68418 v6 I. Introduction Gross agricultural output in Tajikistan grew by 64% in real terms from 1999 to 2003. Increased crop and livestock productivity and higher producer prices were the main source of the growth, whereas the broad structure of crop and livestock output changed little during this five-year period. Sector and economy-wide changes also influenced agricultural growth, including political stability and steady economic growth. This technical annex presents the result of a study that used a growth decomposition model to examine the relative contribution of prices and productivity changes to agricultural growth. To provide a deeper insight into sources of growth, the analysis is decomposed into two time periods (1999-2001 and 2001- 2003), and looks at observed growth in crops, livestock and horticulture and at individual and collective (dehkan and state-owned) farms separately. Further insight is derived by separate analyses of the four main regions (Khatlon, Sogd, GBAO and RRP). The model includes all major crop and livestock commodities, with the exception of cotton, and accounts for approximately 75% of observed sector growth. Cotton, which accounts for 66% of the value of crop production and 53% of overall production, is analyzed in a separate study. The value of cotton output contracted significantly from 1999 to 2003 due to a fall in international prices and the results of growth decomposition analysis are difficult to interpret for “negative growth.” The main characteristics of agriculture sector growth can be summarized as follows: 1. Crop production accounted for 63% of observed growth and livestock production for 37%. This largely reflects the larger contribution of crop production to overall sector output (81%). Yield increases were the major source of growth in crop production, followed by increased producer prices. There was little change in crop area. 2. Growth in livestock production was driven largely by higher prices, in response to increased demand on both domestic and regional prices, but higher productivity and increased livestock numbers also contributed. 3. Horticulture production (fruit, berries and grapes) was erratic due to marked variations in yields and prices. Higher yields appear to lead quickly to lower prices. 4. Household agriculture production accounted for 69% of observed growth and collective (Dehkan and state-owned) farms for 31%. This is consistent with the traditional dominance of household agriculture, which accounts for approximately 55% of total farm output. Increased crop production largely drove this increase in household output. 5. Although there was a substantial growth in Dehkan farm output from 1999 to 2003, this was mainly due to farm restructuring and a transfer of assets from state-owned collective farms to Dehkan farms. The corresponding decline in state farm output substantially offset Dehkan farm growth and lowered overall growth among collective farms. 6. Both farm types responded positively to overall economic growth and political stability with increased crop yields and increased livestock numbers. This response was much more muted among collective farms however, with livestock numbers in particular increasing at a much lower rate. This suggests that individual farmers have much greater incentives to increase livestock numbers, as they are able to appropriate the returns. 7. Khatlon and RRP contributed most to non-cotton agriculture sector growth (39% and 36% respectively), commensurate with their overall contribution to sector output. The minimal contribution of GBAO is consistent with its limited agricultural resource base, but Sogd’s contribution is lower than expected. On a per hectare basis GBAO obtained the highest rate of non-cotton growth, followed by RRP. Regional patterns of growth in crop production differed markedly over time. ƒ In Khatlon, initial growth in 1999-2001 was driven by increased prices. This led to a substantial increase in yields and output from 2001 to 2003, and a subsequent fall in prices. GBAO experienced a rapid crop expansion in 1999-2001, accompanied by a fall in prices; with relatively low subsequent growth in 2001-2003. In both regions this pattern of increased production followed by falling prices demonstrates the risks incurred by expanding production in regions where local markets are depressed or easily saturated, and transactions costs are high. ƒ More sustainable increases in production occurred in RRP and Sogd, which have access to larger markets (Dushanbe and Khujand with its processing capacity and proximity to regional export markets). In both cases increased production in 1999-2001 was followed by higher or stable prices in 2001-2003 and so further increases in production. 2 II. The Context for Agricultural Growth Tajikistan’s agriculture sector experienced a prolonged decline after independence in 1990. Growth resumed in 1999, with gross agricultural output (GAO) increasing by 64% from 1999-2003 (Figure 1). Most of this growth occurred from 2001 to 2003. The crop sector, which accounts for 81% of output, grew by 65% from 1999-2003; while livestock, which account for 19% of sector output, grew by 61%. Household plots were the main source of output and growth (Figure 2). Household GAO increased by 1.6 times from 1999-2003, with strongest growth from 2001-2003. By 2003, household plots accounted for 94% of gross livestock output and 44% of gross crop output. Dehkan farm output increased faster than household plots, however their contribution to total sector output remained modest (Figure 2). A fourfold increase in dekhan farm GAO from 1999-2003, raised their contribution to sector output from 10% to 24%. Most of this growth occurred from 1999-2001, and was driven by increased crop production. Dekhan livestock production is low. Figure 1. Gross Agricultural Output Value (in Figure 2. Composition of Gross Agriculture constant 2000 prices) Output 60 1 200,0 50 1 000,0 In m illio ns o f So m o n i 40 Percent 800,0 30 600,0 20 400,0 10 200,0 0 1996 1997 1998 1999 2000 2001 2002 2003 0,0 Years enterprises Other agricultural 1996 1997 1998 1999 2000 2001 2002 2003 Dekhkan farms Years Household plots The structure of crop production changed from 1999-2003, with an increase in the area of cotton and potatoes and a reduced area of grain, vegetables and melons. Cotton cultivation increased by 15%, although it remained below the area sown in 1991. Potato cultivation increased by 27%, with most of this growth from dekhan farms and other larger farms. For both crops, most of the increase in area cultivated occurred from 2001-2003. Grain cultivation decreased from 1999-2001, but then increased after the drought in 2000. Increased yield was the major determinant of increased crop production, except for tobacco where both yield and output declined. Household plots obtained the highest yields for most crops. Dekhan farms achieved the highest yields for potatoes and fine-staple cotton (households do not grow cotton); and state and collective farms obtained the highest yields for vegetables. The impact of yield and other factors on output growth is discussed in more detail in Chapter 4. Horticulture production increased from 1999-2001 but then declined sharply from 2001-2003 due to falling yields of fruit, berries and grapes. Household plots account for 70% of horticulture output value. Despite these changes in production, the overall composition of crop output value changed little from 1999 to 2003 (Figs 3 and 4). The share of cotton, grain and potatoes increased slightly; and the share of fruit, berries, grapes and tobacco fell. 3 Figure 3. Structure of Crop Output Value, Figure 4. Structure of Crop Output Value, 2003 1999 Fruit, Fruit, berries & berries & grapes Vegetables grapes 4% Tobacco & melons 2% Tobacco Vegetables & melons 2% 5% 0% 6% Potatoes Potatoes 5% 4% Wheat, rice & corn Wheat, rice 22% & corn 20% Cotton Cotton 64% 66% Note: Constant 1999 prices were used to calculate the crop output value for both years. The structure of livestock output changed in response to increased poultry numbers and a marked reduction in the number of sheep and goats. This decrease was greatest in state, collective and dehkan farms. These changes contributed to a decline in output value for beef and veal, and increased output value for eggs, mutton and lamb. Milk remained the largest component of the value of livestock sector output. Figure 5. Structure of livestock commodity Figure 6. Structure of livestock commodity output value, 1999 output value, 2003 Wool Eggs Eggs Wool 1% Beef and Beef and 1% 4% 1% veal veal 24% 17% Milk Mutton and 54% lamb Mutton and Milk 24% lamb 54% 20% Note: Constant 1999 prices were used to calculate the livestock output value for both years. 4 III. Methodology Growth decomposition analyses the contribution of key parameters to growth. The parameters of interest for this analysis were cropped area and livestock numbers (the expansion effect), crop yield and livestock performance (the productivity effect) and producer prices. For purposes of analysis the impact of each parameter was determined separately, while holding the other parameters constant, and this process was repeated for each parameter. Growth was measured as the change in real gross value of agricultural Figure 7. Gross Output Value versus Value Added for production for the period 1999-2003. Agriculture While trends in GAO overstate actual growth, as measured by change in 2 500 agriculture value-added (Figure 7), GAO was used because it is easier to 2 000 Millions of Somoni decompose. 1999 was chosen as the starting point for analysis as political 1 500 and economic stability had been re- established following the end of civil 1 000 war in 1998. At the time of analysis, 2003 was the year for which most 500 recent data were available. The resultant 5 year time-period was then 0 arbitrarily split into two sub-periods 1999 2000 2001 2002 2003 (1999-2001; 2001-2003), to provide Years insight into changing patterns of growth over time. Data were drawn from official statistics on agricultural Value added Gross output production and prices (Annex 1), with prices deflated to remove the influence of inflation. Observed growth was decomposed by sub-sector (crops, livestock, horticulture); region (Sogd, Khatlon, Gorno-Badakhshan autonomous oblasts (GBAO), and raions administratively subordinated to republican level (RRP); and farm type (household plots and other farms). Both dekhan farms and state and collective farms were grouped together as “other farms” for purposes of analysis, as the transformation from one to the other was still on-going during the period of analysis. Trends in cotton production are analyzed separately from the growth decomposition analysis (see next section). Cotton output contracted from 1999-2003 and this analytical technique is only effective for positive growth. The decomposition analysis thus explains 95% of the growth in GAO for non-cotton products, and 47% of all agriculture sector output. 5 IV. Trends in Cotton Production The gross value of cotton production declined, in both nominal and real terms, from 1999 to 2003 (Table 1). As cotton output actually increased by more than 70% during this period, the falling value of cotton output was due to a significant fall in world prices for cotton fiber and the impact of this price fall on domestic producer prices1. Table 1. Cotton Production in Tajikistan Indicators 1999 2000 2001 2002 2003 Nation Sown area (‘000 hectares) 247,8 238,6 257,4 269,2 284,0 Output (‘000 tons) 313,1 335,4 452,7 515,5 537,4 Yield (centners per hectare) 12,6 14,1 17,6 19,3 19,0 Real price (Somoni per ton) 46,3 40,9 21,4 18,0 25,1 Value of cotton output (‘000 Somoni) 14497,8 13733,3 9695,5 9255,1 13463,8 Sogd oblast Sown area (‘000 hectares) 72,1 69,1 75,3 80,8 85,5 Output (‘000 tons) 115,1 121,0 145,3 154,9 148,5 Yield (centners per hectare) 16,0 17,5 19,3 19,4 17,6 Value of cotton output (‘000 Somoni) 5329,9 4952,4 3112,9 2780,6 3719,6 Khatlon oblast Sown area (‘000 hectares) 155,0 150,3 160,5 165,4 173,5 Output (‘000 tons) 174,4 178,0 255,1 305,8 334,1 Yield (centners per hectare) 11,1 11,8 19,9 19,4 19,3 Value of cotton output (‘000 Somoni) 8075,7 7285,8 5463,3 5489,6 8369,5 RRP Sown area (‘000 hectares) 20,7 19,2 21,6 23,0 25,3 Output (‘000 tons) 26,4 36,5 52,3 54,9 54,8 Yield (centners per hectare) 12,7 19,0 24,2 24,9 21,7 Value of cotton output (‘000 Somoni) 1220,6 1495,2 1120,0 984,8 1373,7 The increase in cotton output was driven by a 15% increase in the area sown and a 51% increase in cotton yields. Increased yield was the major determinant of increased production from 1999-2001, and area expansion from 2001-2003. The largest increase in cropped area occurred in the RRP (22%) and Sogd oblasts (19%). Figure 8 shows the magnitude of this increase in production and the extent to which it was offset by the impact of falling prices on the value of cotton output. Figure 8. Trends in Cotton Production and Cotton Output Value 600 16 14 500 s 12 ni n o d to 400 m 10 o n nS 300 8 sa illio u 6 o 200 M Th 4 100 2 0 0 1999 2000 2001 2002 2003 Years Cotton production Value of cotton output 1 Producer prices are determined by adjusting world cotton fiber prices back to equivalent domestic seed cotton prices, using actual domestic processing and marketing margins. 6 Given that cotton accounted for 66% of the value of crop output in 2003, this decline in the value of cotton output dominated trends in overall crop production. This is evident from Figures 9 and 10 below, which decompose growth with and without cotton. Figure 9 shows that price changes had a positive impact on growth when cotton is excluded, while Figure 10 shows that prices had a negative impact – due to the decline of cotton prices. Cotton was thus excluded from the decomposition analysis, to allow a clearer picture of trends in other commodities. Figure 9. Sources of Crop Figure 10. Sources of Crop Growth for Commodity Mix Without Cotton Growth for Commodity Mix With Cotton 1999-2001 1999-2001 Period Period 2001-2003 2001-2003 1999-2003 1999-2003 -50 0 50 100 150 -100 -50 0 50 100 150 Share(%) Share(%) Prices Expansion Productivity Prices Expansion Productivity 7 V. Growth Decomposition Analysis Growth decomposition analysis focused on the impact of three factors: Prices (real producer prices) Productivity (crop yields and livestock performance) Expansion (crop areas and livestock numbers). The impact of these factors on sector output is examined first at national level, and then further decomposed by farm type (household plots versus larger individual and collective farms) and region (the four oblasts). 5.1 Crop and Livestock Production Crop production was the major engine of agriculture sector growth. It grew steadily from 1999-2003 and accounted for 63% of observed growth in the value of output (Fig. 11). Livestock output also grew steadily, but at a much lower rate, accounting for 37% of total growth. Growth was evenly shared between 1999-2001 and 2001-2003. Figure 11. Contribution of Crop and Livestock Figure 12. Sources of Agricultural Growth Production to Output Growth 33% 18% 1999-2001 1999-2001 Period Period 30% 19% 2001-2003 2001-2003 1999-2003 63% 37% 1999-2003 0% 50% 100% 0 20 40 60 80 100 Share Share, % Crop output Livestock output Prices Expansion Productivity Increased productivity was the major source of growth, with higher crop yields being the dominant factor (Figure 12). Higher real product prices were the second most important source of growth, exerting a strong influence during the first period. Further analysis by sub-sector shows that higher yields were the main source of crop growth, and that the expansion effect was slightly negative in 1999-2001 due to a contraction of the area planted (Figure 13). In contrast, livestock sector growth was driven largely by increased producer prices, with increased livestock numbers and higher productivity making a lesser contribution (Figure 14). 8 Figure 13. Sources of Crop Growth Figure 14. Sources of Livestock Growth 1999-2001 1999-2001 Period Period 2001-2003 2001-2003 1999-2003 1999-2003 -50 0 50 100 150 0 20 40 60 80 100 Share(%) Share(%) Prices Expansion Productivity Prices Expansion Productivity 5.1.1 The Impact of Farm Type Household plots accounted for more than two thirds of agricultural output growth, dominating growth in both the crop and livestock sub-sectors (Figure 15) Figure 15. Contribution of Crop and Livestock Production to Output Growth by Farm Type 29% 3% Other Farm type HHs 40% 29% Total 69% 31% 0% 20% 40% 60% 80% 100% Share Crop production Livestock production Further analysis shows that increased crop production accounted for more than 90% of the growth observed for other farms and almost 60% of the growth on household plots (Figure 16 and 17). It also shows that these patterns were constant for both the 1999-2001 and 2001-2003 time periods. 9 Figure 16. Contribution of Crop and Livestock Figure 17. Contribution of Crop and Livestock Production to Output Growth for Household Production to Output Growth for Other Farm Plots Types 57% 43% 1999-2001 89% 11% 1999-2001 6% Period Period 2001-2003 59% 41% 2001-2003 94% 1999-2003 58% 42% 92% 8% 1999-2003 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Share Share Crop output Livestock output Crop output Livestock output Increased productivity is the major source Figure 18. Sources of Agricultural Growth by Farm of growth for both farm types, but its Type impact is much greater for household plots than for other farms (Figure 18). Larger price and expansion effects also result in more balanced growth for household plots, Other compared to other farm types where these Farm type factors account for less than 10% of overall growth. Note that crop “expansion effects” HHs are also associated with changes in crop composition. Among other farms the area cropped fell for all crops except potatoes, Total which increased more than 1.5 times. On household plots, there was a reduction in 0% 20% 40% 60% 80% 100% the area sown to rice, grain corn, melons and vegetables; but this was compensated Share by increased cropping of potatoes, fruit, berries and grapes. Prices Expansion Productivity The relative importance of these factors also varies over time, as demonstrated in Figures 19 and 20 below. Expansion was the main influence on growth among household plots from 1999-2001, while productivity increases drove growth from 2001-2003. Productivity and price effects drove growth among other farms from 1999-2001, but productivity gains were the main influence from 2001-2003. The changes in sown areas mentioned above explain the changing importance of the expansion effect for different farm categories and different sub periods. 10 Figure 19. Sources of Agricultural Growth for Figure 20. Sources of Agricultural Growth for Household Plots Other Farm Types 1999-2001 1999-2001 Period Period 2001-2003 2001-2003 1999-2003 1999-2003 0 20 40 60 80 100 -50 0 50 100 Share Share (%) Prices Expansion Productivity Prices Expansion Productivity Decomposition by sub-sector shows that yield increases drove crop output growth for all farm types, with price increases as a lesser influence (Fig 21). The crop expansion effect was zero for household plots and slightly negative for other farms. Further analysis showed that yield increases accounted for 77% of crop output growth for household plots, and price increases for 22%. Growth was broadly based from 1999-2001, but was then driven largely by yield increases from 2001-2003. A similar pattern occurred for crop production by other farms. Figure 21. Sources of Crop Growth by Farm Figure 22. Sources of Livestock Growth by Farm Type Type 6% -1% 35% 0,2% Other Other 8% 2% Farm type Farm type 12% 44% 22% 47% 23% HHs HHs 1% 21% 80% 24% 53% 23% Total Total -50% 0% 50% 100% 150% 0% 20% 40% 60% 80% 100% Share Share Prices Expansion Productivity Prices Expansion Productivity Livestock sector growth was due mainly to increased livestock numbers (expansion), for all farm types (Fig. 22). It accounted for 51% of growth for household plots and 72% of growth for other farms. On other farms the expansion effect was due mainly to increased poultry numbers (2 times) and more pigs (1.6 times,) as the inventory of most other livestock actually fell. Pronounced differences were evident in the nature and magnitude of growth, however, with low growth among other farms; and strong, more balanced growth from household plots. Note also that other farms exhibited no discernable increase in productivity, while productivity growth occurred in addition to inventory growth for household plots. 11 5.2 Regional Analysis Regional characteristics of agricultural output are summarized below to provide the context for regional analysis of agriculture sector growth. Figures 23 and 24 describe the contribution of each region to total sector output in 2003, with and without cotton. As expected, the two major agricultural regions, Khatlon and Soghd, are the major sources of sector output, followed by RRP and then GBAO. The major contribution of Khatlon to sector output indicates how important it is to overall sector growth. Soghd and RRP both have more diversified agricultural production bases than Khatlon, and so are less dependent on cotton production. The limited agriculture resource base in GBAO is reflected in its small contribution to total output. Figure 23. Gross Agricultural Output Value with Figure 24. Gross Agricultural Output Value Cotton by Region, 2003 without Cotton by Region, 2003 GBAO GBAO 2% 4% RRP 18% RRP 25% Khatlon 40% Khatlon 51% Soghd 29% Soghd 31% Within regions, crop composition changed little from 1999 to 2003. The share allocated to cotton increased slightly in all cotton growing regions, with the largest increase (4%) occurring in Khatlon. In most regions this increase was offset by a decline in the share of fruit, berries, grapes and fodder. The share of cultivated land allocated to potatoes also increased slightly in Soghd, RRP and GBAO, with the largest increase (4%) in GBAO. The grain share increased in Soghd oblast in addition to cotton and potato shares, but this was offset by a reduced area of vegetables – in addition to less fruit, berries and grapes. Only in RRP was there an increase in the share of fruit, berries and grapes. This was offset by lower shares of grain, legumes and vegetables. There was no change in the share of fodder crops, despite their importance to overall sector production. Indeed, in Khatlon and GBAO the share of fodder crops actually decreased. Regional differences in growth thus reflect regional differences in productivity, prices, and livestock numbers, and the differing responses of household plots versus other farm types. 5.2.1 Regional Growth Patterns Regional contributions to agriculture sector growth differ somewhat from their share of output (Figure 25 and 26). While Khatlon accounts for the largest share of overall sector growth, commensurate with its share of output; Soghd’s contribution to growth is much lower than expected and RRP makes a very strong contribution – comparable to that of Khatlon. With the exception of RRP, the value of agriculture output grew most in the first period and then slowed from 2001-2003. The slower growth in 2001-2003 was particularly marked in Khatlon. Crop production was the major source of growth in all regions, particularly in Khatlon and RRP. 12 Figure 25 Regional Contributions to Output Figure 26. Regional Contributions to Output Growth by Time Period Growth by Sub-Sector (1999-2003) 40% 26% 10% 35% 10% RRP 30% 23% 3% 25% GBAO Regions Share 2% 20% 7% 15% 29% 31% 8% Khatlon 10% 13% 13% 5% 2% 12% 9% 3% Sogdi 0% Sogdi Khatlon GBAO RRP 0% 10% 20% 30% 40% Regions Share 1999-2001 2001-2003 Crop production Livestock production Analysis of per hectare growth rates shows a different picture, with GBAO exhibiting the highest growth rate on a unit basis, followed by RRP (Table 2). Khatlon and Soghd had much lower growth rates, excluding cotton, and negative growth including cotton. Note that the value of output per ha in GBAO was still extremely low, despite the high observed growth rates, due to its limited agricultural potential. The analysis is useful nevertheless in demonstrating the potential for growth in all regions, particularly for commodities other than cotton. Table 2. Regional Unit Growth Rates 1999-2003 (change in value of output/haa) With Cotton Without Cotton Khatlon -23.5% 50.0% Soghd -35.8% 52.3% RRP 10.1% 97.3% GBAO 238.3% 238.3% National -21.6% 63.3% a In real prices (1995=100) 13 Increased yields and higher prices were the main Figure 27. Sources of Growth by Region (1999- determinants of growth in all regions, except for GBAO (Fig 27). The expansion effect was limited in 2003) all regions. However the relative impact of these factors varied considerably by time period. Price changes were the main determinant of growth RRP in Khatlon from 1999-2001, but then became a negative influence on growth from 2001-2003 GBAO Regions (Figure 28 and 29). This was due to a decline in the real price of milk and many crop commodities. In contrast, yield and expansion effects drove growth in Khatlon RRP during the first period, but price increases became the major influence in the second period. Sogdi The differing impact of prices, by time and by region, indicates both the advantages of expanding 0% 10% 20% 30% 40% production in an environment of growing regional Share markets (RRP), and the dangers of expanding production when regional markets are weak Prices Expansion Productivity (Khatlon). Figure 28. Sources of Growth by Region (1999- Figure 29. Sources of Growth by Region (2001- 2001) 2003) RRP RRP -0,3% GBAO GBAO Regions Regions -1% Khatlon Khatlon Sogdi Sogdi -20% 0% 20% 40% 60% -20% 0% 20% 40% 60% Share Share Prices Expansion Productivity Prices Expansion Productivity Crop growth was driven largely by yield increases in all regions (Fig. 30). Prices also had a generally positive impact, although this varied by time period and by region as noted above. In Khatlon, prices had a strong positive impact in the first period and a negative impact in the second period; while the opposite pattern occurred in Soghd. The negative expansion effect from 1999-2003 in Soghd was due to a decline in the area sown to vegetables, maize, rice and melons. Livestock growth was also driven by increased productivity, in all regions except for Sogdi where it exerted a negative influence (Fig. 31). This fall in productivity occurred during the first period, but even in the second period there was no observable increase in livestock productivity in this region. The strong livestock expansion effect observed in Soghd was largely a result of increased poultry numbers, a trend also apparent in Khatlon and RRP. In general it appears that farmers in Soghd have chosen to increase returns to livestock by increasing livestock numbers, while farmers in the other three regions have chosen to increase productivity. 14 Figure 30. Sources of Crop Growth by Figure 31. Sources of Livestock Growth by Region Region RRP RRP GBAO GBAO Regions Regions Khatlon Khatlon Sogdi Sogdi -10% 0% 10% 20% 30% 40% 50% -10% 0% 10% 20% 30% 40% Share Share Prices Expansion Productivity Prices Expansion Productivity Analysis of the impact of farm type on regional agriculture sector growth was limited to crop production. (No analysis was made of livestock as household plots account for 94% of livestock ownership). Figure 32 and 33 below show that crop growth in Khatlon was driven by yields first and prices second, for both farm types. A more balanced combination of factors contributed to growth in RRP, increased yields and crop expansion drove crop growth in Soghd, and crop expansion was the main source of crop growth in GBAO. In all regions, price and expansion factors were the dominant influence on crop growth from 1999-2001; and crop productivity growth and price increases drove crop growth from 2001-2003. Khatlon farmers faced declining prices in 2001-2003, however, which meant that most growth was due to increased yields. Figure 32. Regional Sources of Crop Growth Figure33. Regional Sources of Crop Growth for for Household Plots (1999-2003) Other Farm Categories (1999-2003) RRP RRP GBAO GBAO Regions Regions Khatlon -0,4% Khatlon Sogdi Sogdi -0,8% Total Total -50% 0% 50% 100% -50% 0% 50% 100% Share Share Prices Expansion Productivity Prices Substitution Productivity 15 5.2.2 Khatlon Household plots were the dominant source of crop growth in Khatlon, contributing far more than in any other region (Figure 34). Given slow privatization and low returns to cotton, farmers were obliged to focus on their own land in order to survive. Figure 34. Crop Growth Contributions by Farm Figure 35. Structure of Crop Growth by Farm Type in Khatlon Oblast (1999-2003) Type in Khatlon Oblast (1999-2003) -0,4% 10% 24% Other 33% Other Farm type Farm type 0,4% 15% 52% 67% HHs HHs 0,02% 25% 75% Total 100% Total -50% 0% 50% 100% 0% 20% 40% 60% 80% 100% Share Share Prices Expansion Productivity Yield increases were the major source of crop growth for both household plots and other farms, with price increases as the next most important factor (Figure 35). Indeed yield increases by household plots accounted for half (52%) of all observed growth for 1999-2003. Further analysis by farm type shows that price increases were the dominant influence in the first period, for both farm categories, but yield increases dominated in the second period when real prices actually fell (Figure 36 and 37). Figure 36. Sources of Crop Growth for Figure 37. Sources of Crop Growth for Other Household Plots in Khatlon Oblast Farms in Khatlon Oblast 1999-2001 1999-2001 Period Period 2001-2003 2001-2003 -1 1999-2003 1999-2003 -50% 0% 50% 100% -50% 0% 50% 100% Share Share Prices Expansion Productivity Prices Expansion Productivity 16 5.2.3 Soghd Household plots and other farm types contributed equally to crop growth in Soghd (Figure 38), a pattern that stayed the same in both time periods. Crop expansion and yield increases each accounted for approximately half of observed growth. Prices actually declined slightly in the first period, but then increased from 2001-2003. Figure 38. Crop Growth Contribution by Farm Figure 39. Structure of Crop Growth by Farm Type in Sogd Oblast (1999-2003) Type in Sogd Oblast (1999-2003) -2% 24% 27% 49% Other Other Farm type Farm type 1% 30% 20% 51% HHs HHs -1% 54% 47% Total Total 100% -50% 0% 50% 100% 150% 0% 20% 40% 60% 80% 100% Share Share Prices Expansion Productivity These trends varied over time, although the pattern was the same for household plots and other farms (Figures 40 and 41). A larger increase occurred in the first period, driven by expansion and yield effects, while prices fell. Slower but more balanced growth occurred in the second period, with increased yields as the main source of growth. Figure 40. Sources of Crop Growth for Figure 41. Sources of Crop Growth for Other Household Plots in Sogd Oblast Farms in Sogd Oblast -3 -8 1999-2001 1999-2001 Period Period 12 2001-2003 2001-2003 -4 1999-2003 1999-2003 2 -50 0 50 100 -50 0 50 100 150 Share (%) Share (%) Prices Expansion Productivity Prices Expansion Productivity 17 5.2.4 RRP Household plots accounted for 57% of the observed crop growth in RRP (Figure 42), due to a marked increase in output growth during 2001-2003. This response is attributed to the growing demand for agricultural commodities in Dushanbe, and the greater freedom of households to respond to this demand. Increased yields accounted for almost half (45%) of observed growth, with crop expansion and price increases explaining the balance (Figure 43). Figure 42. Crop Growth Contributions by Figure 43. Structure of Crop Growth by Farm Farm Type in RRP Oblast (1999-2003) Type in RRP Oblast (1999-2003) 10%12% 21% 43% Other Other Farm type Farm type 18% 15% 24% HHs HHs 57% 28% 27% 45% Total Total 100% 0% 20% 40% 60% 80% 100% 0% 20% 40% 60% 80% 100% Share Share Prices Expansion Productivity These growth patterns varied markedly by farm type and over time (Figure 44 and 45). Crop growth was much slower in the first period, for both farm types, although it was driven by crop expansion among household plots and a combination of yield and expansion effects among other farms. Crop expansion ceased in the second period and growth was driven by a combination of yield and price increases for both farm types. Figure 44. Sources of Crop Growth for Figure 45. Sources of Crop Growth for Other Household Plots in RRP Farms in RRP -2 -0,01 1999-2001 1999-2001 Period Period -3 2001-2003 2001-2003 1999-2003 1999-2003 -50 0 50 100 -50 0 50 100 Share Share Prices Expansion Productivity Prices Expansion Productivity 18 5.2.5 GBAO Other farms made the greatest contribution to crop growth in GBAO (Figure 46). Indeed their contribution increased from 54% in 1999-2001, to 96% in 2001-2003. This is attributed to rapid land privatization, and the greater freedom of farmers to produce and market what crops they choose. Crop expansion explained 64% of this growth, and yield increases the rest (Figure 47). Declining prices offset the impact of these effects. Figure 46. Crop Growth Contribution by Farm Figure 47. Structure of Crop Growth by Farm Type in GBAO Oblast (1999-2003) Type in GBAO Oblast (1999-2003) 4% 40% 23% 67% Other Other Farm type -11% 24% 20% Farm type HHs HHs 33% -7% 64% 43% Total Total 100% -50% 0% 50% 100% 150% 0% 20% 40% 60% 80% 100% Share Share Prices Expansion Productivity Analysis by farm type and time period (Figures 48 and 49) shows that growth was strong from 1999- 2001 and was driven by crop expansion for both farm types. Prices decreased significantly during this period. Second period growth was driven by yield increases for household plots but was negligible. Among other farms, growth was due to yield increases and crop expansion. Figure 48. Sources of Crop Growth for Figure 49. Sources of Crop Growth for Other Household Plots in GBAO Farms in GBAO 1999-2001 1999-2001 Period Period 2001-2003 2001-2003 1999-2003 1999-2003 -50 0 50 100 150 -100 -50 0 50 100 150 Share Share Prices Expansion Productivity Prices Expansion Productivity 19 4.3 Horticulture Figure 50. Sources of Horticulture Growth Horticulture production includes fruit, berries and grapes. Decomposition analysis for these crops focused on changing patterns of response between the two time periods (Figure 50), changes which 1999-2001 reveal important insights into development of this Period sub-sector. As the area planted to these crops 2001-2003 remained relatively stable from 1999-2003, changes in the real value of output were due to changes in yield and prices. 1999-2003 The period from 1999-2001 was characterized by a -200 -100 0 100 200 pronounced increase in yields and a consequent Share increase in production. But as the markets for horticulture commodities are small, this production Prices Substitution Productivity increase led to a 30% fall in real prices, which was especially marked from 2000 to 2001. The real value of output increased significantly nevertheless, by 51% for fruit and berries, and 108% for grapes. During the second period, from 2001-2003, yields and so production fell sharply. This effect was especially marked for grapes, where a 71% fall in yield occurred in addition to a slight reduction in the area cultivated. The yield of fruit and berries fell by 43%. The resultant fall in production led in turn to a 70% in increase in real prices. Total real output value fell by 24% nevertheless, although it was still 30% above the output obtained in 1999. This close relationship between yields and prices is evident in Figures 51 and 52 below. Figure 51. Trends in Real prices and Yields for Figure 52. Trends in Real Prices and Yield for Fruit and Berries Grapes 30 20 35 20.0 18 18.0 25 30 16 16.0 R e a l P ric e (s o m /to n ) 25 R e a l P ric e s o m /to n 14 14.0 20 Y ie ld (c e n t/h a ) Y ie ld (c e n t/h a ) 12 12.0 20 15 10 10.0 15 8 8.0 10 6 10 6.0 4 4.0 5 5 2 2.0 0 0 0 0.0 1999 2000 2001 2002 2003 1999 2000 2001 2002 2003 Yield Real Price Yield Real Price These trends demonstrate two important, related characteristics of the horticulture sector. First, the strong producer response to changing market conditions, by influencing yields. Yields rose steadily during the first period in response to favorable prices, and then fell abruptly when prices fell. Second, the marked changes in producer prices, in response to changes in supply (yield) indicate the inherent volatility of the market for horticultural products. Domestic demand is limited, and there is limited capacity to manage seasonal production flows and/or store output. The market is thus easily saturated, and prices fall substantially when it is saturated. 20 Crop and Livestock Commodity Mix Indicators for Tajikistan Gross Output Value (‘000 CROP PRODUCTION Sown Area (‘000 ha) Gross Output (‘000 tons) Somoni) 1999 2001 2003 1999 2001 2003 1999 2001 2003 Wheat 326.1 326.4 334.8 365.1 387.3 660.2 8,816.7 10,922.8 16,390.5 Corn for grain 11.8 9.0 8.3 35.7 34.3 94.9 531.7 635.7 1,311.1 Rice 17.5 14.2 11.4 47.3 39.8 59.4 1,538.5 1,426.3 2,047.7 Potatoes 20.5 23.2 26.0 239.7 308.2 473.3 4,058.6 3,670.9 10,846.9 Vegetables 34.2 30.5 30.2 386.7 396.8 583.0 2,516.1 3,718.6 5,725.9 Fruit and berries 54.3 57.6 58.5 77.6 143.8 88.9 1,275.7 856.8 1,665.2 Grapes 32.5 33.2 30.0 53.5 109.7 28.4 825.9 584.3 534.7 Melons 11.8 10.5 10.7 84.0 96.9 138.5 369.3 448.8 793.6 Tobacco 4.9 3.1 0.7 6.9 4.1 0.7 816.7 471.9 32.7 Gross Output Value (‘000 LIVESTOCK/PRODUCTS Livestock Inventory (‘000 head) Gross Output (‘000 tons) Somoni) 1999 2001 2003 1999 2001 2003 1999 2001 2003 Cattle/Beef and veal 1,036.7 1.090.7 1,219.0 26.3 25.5 28.4 2,677.0 2,629.7 4,583.5 Sheep/Mutton and lamb 2,178.0 2,269.2 2,592.5 31.6 32.1 57.1 2,244.9 1,845.2 6,653.6 Cows/Milk (liters) 529.5 559.1 636.3 302.3 382.6 459.2 5,881.9 8,858.2 10,397.0 Chicken/Eggs (million units) 770.9 1,319.6 1,887.8 11.1 41.9 53.8 145.6 487.3 679.7 Sheep/Wool 2,178.0 2,269.2 2,592.5 2.1 2.7 3.6 144.9 167.2 246.4 21