DISCUSSION PAPER Report No.: ARU 58 Population Density, Market Access and Farmer-Generated Technical Change in Sub-Saharan Africa By Prabhu L. Pingali and Hans P. Binswanger Research Unit Agriculture and Rural Development Department Operational Policy Staff World Bank September 1986 The views presented here are those of the author(s), and they should not be interpreted as reflecting those of the World Bank. The authors are staff members of the World Bank. However, the World Bank does not accept responsibility for the views expressed herein which are those of the authors and should not be attributed to the World Bank or to its affiliated organizations. The findings, interpretations, and conclusions are the results of research supported in part by the Bank; they do not necessarily represent official policy of the Bank. The designations employed and the presentation of material in this document are solely for the convenience of the reader and do not imply the expression of any opinion whatsoever on the part of the World Bank or its affiliates concerning the legal status of any country, territory, area or of its authorities, or concerning the delimitation of its boundaries, or national affiliation. POPULATION DENSITY, MARKET ACCESS AND FARMER-GENERATED TECHNICAL CHANGE IN SUB-SAHARAN AFRICA Prabhu L. Pingali and Hans P. Binswanger Introduction The history of agricultural growth of the developed world illustrates that the rate and direction of technical change are influenced by the economy's land and labor endowments and by the conditions of demand for final agricultural products. The responsiveness of science-based invention and innovation to economy-wide factors has come to be known as the process of induced innovation. Historical and econometric inquiries of U.S. and Japanese agriculture by Hayami and Ruttan (1970) and Binswanger (1974) have verified this endogenous nature of science based technical change. Farmer-generated technical change is simiiarly influenced by an economy's population, land endownments and market infrastructure. Population induced changes in traditional food supply systems were examined by Boserup (1965). She concluded that traditional societies across the world have historically devised remarkably similar means of coping with reductions in per capita land availability. Moreover, output growth resulting from these production responses has often been sufficient to cope with the relatively low historic rates of growth in demand from increasing populations. Boserup's view on the beneficial effects of population growth on total agricultural output has been shared by many economists: Kuznets -2- (1960), Clark (1968), Phelps (1966), Simon (1977) and Lee (1984) among others. Within a neoclassical framework a few attempts have also been made to formalize the concept of population induced technical change in pre- industrial societies (Stryker, 1976; Dari-cy, 1980; Pryor and Maurer, 1982; Robinson and Schutjer, 1984). We, however, are not aware of any attempts to empirically test the hypothesis of population-induced technical change proposed by Boserup. In Boserup's analysis population growth has eight principal effects on farming systems, many of which have also been described systematically by Ruthenberg. These effects are : (1) it increases the intensity of Land use, i.e., causes the Ynovement from shifting cultivation to permanent cultivation of land; (2) it increases investments in land improvements especially by drainage, irrigation and terracing; (3) it encourages the shift from handhoe cultivation to animal traction; (4) it encourages soil fertility maintenance via manuring; (5) it reduces the average cost of infrastructure; (6) it permits more specialization in production activities; (7) it induces a change from general to specific land rights; and (8) it reduces the per capita availability of common property resources (forest, bush and/or grass fallows, and communal pastures) (Boserup, 1965, 1975, 1981). The discussion in this paper will concentrate on effects (1), (2), (3), (4) and (7). The first four of these effects result from the necessity to raise land productivity and to offset the increase in labor requirements associated with more intensive cultivation. The final effect on the transition to specific land rights provides incentives to undertake investments in specific plots, investments which are required for the -3- intensification of production and the preservation of soil fertility. This paper first provides a highly stylized utility maximizing model of the following decision variables: the intensity of land use; land investments; soil fertility maintenance; and power source. We show analytically that these decisions can be interpreted as responses to exogenous changes in population density and improvements in market infrastructure. We also show that these decisions vary spatially due to differences in agro-climatic potential. The major hypothesis consistent with the model are summarized and then tested using two distinct data sets. The first set consists of data from 56 villages in 10 countries across Sub-Saharan Africa and India. This data was collected during field visits in 1983-84 using the group interview technique as part of a larger research effort on agricultural mechanization in Sub-Saharan Africa. The second data set consists of worldwide data on labor use and labor productivity assembled from several research studies and surveys. -4- Model Consider a traditional land abundant agricultural society with the following characteristics: (i) Utility functions in output and leisure of all persons in the society are identical; (ii) All farm households have equal access to cultivation land; (iii) There is no locally resident non-cultivating labor class; 1/ (iv) There is no hiring or exchange of labor among residents during the peak labor season, which in this case is the sowing and weeding season; 2/ (v) Rental markets in land do not exist; and (vi) There are no rental markets for animal draft equipment and for tractors. The total amount of land available to the farmer is A=A/N where A is the total arable land of the community and N denotes the number of people in th- community. Denote R to be the land cultivated by the farmer in any given year, and define intensity of cultivation as follows (1) r = R/A Values of r less than one are observed in farming systems that periodically leave a proportion of their land fallow. For example, 1/ In land abundant environments the ease of access to land and simple technology imply that a worker's output is at least as large on his own plot as it is on his employer's plot. Therefore, given supervision costs, the employer cannot compensate a worker for the latter's foregone output on his own plot (Binswanger and McIntire, 1986). 2/ The absence of hired or exchange labor during peak labor season is because the sowing and first weeding operation are highly time-bound and time-synchronic in most environments. Postponing one's own sowing to work for someone else implies a reduced expected yield because it reduces the expected length of the already short growing season. It also implies higher risks. The optimal timing of the first weeding follows from the timing of sowing; weeding delays reduce yield and increase weeding effort (Binswanger and McIntire, 1986). -5- r = .1 implies that each plot of land is cultivated for 1 year and then reverted to fallow for 9 years, while r = 1 implies that each plot of land is cultivated once a year. Values of r greater than one imply multi- cropping systems, for instance, r = 2 implies that each plot of land is cult±vated twice a year. Assume that each farmer has a choice between two cultivation technologies: (a) handhoe technology; or (b) animal draft technology. The salient features of each of these technologies are discussed below. In this section of the paper we will attempt to model the farmer's choice of these technologies. We will show that this choice is determined by the land available to the farmer and on the incentives a farmer has for increasing output (for-going leisure). In other words, we will show that the transition from a handhoe based to an animal drawn plow based technology is directly related to population density and/or improvements in market infrastructure. Hand cultivation regime Consider a utility maximizing farmer who wishes to determine the optimal levels of output and labor input within a hand cultivation regime. Define total output per farm as follows: (2) QH Rq (r,m) Arq (r,m) where qH(r,m) denotes output per hectare, which is a function of the intensity of land use (r) and the amount of organic fertilizer used per hec'are (m). Subscript H refers to the use of the handhoe technology. Organic fertilizer has to be either in the form of ash, compost or manure. Where animals are not available, manure cannot be used. The production function qH(r,m) is defined as follows -6 (3) q (r,m) m (-)ma2 , 0 < a,, a2 < 1 and ai+ a2 < 1. r Where I is an exogenous index of the inherent fertility of the soil and the variable I/r is the actual soil fertility which is positively related to the inherent soil fertility and inversely related to the intensity of land use. Soil fertility declines due to increases in the intensity of land use can be restored by using organic fertilizers. Output is a Cobb-Douglas function of actual soil ferility, organic fertilizer used, and the intensity of land use. In a handhoe regime the two major tasks for which labor is required are (a) organic fertilizer production; and (b) crop cultivation. Tie labor requirements for organic fertilizer production are assumed to be a direct proportion of the quantity of fertilizer used per hectare (Om). Where e is defined as a fraction of a man year spent producing m units of organic fertilizer. The per hectare labor requirement for crop S cultivation is given by Xr , where X a positive constant can be interpreted as a fraction of a man year spent on crop cultivation and 0 < 6 < 1. In other words, per hectare labor requirements for crop cultivation rise with the intensity of land use. This increase in per hectare labor input occurs because: (a) the intensity with which certain tasks have to be performed increases (e.g., as fallow periods become shorter, the land under fallow becomes grassy and higher labor input is required during land preparation in order to get rid of grass roots); and (b) the number of operations which have to be performed increases (e.g., weeding). In order to get a closed form solution for the indirect utility function a simple utility function is chosen which is concave in output and a negative linear function of labor input LH -7- (4) U = U(QN)L ) = [R(r) R2] - - R = AI ILr(1 G1)aMa2- emrA - Xr &1A where S is the elasticity of utility with respect to output, O<8K1, a, and a2 are elasticities of output with respect to soil fertility and organic fertilizer respectively, 0 < a,, a2 < 1 and a, + a2 < 1. The first order conditions for utility maximization are (5) U = a= = (i-al) IA&I lSr(1ad) $lma2 = eA + ( +1) Xr A au a, I1 (1- a,) I S - U a bU = = =Irm erA m from which we get the optimal levels of r and m, denoted by r* and respectively. 1 1-5 (6) rH= GA 8 =A X2 H a aI aj8 X 6 +1) C~ la21 5 11 (1-S)6 (7) H = (A 2 i) i ) 1 =A where A= t(1-a,) I-1 + 6 (I28-1)} < 0 and XH and YH are implicitly defined in equations 6 and 7. Comparative statics on r and * provide the following results: dr* a) < 0 A decrease in land area per farmer results in an increase in agricultural intensity. dr* b) -a. > 0 Areas of higher inherent soil fertility are more intensively cultivated. -8- dr* c) * > 0 An increase in the marginal utility of output - (resulting perhaps from an improvement in market access) leads to an increase in the intensity of land use. dmA d) dA < 0 A decrease in land area per farmer results in an increase in organic fertilizer use per hectare. d < e) dI > 0 Areas of higher inherent soil fertility use higher levels of organic fertilizer per hectar'e because these areas have shorter fallow periods. d m f) M« > 0 An increase in the marginal utility of output leads to an increase in organic fertilizer use per hectare. The indirect utility function, U* can be obtained by introducing the H expression for r* and * into equation (4). H Mf int qain() (8) U* = A { a1, (1- a)$ ya2$ - a(y - E () H H H RHR H dUH* d2U* H> 'H < °. U* is a monotonically rising function of A. dA As land becomes scarce due to rising population densities, individuals' utility levels decline because the marginal disutility associated with additional effort exceeds the marginal utility associated with additional output produced. It is at this stage that farmers will have an incentive to switch to technologies that increase labor productivity and thereby arrest further declines in utility levels. Animal traction regime Let us suppose that a farter can choose to switch from a handhoe to an animal traction regime at any point in time. Assume that there are no problems in the acquisition of oxen and a plow and that the 2iarning costs associated with animal maintenance and use are minimal.3/ A farmer contemplating a switch from a handhoe to an animal draft regime faces two additional tasks that require his time and effort, these are: (a) labor costs associated with year-round animal care and maintenance; (b)labor requirements for the complete destumping of cultivated land. The cost of destumping (in terms of labor used) is extremely high under forest and early bush fallow systems due to the high density of stumps per unit area and due to the highly developed root network that is difficult to remove. As the length of fallow decreases the cost of destumping declines becoming minimal by the grass fallow stage (fallow periods of 2-3 years). Let us specify the additional labor costs as follows, the costs for animal care: IC and the cost of destumping land: - R = A. When a farmer in a handhoe regime switches to animal draft power he experiences substantial labor savings in organic fertilizer productior since manure production is possible, he also saves labor in land preparation. We account for these labor savings in the following way: (i) the per hectare labor requirements for organic fertilizer production under an animal traction regime is given by rt, where n is less than e the corresponding coefficient in the hand hoe regime and (ii) the per hectare labor requirements for crop cultivation are given by r , where y is less 3/ We discuss these problems in detail in Pingali, Bigot and Binswanger (1986). - iO,- than X the corresponding coefficient in the handhoe regime. The utility function of a farmer in an animal traction regime is given by equation (9) (9) UT = AaIa,r a,)$ma2 6 A-A- 1C The optimal levels of r and m and the corresponding indirect utility functions are as follows: 1 rA I- A n Ia2 (10) r* ~ ' )= XT a~ (1-S) 6 (T a( 1-y6) 6 BF Y where A = (I-aj) B-1 + S ( a2fi1)} < 0, and and AT are implicitly defined in equations 10 and 11. d a2- I)- a, (12) U* = A I tI x(I-a,) ay a28 - YT A- } TA'C. 2 dUT* d U* T 0 ° T dA 2A Unlike U* which is a monotonically rising function in A, UT reaches a HT maximum at AT. (13) [,all66 X1 (1-a6] . where Z [Iai8XT( l)5yT X n FTYT Y YXT ] and A, XT and YT are defined as before. By substituting the expression for X in place of A in U* and U* and taking the difference between the two, one can show that UT H TT dominates UH at AT. A detailed comparison of the two utility functions and the farm size ranges over which one dominates the other is provided in the following sub-section. The Switch from a Hand Cultivation Regime to an Animal Traction Regime The switch from a handhoe based to an animal traction based farming system occurs when the indirect utility function associated with animal technology (UT) dominates the one associated with hand technology (U*). Graph 1 depicts this situation. H The extreme right end of the horizontal axis represents areas where per capita land availability is very high, these are typically sparsely populated areas under shifting cultivation. Movement along the horizontal axis towards the origin indicates a decline in per capita land availability (due to higher population densities) and therefore higher intensities of farming, For given parameter values, UT dominates U* for all values of A between S1 and S2' In other words, handhoe farmers switch to animal - 12 - Graph 1:: Indirect Utility Functions Under.Handhoe and Animal Traction Regimes Utility U* , f L IAU* AF. 2 ATS. // Increasing Land Scarcity U* : Indirect utility function associated with a handhoe technology. H U* : Indirect utility function assoicated with an animal draft technology. T UF : Indirect utility function with very high fixed costs for animal F maintenance S : Switch point from handhoes to animal draft technology. S 0: Switch point from handhoes to animal draft technology when permanent fallows are allowed. S2 : Switch point from animal draft back to handhoes. AN: The point where UT reaches a maximum. T T - 13 - draft technology at the point Si in order to avoid further declines in utility levels. A switch to animal draft technology for values of A greater than Si would not be desirable since the marginal disutility associated with the increased labor requirements for destumping fields and for year round maintenance of animals would exceed the marginal utility of output produced with the technology. The above result holds on the assumption that all the land available to the farmer is either under cultivation or under fallow. Suppose we allowed a third category -- land that the farmer chooses not to cultivate at all, i.e., permanent fallow land, then the switch point to animal draft power is at S1'. At Si' the farmer is faced with the choice of declining utility levels under a handhoe technology or maintaining his utility levels by switching to animal draft power. Such a switch would be possible if the farmer cultivates land area AT while land area (Sl'-AT) is left under permanent fallow. By making the switch at Si' th-e farmer avoids the decline and the subsequent rise in utility levels associated with a switch in technology at Si. One can show that an increase in the value of 8 leads to an increase in the va.Jue of Si, i.e., improvements in market access, other things heLd constant, leads to a switch to animal draft power earlier in the intensification process. The increased utility of output makes it more valuable relative to leisure. This result is reinforced by the complimentarity between draft power and manure use. Graph I also shows that for values of A less than S U* 2' H dominates UJ*. When farm sizes are very small the fixed labor costs T - 14 - associated with year round care and maintenance of animals far exceeds the utility of additional output, hence farmers revert back to hand cultivation on very small plots of land.4/ We have no documentation of cases where such reverse transition back to handhoes has occurred. Moreover, such a transition from animal draft back to hand cultivation is not important in relatively land abundant Sub-Saharan Africa and we will therefore not examine it in the empirical section of this paper. The above results are relevant when parameter values are such that U* and U* intersect. There are conditions, of course, where such H T F intersection may not occur. For instance, consider the curve UT, this is an indirect utility function associated with animal power technology in an environment with extremely high fixed costs of animal care and maintenance. Some of the animal disease infested zones of Sub-Saharan Africa would be good examples of this situation (trypanosomiasis zones, for example). Under such conditions, hand cultivation would be the dominant technology for all values of A. Now suppose a farmer had a choice between handhoes and tractors. The switch from handhoes to tractors would require: (a) higher quality of destumping; (b) higher learning costs; and (c) higher fixed costs for maintenance relative to animal traction technology. Moreover, tractor farmers have to forego the manuring benefits available to animal draft farmers. Under these conditions of higher fixed costs, one can show quite easily that the s; -.tch fom handhoes to tractors would occur later tharn the switch from handhoes to animal draft (assuming no tractor rental markets). 4/ Note that in the real world one might expect draft power markets or inequality in land holdings to emerge, and this result may therefore be a bit artificial. - 15 - Therefore, in most situations, a direct transition from handhoes to tractors is not economical and animal draft power is a necessary intermediate stage. Hypothesis The following testable hypothesis are consistent with the model presented above: 1. Other things held constant, increases in population density and/or improvements in market access will lead to an increase in the intensity of land use. 2. Other things held constant, increases in population density and/or improvements Ln market access will lead to the substitution of organic fertilizers for fallow periods in the maintenance of soil fertility. 3. Other things held constant, increases in population density and/or improvements in market access will lead to the substitution of mechanical power for human labor (animal draft power or tractor power). 4. Other things held constant, areas of higher inherent fertility (higher agro-climatic potential) are more intensively cultivated, use higher levels of organic fertilizer and are more likely to switch to mechanical power. 5. Holding the power source constant, an increase in the intensity of land use is associated with declining labor productivity. - 16 - 6 A change in the power source can be motivated entirely by labor saving benefits and could therefore occur even in the absence of yield benefits. 7. In addition to the above, we will test the Boserup hypothesis on land rights: rights to the acquisition and the use of land become well defined with population growth and/or improvements in market access. - 17 - II. Data and Empirical Results Empirical tests of the above hypothesis were performed on two distinct data sets. Hypotheses 1, 2, 3, 4 and 7 were tested using data from fifty-six villages in ten countries of Sub-Saharan Africa and India. This data was collected using the group interview technique during field visits in 1983-84. See Pingali, Bigot and Binswanger (1986) for further detailed description of this data. Presentation of the hypothesis in terms of two-way tables can also be found in that source. Hypotheses 5 and 6 on labor use and labor productivity were tested using a world wide data set assembled from several published research studies and surveys (see Pingali and Binswanger, 1984, for a detailed description of the data). Discussion of the empirical results will accordingly be separated into two parts based on the data set used. A. Intensity of Land Use, Technology and Land Rights The independent variables used in testing hypotheses 1-4 and 7 are population density, market access and agro-climatic potential. The dependent variables are intensity of land use, power source, organic fertilizer use and land availability. Ordered multinomial logit regressions were run on each of the dependent variables relating them to the three independent variables. The following is a description of the variables and the empirical results. Population density is measured as the rural population per kilometer square within the administrative unit (district) in which the - 18 - village is located. This information was obtained either from the agricultural officer in charge of the district or from census records. Market access is a scale variable from 1 to 5, where 1 indicates poor market access and 5 denotes excellent market access. The relative position of the village on this scale was determined by the investigators during the field visits in terms of the distance of the village from an inter-regional road/railroad and the quality of the connecting road (all weather road or dry weather road). Inherent fertility or agro-climatic potential of a location is approximated by the average annual rainfall. The relationship between agro-climatic potential and rainfall is not monotonic, it first rises with rainfall and then begins to decline. Descriptions of the dependent variables are presented along with a discussion of the empirical results. Table 1 presents the multi-nominal logit estimates. Intensity of Land Use is defined as the frequency with which a plot of land is cultivated and is measured as a discreet variable from 0 to 5 (0: Forest Fallow; 1: Bush Fallow; 2: Grass Fallow; 3: Emerging Annual Cultivation; 4: Annual Cultivation; 5: Multi-cropping systems). The first multinumial logit estimate in Table 1 shows a highly significant positive relationship between population density and farming intensity. Forest and bush fallow systems are predominant under sparse population densities. As population densities increase one observes the transition to more intensive systems of land use, first to grass fallow, then to annual cultivation and finally to multi-cropping where two or more crops are - 19 - cultivated (sequentially) on the same plot each year (hypothesis l). We have not come across any cases of sparsely populated areas under permanent cultivation systems or cases of very densely populated areas under shifting cultivation systems. Presumably, the former could occur under sparse population densities if market access is excellent, while the latter could not occu7 even under poor market conditions. The first multinomial logit regression also shows a very significant positive relationship between market access and the intensity of land use (hypothesis 1). For given population density, an improvement in market access results in more intensive use of existing land resources. Improvements in market access allow for better terms of trade between farm output and imported consumer goods, and thereby accelerate the pace of intensification. Contrary to our expection, intensity of land use had a significant negative association with rainfall. This was possibly because of the inclusion of several locations from the humid tropical zone in our sample. Field crop production in this zone is traditionally carried out at low intensities due to che susceptibility of the soils in this zone to high levels of leaching and acidification. Evolution in Tool Systems Tools used for tillage are defined by a qualitative variable as follows: 0 represents handhoe cultivation systems; I represents the emergence of animal traction; and 2 denotes well established animal traction and the emergence of a few tractors in the community. Multinomial logit estimates in column 2 of Table I present the 20- relationship between tools used, population density, market access and agro-climatic potential. There is a significant positive relationship (at 10% level of significance) between population density and the evolution from handhoes to animai draft power. Also, for given population density, improvements in market access have a very significant positive effect (at 1% level of significance) on the transition to animal draft power. These results verify hypothesLs 3. Rainfall level has a very significant positive effect and the square on rainfall has a significant negative effect on the evolution in tools systems. This indicates that higher rainfall areas (up to a maximum of approximately 1300mm. per annum) tend to switch to animal draft power sooner that areas of lower rainfall. Organic Fertilizer Use is defined in terms of the labor input required for its production. The movement being from less to more labor intensive techniques for maintaining soil fertility. The variable is denoted as follows: 0 represents no organic fertilizer; 1 represents the use of kraal dust which can be produced with very low labor input; 2 represents limited use of household refuse which requires a moderate level of labor input; and 3 represents manure production. The very significant positive effect (at 10% level of significance) of population density and market access (at 5% level of significance) are consistent with hypothesis 2. As the intensity of land use increases due to population growth and/or improvements in market - 21 - infrastructure, fallow periods are replaced by more and more labor intensive fertilizing techniques such as composting and then manuring. Changes in Land Rights During our field visits we asked a series of questions that allowed us to determine if and how outsiders. could acquire land in a particular village. Outsiders are defined as individuals who are not members of the lineage to which the residents of the village belong. We categorized land acquisition as follows: 0 represents easy access to land, where outsiders could obtain land merely by asking; I represents delayed access to land, where outsiders would have to work for a year or more as farm labor before acquiring the privilege to cultivate their own plots; 2 denotes no access to land, where outsiders cannot acquire any land in the village but there are as yet no direct sales of land and 3 represents private property, where direct sales of land are possible and prevalent. There is a significant positive relationship between access to land, population density and market access (10% and 1% level of significance, respectively). This result verifies hypothesis 7. It is only in sparsely populated locations that we find easy access to land to be generally true. As population densities increase one observes a movement towards privatization of agricultural land. For given population density, areas with better access to the market are more likely to deny outsiders access to land and move more rapidly towards codified private property. B. Labor Use and Labor Productivity with Farming Intensity Empirical tests of hypothesis 5 and 6 on labor use and labor D 22 - productivity are presented in this sub-section. The data used for this part of the-analysis was compiled from the literature. -This data set, presented in Appczndix 1, contains information on 52 specific locations in Africa, Asia and Latin America. The following variables were used in the analysis: intensity of land use, labor use per hectare, yields, mechanical inputs and land investments. Intensity of land use was measured by the frequency with which a plot-of land was cultivated. Mechanization was defined as the substitution of animal and tractor power for hand cultivation. Dummy variables were used for animal traction, tractor use, and land investment. Land investments include destumping, leveling, drainage, irrigation, etc., the fertilizer use variable includes both organic and chemical fertilizers. Labor input was measured in hours per hectare and yield was measured per hectare and per labor hour. Labor Use and Intentsty of Land Use As discussed by Boserup (1965) and Ruthenberg (1980), holding tools constant, the total labor input per hectare on a given crop is positively correlated with the intensity of farming. The increase in labor input occurs because: (a) the number of operaticns which have to be performed increases (e.g., manuring, irrigation, etc.), and (b) the intensity with which certain tasks have to be performed increases (.e.g, land preparation and weeding). Table 3 shows how the operations performed increase with the intensification of the farming system. A log-linear regression was used to test the hypothesis of a positive correlation between labor use and the intensity of land use. .. ..... . -23 - The results of the regression are presented in Table 2. Intercept dummies were used for animal traction, tractor use and land investments. Since our earlier model treats labor input and choice of technique both as decision variables, these regressions represent a correlation analysis, but do not establish causality as did the regressions in Table 1. The hypothesis that the transition to more intensive systems of land use are associated with longer working hours is validated. A 10% increase in intensity of land use is associated with a 4.6% increase in labor-use per hectare. This positive relationship between land use intensity and labor use per hectare holds true irrespective of the types of tools used, The data set was separated by each type of tool and the coefficient on farming intensity estimated. The null hypothesis of equality of coefficients could not be rejected. Animal traction and tractor use act as intecept shifters. For given intensity of farming the switch to animal traction or tractors :ls significantly associated with a reduction in the total labor use per hectare. Land investments were not found to significantly affect labor use in cultivation. However, the labor use data generally did not include the time spent for investments such as destumping, levelling, irrigation, etc., and for the maintenance of animals and machines. This overhead labor obviously does increase with land investments. Results presented in Table 2 also validate the hypothesis that yield per hectare rises with intensity of land use, holding the level of mechanization constant. A 10% increase in land use intensity is associated with a yield increase of 3.9%. The level of mechanization on the other -24- hand is not significantly associated with yield. On the other hand, for given intensity of land use, land investments are associated with yield increases of 30%. Land use intensity was not significantly associated with output per man-hour though the negative sign on the coefficient was as hypothesized. The decline in output per man-hour of cultivation labor may have been prevented by additional capital investments in land. The additional labor required for these investments is not reflected in this data set. It is likely that output per man-hour of total labor (cultivation and overhead) would show the hypothesized declining trend. Labor productivity is positively associated with mechanization: for given intensity, output per man-hour is 78% higher on animal traction farms than on handhoe farms and 42% higher on tractor farms than on animal traction farms. - 25 - Conclusions 1. This paper provides empirical support to Boserup's thesis on population induced technical change in traditional agricultural societies. In addition to population growth we show that improvements in market access through better transport infrastructure have similar effects on technical c'ange. This paper shows that an increase in population density and/or improvements in markec access cause a reduction in fallow periods due to increasing land scarcity and therefore lead to the evolution of the agricultural system from shifting cultivation to permanent cultivation of land. Intensification of land use is generally associated with increased labor requirements per hectare, and therefore diminishing returns to labor when technology is held constant. 2. Far from being immobile and technologically stagnant "traditional" African societies have responded to changes in population densities and external markets with changes in farming systems, land-use patterns, technology and institutions along systematic and predictable patterns. 3. The switch from the handhoe to animal drawn plows and later to tractors is closely associated with the evolution of farming systems. This switch is induced by the labor saving benefits associated with animal draft and tractor power. The switch to animal drawn plows occurs around the short fallow stage and not before because it is only at this stage that the overhead labor costs for destumping and leveling fields and for training, feeding and maintaining animals are offset by benefits in terms of labor - 26 - savings. The substitution of fallowing first with simple and then with more evolved manuring techniques is likewise related to the evolution in farming systems. 4. Institutional arrangements for the acquisition of land by individuals within the group and by "outsiders" are not rigid but do change as increasing population densities or improved market access makes land scarce. Land aquisition which is extremely easy under shifting cultivation becomes more and more difficult as intensification leads to more narrowly defined groups or lineages and therefore results in the exclusion of large numbers of people from acquiring the rights to cultivate. The ultimate institutional change, and one which commonly occurs under high population densities is one of clearly defined private property rights with the ability to buy and sell land. 5, As we know from research on developed countries, declines in labor productivity can be offset and more than offset by mechanical inputs and biological technologies. All countries of the developed world have achieved rates of labor and land productivity sufficient to outstrip their rates of population growth. This cannot, however, be cause for assuming that rapid population growth presents no problem. First of all, developing country population growth rates exceed historical rates in developed countries substantially. Moreover, the biological technologies required often have to be invented or adapted. They cannot simply be pulled from a shelf. In a number of countries the institutional capacity to invent these technologies is rudimentary at best. Even the mechanical technologies require adaptive learning. They also require capital accumulation of a - 27 magnitude which is very hard to achieve given current capital endowments of the societies. 6. The potential problems of declining labor productivity or of environmental degration are not problems of levels of population densities. Given sufficient time it is likely that a combination of farmer inventions, savrings, development of research institutions and institutions to deal with land tenure and soil degradation issues will be able to accommodate much larger than current population in most countries, and especially in many of the low density ones. However, when populations grow rapidly, all these changes in technology and institutions are required rapidly and simultaneously and they may fail to emerge at a sufficiently rapid pace to prevent decline in human welfare. - 28 - Table 1: Ordered Multinomial Logit Estimates Dependent Variable Intensity Tools Organic Land of Land Use Used Fertilizer Availability Independent Variable (1) (2) (3) (4) al 5.04 -4.57 0.024 21.05 (2.53) (2.53) (2.42) (11.456) a2 2.42 -6.66 -1.92 19.70 (2.40) (2.54) (2.44) (11.45) a3 0.77 -9.27 -2.59 16.40 (2.39) (2.73) (2.45) (11.37) a4 -0.03 (2.40) a5 -2.47 - - (2.44) Population 0.019*** 0.007* 0.008* 0,038*** Density (per km2) (0.005) (0.004) (0.004) (0.013) Rainfall -0.007** 0.008*** -0.0004 -0.05 (mm/year) (0.003) (0.003) (0.004) Rainfall Sq. 0.000002 -0.000003*** -0.0000007 0.00003 (0.000001) (0.000001) (0.0000014) Market Access- 0.875*** 0.658* 67*** 0.95** (0.27) (0.26) (0.28) (0.46) R. 0.475 0.315 0.377 0.608 Notes (i) Standard errors are given in parentheses. (ii) *, **, ** denote significance levels of 10%, 5% and 1%; respectively. (iii) a1-a5 are intercepts used in estimating the probability that the dependent variables takes a certain value. For instance, P(YD ) 5 1e - 29 - Table 2: Intensity of Land Use Related to Labor Use Per Hectare Yield Per Hectare, Yield per Hour of Cultivation Labor Log Labor Log Yield Log Yield/ Use Per per Hectare per Hour of Hectare cultivation Labor Log Land Use 0.456*** 0.389*** -0.068 Intensity (0.132) (0.097) (0.15) Animal -0.96*** -0.179 0.78*** Traction (0.23) (0.17) (0.26) Tractor -1.14*** 0.068 1.20*** (0.27) (0.20) (0.31) Land 0.057 0.299** 0.24 Investments (0.20) (0.148) (0.23) Intercept 5.46 5.69 0.24 0.33 0.45 0238 Number of Observations 56 56 56 Notes (i) *** significant at 1%; ** significant at 5% (ii) Standard errors are given in paretheses. (iii) Does not include labor use in overhead activities such as land investments and animal or machine maintenance. - 30 - Table 3: Comparison of Operations across Farming Systems Syst ems Operations --- - -k----- --------- --------- -------------- - - -- -- FB AC meC i.and Clearing Fl re Fi re None Noe None Land Nt, lanid prep.aration I.and I loosened Ilse *,f plow for Aiiflmi dIrawn plows Ammimal dratwl phlows Preparation 6 digging sticks used usiniiig hoes and preparing lind antl tractors andl tractors Plantinig to plant root.sc and digging tticks s5w seeds Hantare Ilse - Ash - Ash - Animal dsung or - Hamiuare somnetlines - H-anutre, sometimes - Perhaps Ihniselkold - Sometlmens ciiitemene Manirtir glimniman waste human waste reftusfe for garden techni qlues - Somptimes composting - CompostIiug - Gomposting - hlotisehold refuse - Cualtivation of - Cultivationt of for garden plots green maliiure crops green mainure crops - Clhemilcal fertilizers - Chemical fertilizers WeedIlag Ml itl I.mat Reilil red as rite Intensilve weedling InitensiI ve weeding Intensive weeding I t ngtht of t a l low reips I red reql, t red reuit I red dec reases llue of Animals NSuie As legLth of -- Plowing - IP lwilng - P'lowing Fariai iog falllow decreases - Transport - Tranisport - Transport animal dIrawI plOws - Interctiltuire - InterculItuire - Interculture begin to) appear - Ioet-harvest tasiks - Post-harvest tasks - Irrigatloni - Irrigation Seasonality of Minimal Wetilng einerges Land lpreparit ion, L.and preparation, Acilte peak arouind land lal)bor demnand as a peak weedilng ;nd itarvestiag weeuling and hlalrvesting preparation, liarvest, and post-lharvest tasks Fodder Supply Nonoe Emeugence Of Abundant Open Open graziig Intensive fodder grazinig lanitd grazinig restriected to maniagemesnt antd fodder ma rgicial lands antd crop produict lon. stObble graizing. - 31 - References Binswanger, H.P. "The Measurement of Technical Change Biases with Many Factors of Production," American Economic Review Vol. 64, No. 6. Binswanger, H.P. and John McIntire. "Behavioral and material Determinants of Production Relations in Land Abundant Tropical Agriculture," Economic Development and Cultural Change (forthcoming, 1986). Boserup, Ester. The Conditions of Agricultural Growth. Chicago: Aldine Publishing Co., 1965. -__ . "The Impact of Population on Agricultural Output," Quarterly Journal of Economics May, 1975. . Population and Technological Change: A Study of Long- Term Trends. Chicago: University of Chicago Press, 1981. Clark, Colin. Population Growth and Land Use. New York: Macmillan, 1968. Darity, William A. "The Boserup Theory of Agricultural Growth: A Model for Anthropological Economics," Journal of Developmental Economics (7):137-157, 1980. Hayami, Yujiro and Vernon Ruttan. "Factor Prices and Technical Change in Agricultural Development: The United States and Japan, 1880-1960," Journal of Political Economy 78 (Sept.-Oct.):1115-1141, 1970. __ _ _ _ _ _ _ . Agricultural Development: An International Perspective (Revised Edition). Baltimore: Johns Hopkins University Press, 1985. Kuznets, Simon. "Population Change and Aggregate Output," in Universities-National Bureau of Economic Research, Demographic and Economic Change in Developed Countries. Princeton: Princeton University Press, 1960. Lee, Ronald D. "Malthus and Boserup: A Dynamic Synthesis." Paper presented at the conference "Forward from Malthus." Cambridge, :ngland, September 1984. - 32 - Pingali, Prabhu L. and Hans P. Binswanger. Population Density and Agricultural Intensification: A Study of the Evolution of Technologies in Tropical Agriculture. Agricultural and Rural Development Dept, Research Unit, Report No. ARU-22. Washington, DC: World Bank, 1984. Pingali, Prabhu L., Yves Bigot and Hans P. Binswanger. Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa. Agriculture and Rural Development Dept., Research Unit, Report No. ARU40. Washington, DC: World Bank, 1985. Pryor, Frederic L. and Stephen B. Maurer. "On Induced Economic Change in Pre-capitalist Societies," Journal of Developmental Economics (10):325-353, 1982. Robinson, Warren and Wayne Schutjer. "Agricultural Development and Demographic Change: A Generalization of the Boserup Model," Journal of Economic Development and Cultural Change 32 (2):355-366, 1984. Ruthenberg, Hans. Farming Systems in the Tropics, 3rd Edition. Oxford: Clarendon Press, 1980. Simon, Julian. The Economics of Population Growth. Princeton: Princeton University Press, 1977. Stryker, J. Dirck. "Population Density, Agricultural Technique and Land Utilization in a Village Economy," American Economic Review 66 (3):347-358, 1976. - 33 - DISCUSSION PAPERS AGR/Research Unit Report No.: ARU 1 Agricultural Mechanization: A Comparative Historical Perspective by Hans P. Binswanger, October 30, 1982. Report No.: ARU 2 The Acquisition of Information and the Adoption of New Technology by Gershon Feder and Roger Slade, September 1982. Report No.: ARU 3 Selecting Contact Farmers for Agricultural Extension: The Training and Visit System in Haryana, India By Gershon Feder and Roger Slade, August 1982. Report No.: ARU 4 The Impact of Attitudes Toward Risk on Agricultural Decisions in Rural India by Hans P. Binswanger, Dayanatha Jha, T. Balaramaiah and Donald A. Sillers, May 1982. Report No.: ARU 5 Behavior and Material Determinants of Production Relations in Agriculture by Hans P. Binswanger and Mark R. Rosenzweig, June 1982 (Revised July 22, 1985). Report No.: ARU 6 The Demand for Food and Foodgrain Quality in India by Hans P. Binswanger, Jaime B. Quizon and Gurushri Swamy, November 1982. Report No.: ARU 7 Policy Implications of Research on Energy Intake and Activity Levels with Reference to the Debate of the Energy Adequacy of Existing Diets in Developing Countries by Shlomo Reutlinger, May 1983. Report No.: ARU 8 More Effective Aid to the World's Poor and Hungry: A Fresh Look at United States Public Law 480, Title II Food Aid by Shlomo Reutlinger, June 1983. Report No.: ARU 9 Factor Gains and Losses in the Indian Semi-Arid Tropics: A Didactic Approach to Modeling the Agricultural Sector by Jaime B. Quizon and Hans P. Binswanger, September 1983 (Revised May 1984). Report No.: ARU 10 The Distribution of Income in India's Northern Wheat Region By Jaime B. Quizon, Hans P. Binswanger and Devendra Gupta, August 1983 (Revised June 1984). - 34 - Discussion Papers (continued) Report No.: ARU 11 Population Density, Farming Intensity, Patterns of Labor-Use and Mechanization by Prabhu L. Pingali and Hans P. Binswanger, September 1983. Report No.: ARU 12 The Nutritional Impact of Food Aid: Critria for the Selection of Cost- Effective Foods by Shlomo Reutlinger and Judith Katona-Apte, September 1983. Report No.: ARU 13 Project Food Aid and Equitable Growth: Income-Transfer Efficiency First! by Shlomo Reutlinger, August 1983. Report No.: ARU 14 Nutritional Impact of Agricultural Projects: A Conceptual Framework for Modifying the Design and Implementation of Projects by Shlomo Reutlinger, August 2, 1983. Report No.: ARU 15 Patterns of Agricultural Projection by Hans P. Binswanger and Pasquale L. Scandizzo, November 15, 1983. Report No.: ARU 16 Factor Costs, Income and Supply Shares in Indian Agriculture by Ranjan Pal and Jaime Quizon, December 1983. Report No.: ARU 17 Behavioral and Material Determinants of Production Relations in Land Abundant Tropical Agriculture by Hans P. Binswanger and John McIntire, January 1984. Report No.: ARU 18 The Relation Between Farm Size and Farm Productivity: The Role of Family by Gershon Feder, December 1983. Report No.: ARU 19 A Comparative Analysis of Some Aspects of the Training and Visit System of Agricultural Extension in India by Gershon Feder and Roger Slade, February 1984. Report No.: ARU 20 Distributional Consequences of Alternative Food Policies in India by Hans P. Binswanger and Jaime B. Quizon, August 31, 1984. Report No.: ARU 21 Income Distribution in India: The Impact of Policies and Growth in the Agricultural Sector by Jaime B. Quizon and Hans P. Binswanger, November 1984 (Revised October 1985). - 35 - Discussion Papers (continued) Report No.: ARU 22 Population Density and Agricultural Intensification: A Study of the Evolution of Technologies in Tropical Agriculture by Prabhu L. Pingali and Hans P. Binswanger, October 17, 1984. Report No.: ARU 23 The Evolution of Farming Systems in Agricultural Technology in Sub-Saharan Africa by Hans P. Binswanger and Prabhu L. Pingali, October 1984. Report No.: ARU 24 Population Density and Farming Systems - The Changing Locus of Innovations and Technical Change by Prahbu L. Pingali and Hans P. Binswanger, October 1984. Report No.: ARU 25 The Training and Visit Extension System: An Analysis of Operations and Effects by G. Feder, R. H. Slade and A. K. Sundaram, November 1984. Report No.: ARU 26 The Role of Public Policy in the Diffusion of New Agricultural Technology by Gershon Feder and Roger Slade, October 1984. Report No.: ARU 27 Fertilizer Subsidies: A Review of Policy Issues with Special Emphasis on Western Africa by Haim Shalit and Hans P. Binswanger, November 1984 (Revised November 1985). Report No.: ARU 28 From Land-Abundance to Land Scarcity: The Effects of Population Growth on Production Relations in Agrarian Economies by Mark R. Rosenzweig, Hans P. Binswanger and John McIntire, November 1984. Report No.: ARU 29 The Impact of Rural Electrification and Infrastructure on Agricultural Changes in India, 1966-1980 by Douglas F. Barnes and Hans P. Binswanger, December 1984. Report No.: ARU 30 Public Tractor Hire and Equipment Hire Schemes in Developing Countries (with special emphasis on Africa). A study prepared by the Overseas Division, National Institute of Agricultural Engineering (OD/NIAE) by P. J. Seager and R. S. Fieldson, November 1984. Report No.: ARU 31 Evaluating Research System Performance and Targeting Research in Land Abundant Areas of Sub-Saharan Africa by Hans P. Binswanger, January 1985. - 36 - Discussion Papers (continued) Report No.: ARU 32 On the Provision of Extensioil Services in Third World Agriculture by Alastair J. Fischer (Consultant), January 1985. Report No.: ARU 33 An Economic Appraisal of Withdrawing Fertilizer Subsidies in India by Jaime B. Quizon, April 1985 (Revised August 1985). Report No.: ARU 34 The Impact of Agricultural Extension: -A Case Study of the Training and Visit Method (T&V) in Haryana by Gershon Feder, Lawrence J. Lau and Roger H. Slade, March 1985. Report No.: ARU 35 Managing Water Managers: Deterring Expropriation, or Equity as a Control Mechanism by Robert Wade, April 1985 Report No.: ARU 36 Common Property Resource Management in South Indian Villages by Robert Wade, April 1985. Report No.: ARU 37 On the Sociology of Irrigation: How do we Know the Truth about Canal Performance? by Robert Wade, May 1985. Report No.: ARU 38 Some Organizations Concerned with Animal Traction Research and Development in Sub-Saharan Africa by Paul Starkey, April 1985. Report No.: ARU 39 The Economic Consequences of an Open Trade Policy for Rice in India by Jaime B. Quizon and James Barbieri, June 1985. Report No.: ARU 40 Agricultural Mechanization and the Evolution of Farming Systems in Sub-Saharan Africa by Prabhu L. Pingali, Yves Bigot and Hans P. Binswanger, May 1985. Report No.: ARU 41 Eastiasian Financial Systems as a Challange to Economics: The Adventures of "Rigidity," with particular reference to Taiwan - by Rober Wade, June 1985. Report No.: ARU 42 Education, Experience and Imperfect Processing of Information in the Adoption of Innovations by Alastair J. Fischer, June 1985. Report No.: ARU 43 A Review of the Literature on Land Tenure Systems in Sub-Saharan Africa by Raymond Noronha, July 1985. - 37 Discussion Papers (continued) Report No.:. ARU 44 Policy Options for Food Security. by Shlomo Reutlinger, July 1985, Report No.: ARU 45 Credit Markets in Rural South India: Theoretical Issues and Empirical Analysis.3 by H. Binswanger, T. Balaramaiah, V. Bashkar Rao, M.J. Bhende and K.V. b.ashirsagar, July 1985. Report No.: ARU 46 The Impact of Agricultural fSltension: The Training and Visit System in India. by Gershon Feder and Roger Slade, June 1985. Revort No.: ARU 47 Methodological Issues in the Evaluation of Extension Impact. by Gershon Feder and Roger Slade, July 1985. Report No.: ARU 48 Escimacion of Aggregate Agricultural Supply Response. by Hans Binswanger, Yair Mundlak, Maw-Cheng Yang and Alan Bowers August 1985 (Revised October 1985). Report No.: ARU 49 Land Values and Land Title Security in Rural Thailand. by Yongyuth Chalamwong and Gershon Feder, June 1985 (Revised October 1985). Reoort No.; ARU 50 Land Ownership Security and Capital Formation in Rural Thailand. by Gershon Feder and Tongroj Onchan, December 1985 (Revised February 1986). Report No.: ARU 51 Land Ownership Security and Farm Productivity in Rural Thailand. by Gershon Feder, April 1986. Reoor ; No.: ARU 52 Social and Cultural Aspects of Land Inheritance and Transactions in Rural Thailand. by Charles B. Mell, June 1986. - 38 - Discussion Papers (continued) Report No.: ARU 53 Land Ownership Security and Access to Credit in Rural Thailand by Gershon Feder, Tongroj Onchan and Tejaswi Raparla, April 1986. Report No.: ARU 54 The Management of Common Property Resources: Collective Action as an Alternative to Privatization or State Regulation by Robert Wade, May 1986. Report No.: ARU 55 Land Policies and Farm Productivity in Thailand's Forest Reserve Areas by Gershon Feder, Tongroj Onchan and Yongyuth Chalamwong, August 1986. Report No.: ARU 56 On the DeterminarLts of Cross-Country Aggregate Agricultural Supply by Hans P. Binswanger, Yair Mundlak, Maw-Cheng Yang and Alan Bowers, September 1986. Report No.: ARU 57 What Can Agriculture Do for the Poorest Rural Groups? by Hans P. Binswanger and Jaime B. Quizon, September 1986. Report No.: ARU 58 Population Density, Market Access and Farmer-Generated Technical Change in Sub-Saharan Africa. by Prabhu L. Pingali and Hans P. Binswanger, September 1986.