irps 38 farm mechanization, employment, and income in nepal: traditional and mechanized farming in...

8/4/2019 IRPS 38 Farm Mechanization, Employment, and Income in Nepal: Traditional and Mechanized Farming in Bara District

1/12


2/12

IRPS No. 38, August 1979

FARM MECHANIZATION, EMPLOYMENT, AND INCOME IN NEPAL:TRADITIONAL AND MECHANIZED FARMING IN BARA DISTRICTl

ABSTRACT

A survey of traditional and mechanized farms in Baradistrict, Nepal, was conducted to assess the impactof mechanization on cropping intensity, timeliness,yields, income, employment, and efficiency. Croppingintensity, yields, income, and employment were higheron mechanized than on traditional farms. However,the much greater use of cash inputs and higher edu-cation levels associated with mechanized farms madeit difficult to attribute yields and income effectssolely to machinery. Tractors could not be clearlylinked with anyon-farm labor displacement and pump-

sets were found to raise farm employment. Tractorownership allowed large farms to achieve highercropping intensity through speedy and timely opera-tions. Increased cropping intensity appeared to puta premium on timeliness for large farms, but did notseem important for small units. Tractorization per-mitted the farmers to nearly eliminate bullocks.The highest levels of efficiency were achieved bypumpset owning and tractor hiring farms rather thanlarge tractor owning farms.

1 ..By Som Prasad Pudasalnl,Ric e Res earch In stit ute.Philippines at Los Banos,

former research scholar, Department of Agricultural Economics, The InternationalThis paper is based on the author's MS thesis submitted to the University of the1976. Submitted to the IRRI Research Paper Series Committee August 1979.


3/12


4/12


FARM MECHANIZATION, EMPLOYMENT, AND INCOME IN NEPAL:TRADITIONAL AND MECHANIZED FARMING IN BARA DISTRICT

Farm mechanization2 in the developed countries hasoccurred primarily as a response to high and risingwage rates. The recent spread of tractorization tocountries with low wage rates such as India, Pakis-tan, and Nepal raises two sets of questions. First,because tractors must have been privately profitableat the prices and incentive structures under whichthey were bought, what is the source of benefitstha t account for the private profitability where wagerates are not high? Are yields raised, does croppingintensity rise, or is the primary benefit the re-placement of animals? Second, does the tractor leadto direct labor displacement or to a slow-down in thegrowth of labor demand available from other innova-tiolls?This paper attempts to answer those questions in thecontext of the Nepal Terai. A survey of differenttypes of f a rms was conducted and the "before andafter" as well as "with and without" approaches wereused to evaluate these issues but the results wereinfluenced by differences between bullock-operatedand tractor and fanns in respects other than trac-tors. In particular, tractor farms are usuallylarger, better capitalized, and also may have morequalified managers. Special attention has thus beenplaced on measuring these other differences. Finallyproduction function analysis is used to "control"these other differences, at least in a statisticalsense.

RESEARCH DESIGNIn 1975 there were slightly more than 2,000 tractorsoperating in Nepal. Fourteen percent (295) wereoperating in Bara District of the Terai. About 12%of the farmers falling in the largest farm size cate-gory owned tractors (Table 1).

S mn p le c ha r ac t er i st i csFive different types of farmers, traditional withbullocks (TDF), pumpset owners (POF), tractor owners(TOF), tractor hirers (THF), and tractor and pumpsetowners (TPOF) were identified and studied to tracedifferences in fann employment and income. A samplesize of 20 or 21 was randomly drawn from each of thefive specified types of farmers and 102 farmers fromBara District in the central Terai of Nepal were

?-Although mechanization in the study refers tointroduction of four-wheeled tractors or pumpsetsor both, the major factor expressed in this paperis tractorization.

interviewed to gather necessary infonnation on pro-duction, costs, labor use, etc., for the periodNovember 1974 to December 1975.The five specified types of fanners differed fromeach other in basic characteristics, such as cul-tivated area, expenses for improved inputs, invest-ment in machinery and tools, and education of thefann operators (Table 2). Most traditional farmerswere part owners of land in contrast to most machineusers being full owners. The differences in expensesfor improved inputs, value of machines and tools,and education between the traditional farmers andall other categories were large, indicating muchpoorer access to capital inputs of any type on thesef a rms

Impact of mechanization on farm operations andc ro pp in g i nt en si tyAssessment of the quality of land preparation per-formed by the tractor was based on the farmer'sjudgment of degree of weed infestation immediatelyafter transplanting, crop stand, and yield. Weedinfestation was less on fanns plowed by tractorsas reported by most tractor users. Most tractorusers also reported better rice and wheat standsand higher rice and wheat yields after use of thetractor for land preparation. Crop yields showedthat both rice and wheat yields were higher onmechanized than on traditional farms (Table 3).But the simultaneous arrival of biological (modernvarieties) and chemical (fertilizers, pesticides,e tc ,) innovations on mechanized fanns makes it dif-ficult to attribute higher yield and better cropstand solely to tractorization. Pumpset owners hadlower yields for rice but marginally higher yieldsfor wheat than the tractor as well as tractor andpumpset owners despite the fact that tractor ownersused almost the same level of improved inputs andtractor and pumpset owners used substantially higherlevels than pumpset owners. Thus, there was no clearevidence in support of the hypothesis that tractorsimprove yields.

Optimization of planting dates depends primarily ontimely land preparation. Delay in land preparationmeans delay in planting, which in turn delays har-vesting. The influence of tractorization on time-liness of fann opera tions was, therefore, based onfanners' reports of how their last rice and wheatplanting dates compared with those of their neigh-bors before and after tractor use. More than 90percent of the tractor owners and tractor and pump-set owners reported earlier completion of wheatplanting and more than 62 percent of them reportedearlier completion of rice transplanting than theirneighbors.


5/12

4 IRPS No. 38, August 1979

Table 1. Distribution of landholdings in Bara District, Nepal.

F arm Gro upAverage size ofl an dh ol di ng s ( ha ) Total cultivated area Total landholdings

Thousand %ha

2.481 8.7

5.473 19.2

20.598 72.1

28.552 100.0

Thousandha

%

Large 9.02

Medium 3.56

Small 1.24

Total 2.36

Source: Agricultural Credit Survey, Nepal, 1972.

22.376 33.2

19.482 28.9

25.541 31.967.400 100.0

Table 2. Selected characteristics of traditional and mechanized farmers of Bara District, Nepal, 1975.

Traditional Pumpset Tractor Tractor T ract or an dCharacteristics farmers owners owners hirers pumpsct owners(N = 20 ) (N = 20) (N = 21) (N = 2f ) (N = 21)

Education (years) 3. 8 5.6 7. 0 6.7 7.9Landholdings (ha) 5. 9 8.8 32.1 9. 4 25.0A re a c ul ti va te d (ha) 5. 3 8.3 21.1 6.2 21. 6Expenses for improifed inputs 64 682 653 388 857(Rs/ha per year)Value of machines and tools (Rs/farm) 254 5,521 47,805 93 8 58,157

a Rs 10.00 US$1. 00.

In contrast, almost no change was observed in therice transplanting schedule of those hiring tractors.In the case of wheat, however, a relatively largerproportion (35%) of the tractor hirers reportedplanting earlier than their neighbors after usingtractors; only 5 percent completed planting earlierbefore the introduction of the tractor.Cropping intensity on all machine-using farms washigher than on traditional farms after mechanization.Differences in cropping intensity were prevalentamong different types of machine using farms (Table4). Their increase in cropping intensity indicatedtwo important points. First, that mechanizationpossesses the potential for increasing croppingintensity. Second, a mechanization strategy com-bining irrigation and efficient land preparationfacilities is likely to result in much higher crop-ping intensity than for either technology alone.

The similar cropping intensity of tractor owners andtractor hirers indicated that the tractor rental mar-ket could be as effective as tractor ownership inac hiev ing hi ghe r c ropp ing in tens ity .

F a Y ' I T Ie m p l o y m e n tFarm employment was measured in days per hectareper year of bullock and human labor used in rice,wheat, and sugarcane. The annual per hectare humanlabor absorbed by machine users was much higher thanthat absorbed by traditional farmers (Table 5). Themaximum labor use of 190.4 days was on pumpset owningfa rms. A ddi tion al or al ter nati ve mec han izat ionresulted in a decline in labor use. Pumpsets thusappeared to be more of an employment-generating formof mechanization than tractor farms, which employedmore labor per hectare than traditional farms. The


6/12

Table 3. Crop yield of sample traditional andmechanized farmers, Bara District, Nepal, 1975.

Type of farmer Yield (t; /ha) aRice Wheat

Traditional 1.72 1.98(.41) (.63)

Pumpset owners 2.09 2.57(.45) (.67)

Tractor owne rs 2.20 2.42(.34 ) ( .52)

Tractor hirers 2.09 2.09(.41) (.67)

Tractor and pumpset owners 2.28 2.50(.37) ( .60)

aStandard deviation in parentheses.

Table 4. Cropping intensity a of sample farmersbefore and after mechanization, Bara District,Nepal, 1975.

Type of farms Cropping intensity(%)bBefore After Increase

Traditional 145.0 145.0

(33.5)Pumpset owners 137.4 155.3 17.9

(38.3) (43.8)Tractor owners 121. 2 165.0 43.8

(42.2) (19.2)Tractor hirers 148.8 166.9 18.1

(22.6) (14.1)Tractor andpumpset owners 155.2 174.1 58.9

(29.3) (36.4)

a Cropping intensity:Total area under different cropsduring a given year x 100Total cultivated area available

b Standard deviations in parentheses.


higher absorption of labor by machine users over thetraditional farms resulted from the increased use oflabor, mainly in weeding, harvesting, and threshing(Table 6). That may have been partly due to in-creased use of modern technology which demandedbetter weeding to attain a better harvest. The in-creased cropping intensity of machine users was alsoan important contributor to higher labor absorptionin case of machine users.Bullock labor employed by tractor owners, tractorand pumpset owners, and tractor hirers was alsolower than that of traditional farmers (Table 7).Pumpset owners were not significantly different fromtraditional farmers in bullock use. This shows thatthe displacement of human labor in land preparationand the reduction in bullock needs on the farm waslarger in total in the case of tractor owners, close-ly followed by tractor and pumpset owners; both dis-placements were nominal in the case of tractor hirersand pumpset owners were relatively neutral.The large reduction in bullock use associated withtractor ownership was one of the primary benefitsof tractorization while wage cost reduction waslimi ted.

F a Y ' r r t i n c o m e

The level of farm income is a measure of the profit-ability of a farm business and reflects the effi-ciency with which farm resources or services areutilized. Farm income can be measured as gross farmrevenue (GFR), farm family income (FFI), returns tofamily labor and capital (RFLC), returns to familylabor and management (RFLM), and returns to familyowned capital (RFOC). Of these, FFI, which is returnabove paid out costs, and represents returns to ownedland, family labor and management, and farm capitalsuch as buildings, machines, tools and equipment, isperhaps most widely used. Machine users had higherincome than traditional farmers as shown by thevalues of GFR, FFI, RFLC, RFLM, and RFOC in Table 8.Among machine users, almost all measures showed thehighest income in the case of tractor and pumpsetowners, followed by tractor owners, tractor hirers,and pumpset owners.

FARM EFFICIENCY AND FACTOR PRODUCTIVITYUnderlying the estimation of marginal productivityand efficiency of the farms under study is the pro-duction function concept that the quantity of outputdepends on the quantities of factor inputs or ser-vices employed. Assuming that the output (GFR) Ydepends on the quantities of factor inputs, Xi,education of a farm operator, E, and types of farms,Dj, the production function can be expressed as

Y (1)

Estimates of a y / a x and a y / a E provide the marginalvalue of products for factor inputs and education of


7/12


Table 5. Annual human labor employed by pamp1e traditional and mechanized farmers, Bara District, Nepal, 1975

a

Hired labor Increase overHired as percent- traditionalage of total farms

104.4 70178.8 94 41.4173.0 98 27.4170.0 91 38.0178.3 93 33.0

Type of farms Total

Traditional 149.0Pumpset owners 190.4Tractor owners 176.4Tractor hirers 187.0

Annual labor use (days/ha)Family Exchange

Tractor and pumpset owners 182.0

37.011. 63. 1

16.83. 6

7. 0

a0. 20. 1

a Less than .01.

Table 6. Human labor for farm operation on sample traditional and mechanized farmers, Bara District, Nepal,1975.

Annual labor used (days/ r)Traditional Pumpset Tractor Tractor Tractor andfarmers owners owners hirers pumpset owners

33.0 31.9 4.7 29.3 7.38.0 9.6 9.8 9.0 13.6

49.7 50. i, 49.3 53.8 52.01. 3 1. 9 2.8 3.3 3.3

22.0 40.4 50.8 39.7 40.335.0 54.2 59.0 51.9 65.5

149.0 190.4 176.4 187.0 182.0

Farm operation

Land preparationDike repair and irrigationTransplanting or plantingFertilizing and sprayingWeedingHarvesting and threshing

Total

Tahle 7. Bullock l abo r absorption by sample tradi-tional and mechanized farmers, Bara District, Nepal,1975.

a, a2 c x3 c x4 Sy X .l X2 X3 X4 e (2 )1where

5S ylX5 + y2X6 + Y3X7 + l: d -D .j 2 J JType of farm

Annual bullock labor (~s/haLTotal Family Exchange Hired

Traditional 63.3 49.3 14.0 aPump set owners 57.4 46.4 9.9 1. 1Tractor owners 3. 4 2.9 0.5Tractor hirers 53.5 40.8 12.4 0. 2Tractor & pumpset 4.8 4.5 0. 3

owners

a Less than 0.1.

andXl total crop area per farm (ha) ,X 2 total human labor used per farm (days),

cash expenses on fertilizer, pesticides,irrigation, and land revenue per farm(Rs)",

X4 total bullock labor used per farm (days),X5 schooling of farm operator (yr),

a farm operator. For my purpose, the followingspecification in natural logarithms was estimated toinvestigate whether selected conventional and non-conventional inputs significantly raised farm revenue.

total pumpset use per farm (h), andtotal tractor use per farm (h),

total gross farm receipts (Rs)".


8/12


Table 8. Farm income of sample traditional and mechanized farmers, Bara District, Nepal, 1975.

Farm income (Rs per ha bper year)Measures of Traditional Pumpsetincome a Tractor Tractor Tractor andfarms owners owners owners pumpset owners

GFR 2167 3533 3687 3110 4527FFI 1484 2031 2222 2277 2635RFLC 790 1001 1481 1062 1666RFLM 594 648 1115 705 1099RFOC 535 868 1447 923 1621

aCFR (gross farm revenue): Sum of the values of all farm crop output, amount received from the sales of live-stock; value of products received by leasing or renting land (if any), and income earned from renting out oftractor and/or pumpsets by a farm family. FFI (farm family income): GFR minus paid out costs representedtotal cash and kind costs actually incurred during the farm revenue generating process. Cash component of paidout costs included expenses on fertilizer, insecticides, land revenue, irrigation fee, fuel and lubricants,repair and maintenance of farm machinery, cash labor cost. The kind component included produce paid to land-lord, seed, kind payment to labor. RFLC (return to family labor and capital): FFI minus imputed rental costsof owned land. The imputed rental cost on the owned land was assumed to be 10 percent of the value of the landRFLM (return to family labor and management): RFLC minus imputed interest cost on family owned capital. Theinterest cost charged on family owned capital was 15 percent. RFOC (return to family owned capital):b RFLCminus imputed costs on family labor. In imputing cost on family labor on-going wage rate was used. Rs 10.00US$l.OO.

The dummy variables are where

D2 1 if pumpset owners, 0 otherwise;y gross farm receipts at its geometric mean

D3 1 if tractor owners, 0 otherwise; x . .th factor input at its geometric meant. 1D4 1 if tractor hirers, 0 otherwise; and u. partial regression coefficient of .thDS 1 if and 0 other- i. input. 1tractor pumpset owners,wise.

The marginal value product (MVP) of a given factorinput is the first derivative3 of the estimatedspecification with respect to that factor input.Thus, the MVP of Xi, with all other inputs at theirgeometric means, was computed as:

7 J _ _ u. .u . i = 1, 2, 3, 4oX . t. X .

,i. t.

and

7 J _ _ u. y , i 5, 6, 7sr . t.t.3Because h d ( )e ependent variable y is in monetaryvalue here, the first derivative directly givesMVP, otherwise it would be only the marginal physicalproduct for the input in question. *Rs 10.00 =US$l.OO as of June 1979.

In addition, farm type dummies were employed in thefunction to see whether the specified types of machineusers were significantly different from traditionalfarmers in terms of efficiency. The results,reported in Table 9, showed that all conventionalresources, viz. land, human labor and bullock labor,contributed significantly to increase farm revenueof traditional farmers as well as machine users.Land had the largest coefficient, followed by humanlabor, indicating that land and human labor were themost important productive resources in the district,contributing a larger share to farm revenue.As expected, the nonconventional inputs, viz. cashexpenses in chemical fertilizers, pesticides, irri-gation, etc., and education of farm operators alsohad a positive and significant impact on farmrevenue.

The contribution of both tractors and pumpsets tofarm revenue was positive but insignificant inEquation 2, whereas the contribution of tractor


9/12


Table 9. Partial regression coefficients obtainedby fitting the specified function to pooled dataof sample traditional and mechanized farmers,Bara District, Nepal, 1975.

Independentvariables

Partial regressioncoefficients of

equations2

0.3710** 0.3823***(0.0956) (0.0942)0.2837*** 0.2708***(0.0965) (0.0951)0.0827** 0.0815**(0.0397) (0.0391)0.0556** 0.0554**(0.0299) (0.0294)0.0137** 0.0126*(0.0080) (0.0078)0.0003* 0.0002(0.0002) (0.0002)0.0004 0.0003(0.0004) (0.0004)

Crop land: Xl

Human labor: X2

Cash: X3

Bullock 'labor: X4

Education: X5

Tractor labor: X6

Pumpset use: X7

Dwnmy variablesPumpset owners: D2 0.1988 0.2540*

(0.1562) (0.1538)Tractor owners: D3 0.0525 0.2308

(0.2210) (0.2177)Tractor hirers: D4 0.4032*** 0.4195***

(0.1312) (0.1292)Teactor and pumpset

owners:D5 -0.2078 0.0048(0.2443) (0.2406)

Constant0.90296.0607

0.90976.1286

*Significant at 10 percent level, ** significant at 5percent level, *** significant at 1 percent level.Figuees within parentheses indicate standard errorsof the coefficients. NOTE: Regressions 1 and 2differ in that the dependent variable (Y) of 2 in-cludes earnings from machine renting out whereasthe other includes earnings only from within a farm.

labors to farm revenue was posltlve and significantin Equation 1. Because the contribution of a machinecan be realized over a period of time, the estimatebased on cross-sectional data for one year might nothave caught the effect of machines on revenue fully.The ambiguity of the results may, therefore, indi-

cate a deficiency in the data, and not permit draw-ing of any strong conclusions about the contributionof machines to farm revenue.The positively significant coefficients of farm typedummies for tractor hirers and pumpset owners (Equa-tion 2) showed that both were more efficient thantraditional farmers. On the other hand, the insig-nificant coefficients of farm type dummies for trac-tor owners and tractor and pumpset owners, regard-less of their size, indicate that those farmerswere not significantly different from traditionalfarmers with regard to their efficiency. Overall,the small tractor use coefficients, together withthe nonsignifica'nt dummy variables for tractorownership on the ,one,hand, indicate that differencesin farm earnings are related more closely to the useof other labor and capital inputs and education thanto tractor ownership per se. The large coefficientassociated with land (crop area), however, indicatesthe possibility that tractors indirectly contributeto increased farm earnings through increased crop-ping intensity and probably through better landpreparation.Marginal analysis showed that the marginal va1:ueproducts of land, human labor, cash expenses, andbullock labor were more than their unit prices. Inaddition, the marginal value products of educationof a farm operator was also substantial (Table 10).Although the marginal value of products for thepumpset was more than the custom rate and for thetractor less than its custom rate, the reliabilityof the estimate is dubious because the coefficientswere insignificant.

DIFFERENCES AND BENEFITSThe production function analysis shows that most ofthe observed differences in gross farm revenue be-tween mechanized and traditional farms are primarilya reflection of the much higher use of cash inputsand higher education levels. This, however, doesnot rule out the possibility of tractors indirectlycontributing to higher farm earnings. There isevidence that the highest efficiency levels areachieved by small pumpset owning and tractor hiringfarms rather than the large tractor owning farms.

Nevertheless, the source of private benefits fromtractor ownership was clear. In the case of BaraDistrict, tractor ownership allows large farms toachieve and surpass intensity levels otherwise asso-ciated with smaller bullock farms. It has alsoallowed them to almost fully eliminate bullocks.Conversely, there is no clear evidence of yieldincreases caused by tractors alone, despite the factthat tractor farmers report fewer weed problems anda higher degree of timeliness. This implied thattimeliness was not a very important premium for thesecond crop on small farms in the area which wasirrigated, or that nontractor farmers may properlytake care of added weeds by later weed controlmeasures or both. However, increased cropping in-tensity appeared to put a premium on timeliness forlarge farms in the area.


10/12

Table 10 . Marginal value products of factorinputs computed from the es tima ted specifica-tion.

Marginal va]ue products(Rs)a

Computed from Rentaleguations rates

Inputs Unit 1 2

Crop lanGb ha 806.85 892.00 894.00Human labor days 11.12 10.53 6.00Cash ex-

penses Rs 1.10 1.08 1. 00Bullock

labor days 17.46 17.40 13.00Tractor

labor hours 11. 53 7.55 40.00Pumpset

use hours 15.10 11. 33 9.20Education years 517.18 475.68

aRs 10.00 = US$l.OO. bThe marginal value productspresented are for crop area whereas the rental ratesare for the physical area. Adjusted for croppingintensity, rental rate per hectare of crop area turnsout to be Rs 555.00.

IRPS No. 38, August 1979 9

On-farm labor displacement did not seem to be largeand was not a primary source of private benefits.This does not imply that tractors do not displacelabor from a br08der societal point of view. Thelarge amount of capital invested in tractors did notcreate as much employment as the pumpsets on all typesof farms. But it must be noted that tractors may in-crease employment on large farms allowing them toraise cropping intensity while the same may not betrue on small farms, wh ich constitute d large bulkof total holdings in Nepal. The small farms appearto be able to attain high intensity even withouttractors.If the capital invested in tractors had been investedin additional irrigation facilities, however, laborrequirements would have increased. It is also like-ly that most nonagricultural investments would alsohave led to more employment creation.Considering the foregone opportunities for employmentcreation, as well as potential on-farm displacement,investment in irrif8tion facilities such as pumpsetmechanization rates priority in the context of low-wage (labor abundant), capital-scarce economies suchas Nepal.

REFERENCES CITEDAgricultural Credit Survey, Nepal. 1972. The survey

report, Vol. 1. Nepal Rastra Bank.Pudasaini, Som P. 1976. Resource productivity,

income, and employment in traditional andmechanized farming of Bara District, Nepal.Unpublished MS thesis, University of thePhilippines at Los Banos.


11/12


12/12

Other papers i n this seriesNo. 1No. 2

No. 3

No. 4

No. 5

No. bNo 7

No 8No. 9

No. 10No. 11

No. 12

No. 13No. 14No.15No 16No. 17

No. 18

No. 19

Recent studies on the rice tungro disease at IRRISpeci fic soil che mical ch.rructcrixt ics for rice production inAsinBiological nitrogen f ix a tio n in paddy field studies by in situace ty lcnc-rcd uctiori assay sTransmission of rice tungro virus at various temperatures: At r.msi tory virus-vector interactionPhysicochemical properties of submerged soils in relation tofertilityScreening rice for tolerance to mineral stressesMulti-site tests environments and breeding strategies for newrice technologyBehavior of minor elements in paddy soilsZinc deficiency in rice: A review of research at theInternat ional Rice Research InstituteGenetic and sociologic aspects of rice breeding in IndiaUtilization of the Azolla-Anabae na complex as a nitrogenfertil izer for riceScientific communication among rice breeders in 10 Asianna tionsRice breeders in Asia: A 10-country survey of theirbackgrounds, attitudes, and use of genetic materialsDrought and rice improvement in perspectiveRisk and uncertainty as factors in crop improvement researchRice ragged s tunt disease in the Phil ippinesResidues of carbofuran applied as a systemic insecticide inirrigated wetland rice: Implications for insect controlDiffusion and adoption of genetic materials among ricebreeding programs in AsiaMethods of screening rices for varietal resistance toCcrcospora leaf spo t

No. 20No. 21No. 22

No. 23No. 24No. 25No. 26

No.27

No. 28

No. 29

No. 30

No.31

No 32

No. 33

No. 34No. 35

No. 36No. 37

Tropical climate and its influence on riceSulfur nutrition of wetland riceLand preparation and crop establishment for rainfed andlowland r iccGenetic interrelationships of improved rice varieties in AsiaBarrier s to efficient capi tal investment in Asian agricultureBarriers to increased rice production in eastern IndiaRainfed lowland rice as a research priority - an economist'sviewRice leaf folder: Mass rearing and a proposal for screeningLH varietal resistance in the greenhouseMeasuring the economic benefits of new technologies tosmall r ice farmersAn analysis of the labor-intensive cont inuous rice productionsystem at IRRIBiological constraints to farmers' rice yields in threePhilippine provincesChanges in rice harvesting systems in Central Luzon andLagunaVariation in varietal reaction to rice tungro disease: PossiblecausesDetermining superior cropping patterns for small farms in adry hind environment: Test of a methodologyEvapotranspirat ion from rice f ieldsGenetic analysis of traits related to grain characteristics andquality in two crosses of riceAliwalas to rice garden: A case study of the intensificationof rice farming in Camarines Sur , PhilippinesDenitrification loss of fertilizer nitrogen in paddy soils - itrecognition and impact

The International Rice Research InstitutePO. Box 933. Manila, PhilippinesStamp

ISSN 0115-3862

irps 38 farm mechanization, employment, and income in nepal: traditional and mechanized farming in...

Documents