to a disparate array of wheat-growing areas.

A growing body of research reports that when two or more crop varieties are grown in close proximity and neither has complete resistance to a particular disease, a "mixture effect" tends to occur in which each variety in the mixture appears to confer additional resistance on the other varieties in the mixture. Mixture effects have been documented for beans (Trutmann et al. 1993 and 1994; Mikishi and Trutmann, 1992), rice (Zhu et al. 2000), and barley (Chin & Wolf, 1984) and are reviewed by Wolfe (2000 and 1985). Zhu et al.'s (2000) research on rice mixtures in China was carried out on thousands of hectares in an already highly productive rice-growing area. A two-variety mixture not only out-yielded a hybrid rice grown alone, but also removed the need for pesticides.

This literature suggests that there is potential for synergy between the informal and formal seed sectors as long as each is able to tap into the strengths of the other.

As suggested in the SCA report cited earlier, variety mixtures appear to have been a modus operandi in the informal seed sector in Afghanistan for many years, most probably due to lack of good techniques to create or maintain single-variety seed populations. Without the incorporation of modern genes for fertilizer responsiveness and control of lodging, and with apparent absence of land race enhancement by pure line selection at government research stations, the potential for utilizing the existing crop biodiversity in Afghanistan in the formal seed sector appears to have gone largely untapped. FAO reports on seed multiplication in Afghanistan (N.S. Tunwar, 1998 & 2000) mention not a single local variety wheat. Reports received also indicated that the introduced improved varieties out-yielded local varieties.