lecture 8 reading---Bilsborrow,+2001+_CAB+Intl_

lecture 8 reading---Bilsborrow,+2001+_CAB+Intl_ -...

Info iconThis preview shows pages 1–21. Sign up to view the full content.

View Full Document Right Arrow Icon
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 2
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 4
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 6
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 8
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 10
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 12
Background image of page 13

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 14
Background image of page 15

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 16
Background image of page 17

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 18
Background image of page 19

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Background image of page 20
Background image of page 21
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Population, Agricultural Land Use and the Environment in Developing Countries RICHARD E. Buseoa‘aow1 AND DAVID L. CARR2 ’Carolina Population Center and 2 Department of Geography, University of NOrth Carolina at Chapel Hill, Chapel Hill, North Carolina, USA Introduction The earth’s human population exceeded 1 billion for the first time in 1850. By 1999, the world’s population reached 6 billion. The primary reason for such rapid growth during the past century and a half is that, despite the ongoing declines in mortality due to improved health conditions, fertility remained high for another generation or two in most countries (Teitelbaum, 1975). It is only since the 19703 that fertility has declined in most parts of the developing world. in fact, the developing nations will be responsible for the entire increase of 24 billion persons that will be added to the global population during the next 50 years (based upon the United Nations (UN) medium fertility projections). _ This will increase theirlshare of‘the global population from three—quarters to eightmninths. Concomitantly, the loss of global forest cover is linked solely to deforestation in the developing world.1 Once covering most of the earth, global forests have fallen to approximately a quarter of the terrestrial surface (FAQ, 1995). If rates continue to accelerate as they did from 1984 to 1994 (by 40%), all primary tropical forests will be destroyed in another 50 years (Houghton, 1994). ~ Since sedentary agriculture dawned on the fertile banks of the Ganges, _‘ Tigris and Nile Rivers some 10,000 years ago (although recent evidence indi~ (cates that the first cultivated maize in central Mexico goes back nearly as far), growing populations have been fed by the expansion of agricultural land, mostly at the expenSe of forested areas. Despite technological advances and an increase in crop intensification, agricultural conversion was most dramatic dorm .. ing the 20th century. In 1900, 40 Mha of land were in crops around the world; - by 1993, cropland had risen to 248 Mha (Brown, 1997). Resource degradation ~':_ linked to rapid deforestation has led to chronic underdevelopment in rural areas ds‘ DR. tee and CB. Barrett} . 35 Where prima Dasgupta, 1995b). Since the earth’s land is finite there as in any other world region (Wm-I case studies illustrates th change and land—use ch Finally, we sfimmarize Population, Land Use and the Environment 37 ______________________________.__._._——-——~——-—-—— that is, as available arable land becomes scarce relative to labour, societies adopt more labouruintensive techniques, which take advantage of increased labour—- land factor ratios. Boserup enumerated five stages of land intensification through which agrarian societies evolve over time: (i) forest or long fallow (29—25 years between crops); (ii) bush fallow (6w10 years); (iii) short fallow (1%! years); (iv) annual cropping; and (v) multiple cropping (more than one crop per I year on the same land). These forms of increasingly intensive use of land embody increasing inputs of labour per unit of land per unit of time.2 Based on a review of historical processes in Japan and Europe, Davis (1963) proposed that demographic responses to population pressures on the land may occur: that is, rural populations may respond by reducing fertility through postponement of marriage and increased celibacy, increased regulation of fertility Within marriage and/or increasing abortions. Davis considered out— migration as a last resort if these responses proved inadequate. He also proposed that the various responses may occur simultaneously (or ‘multiphasically’), and that the more one response occurs -— and the more the effects of population pressures on the land are thereby relieved “- the less likely other responses are to occur. 7 In support of Davis’s hypothesis, Friedlander (1969) found that fertility reduction preceded out—migration in preindustrial France. Scarce employment opportunities in French urban centres deterred rural—urban migration. With shrinking farmland per capita due to high fertility and the absence of out~ migration possibilities, France thus became the first European nation to experience declining marital fertility, even prior to industrialization. According to Friedlander, fertility will decline to the extent that communities are constrained in relieving population pressures through out-migration. However, studies of demographic responses (e.g. Davis, 1963; Friedlander, I969; Mosher, 1980) neglect economic and land-use responses and vice versa. An attempt at synthesizing theoretical views is found in Bilsborrow (1987), who classified responses as economic (land intensification 'and‘ extensification); ’ demographic (fertility responses) and economicwdemographic (out-migration). file hypothesizes that households traditionally exhaust economic options first, beginning with land expansion. If that is insufficient, then available land intensification technologies may be adopted, involving a decline in leisure time. If such adjustments together still prove inadequate, the next reaction is likely to be out—migration, temporary or seasonal migration at first, since that does not require giving up one’s home and community. Permanent out—migration is likely to follow if prior adjustments, along with the remittances and savings brought by the temporary migrant, prove inadequate. This may take the form of either migration to urban areas, especially if wage levels are much higher there, or migration to other rural areas. The latter is more likely in countries with lands that have potential value for agriculture but are still unoccupied. This is particularly true of countries with significant endowments of tropical forests —— for example, in South America and Central Africa. Finally, since customs producing high fertility are so deeply rooted in culture, fertility reduction is postulated to occur in traditional societies only as a last resort, when other Rural growth population responses: (i) quantity and propinquity of potentially arable land (Boserup, 1965, 1981; Bilsborrow and Geores, 1994; Sambrook et al., 1999); (ii) quality of natural resources, as manifest in climate, rainfall, topography and soil quality (Sanchez and Cochrane, 1980; Blaikie, 1987; Barbier, 1990; Buol, 1995); (iii) type of land tenure regime (Forster and Stanfield, 1993; Utting, 1993; Schmink, 1994); (iv) land distribution and the extent of iandlessness (Stonich, 1989; World Bank, 1995a; Valenzuela de Pisano, 1996); (v) urban employment opportunities that may attract migrants (Stern, 1976; Alberts, 1977; Gilbert, 1994; Altamirano et 41., 1997); (vi) proximity and accessibility to product and labour markets (Rudel, 1983; Rudel and Richards, 1990); (vii) knowledge of and accoss to alternative forms of agricultural technology (Boserup, I965; Zimmerer, 1993); and (viii) government policies, including those related to land distribution and tenure; tax and expenditure policies (including subsidies), especially those ' ' ' extensification and intensification; transportation infrastructure (Bilsborrow, 1.987;, Hecht and Cockburn, 19%; Rudel and Richards, 1990); the fixing of prices for agricultural products (Garland Bedoya, 1991; Stewart, 1994); and family planning, education and health pol1c1es which affect fertility and mortality (Bilsborrow and Geores, 1992). With these Rural ~rurai migration and land extensification Agricultural intensification ' technology, modern 5 agricultural inputs : and credit Pressures lo respond Hurrah—urban migration Fertiiiiy decline . and health i j programmes : 21‘ ‘I? "n a Population, Land Use and the Environment 39 W conditioning mechanisms in mind, what does empirical evidence tell us about responses to population change among the major world regions? A Glanceat Global Relations in an earlier paper, the first author looked at continent—wide relationships from a broad multiphasic perspective, based on estimates of trends over time (Bilsborrow, 1987). Data on demographic and economic changes in five devei- oping country-regions from the 19603 to 1980 were presented. Demographic change was measured by the change in the total fertility rate and economic- demographic change by the annual rate of out—migration from rural areas. Economic (or agricultural) changes comprised land extensification, measured by the percentage change in the share of arable and cropped land in total land, and land intensification, measured by the percentage change in rural labour productivity. The developing-country regions with the largest responses of one type did indeed tend to have smaller responses of the other types, as predicted by the multiphasic approach, and major deviations from the expected tradeoffs were generally explainable by exogenous factors, such as differences across regions in natural resource endowments or government policies. Now that over a decadehas passed, it is desirable to update that overview. We have prepared estimates of changes in the same four factors, up to and including 1995 (Table 3.1), revising the earlier period estimates using data sources appropriate for the latter period. The measures of all variables are the same as earlier, except that better estimates of rural out—migration have become available recently from the UN (1997). The caveats pertaining to the earlier study — especially the riskiness of cross—country comparisons, due to incon— sistencies across countries and deficiencies in the underlying data — should still be kept in mind. ‘W e have presented data here for medians as well as fornmeans, and we note where the means would be considerably different with the exclusion of outliers}5 Note that what we are concerned with here is the association between the change in rural population density and the associated demographic, economic— demographic and agricultural changes over two time periods. We observe at the outset that, at a continent~wide level, there was a slight increase in rural population density in Latin America in the first period (19654980) but none _ in the second (19814995), as fertility and natural population growth declined ton-equal out-migration from rural areas.- In all other regions, however, rural population density rose substantially, most notably in China. In Africa, the Near East, China and especially ‘Other Asia’, the increase in rural density was much greater in the second period than in the first, implying that there shouid be greater responses. Regarding economic or land—use responses, we expect agricultural land to increase where there is untapped land, which occurred in Latin America and ‘Other Asia’, as‘expected. Secondly, agricultural labourproductivity rose _‘-substantially, except in sub—Saharan Africa, but this occurred more in the first time period (except for the Near East), probably due to the earlier exploitation .Rmm a 23 E9“ cmxfl 9m 8mm :25me .83: Z: Eoc .oEmEURU Emu \Emtwm . _ .623 039 $me aha: 04E 86¢ 8532mm. Smt 523305 3.5 mm: Ucfl :< .3285 g mmm Tomm M gov. 93meth 9:. .3533th cm nonmmaaoa méum >2mumEOcoum womb En. waE coauszth Eamatmm mam :kEEEuzmm E cosmiqoa m>zum xzmumEocoum mom F “ma xwvcm cozunuoi _S:::urmm wmm :imhssmutwm E c9338 gaum EmumEocoum ommw ba 5?: :96:an “Eamnucwm owm: “89:83 how 950:8 mm 322:2ng $3.5 new 3th SE mam .6333 :30 new: @3333 Much EwcmELmi‘ Emmy m 53:: Scamp“ E a mum. 1cm; ham 222mm gamma “m5 352 mm chmm «50305 $52953 bum mEmE 6 may} v.» ad ed mcmfiwfi EBB mmmqoqwxam Md: wd; El ml of w m o “l md wd 8:? m. T. m. T 8 ml mm: o w: m w o; No Nd “mam 2mmZ m... 7 od 0 mm NW? NY. m; 3.. Ym \ md QM SEQ m. w... md: u; mm val m T m m m; md md $3 .650 m. 7 n91 m p mm mm! 3.... m w vd m6 m6 $58.4 cum.“ m. T w. T m _ mm mm: A mm... m m1 QN m; md 385 3.63 $33960 m. 7: m6: m T _ m1 m . m hr A: o; «d at? m. T m. T K m Wm: Z: E El a; No md m3 52 m. 7 ed 0 mm ET 31 ml m 7 Ym wd 9m EEG 0&1 50.; m mm wml m? m" o n; N; E £m< 550 m. T m. T 3 mm mm: mm: m 3 m6 md md 8me< 523 memo" mommy mm? 83 mmm Tammi owm Flmmmw vamp mum» mmmh away 39 mcowmwm Imam? 18mm” I39 133 3 0mm Tmmmw 9 mam wiommfi A..an imam” swamp Immm» Imam". flailflilzll I]!!! Eli IIJJIIII Elli]! comfiwmfiézo 533.55%th Emu 323.5% Em“ Enact «new. mask 1:33 *0 . Boss EB E “Sow F. wmcmgu rucwcmctma bcm £39m \cozmfisuoa WEE 3mm mmcmzu 343:8me mmficmukmm cm wmzmgu ammucmuhmnm gauw 3mmumE0coum .mmmcoamm: quEmoEmt mum uEQEmOvaluéocoum ~3;ng new bacwv conmSaom “Wm «Bah Population, Land Use and the Environment 41 of Green Revolution technologies and the associated use of inputs. Fertility changes mainly occurred starting in the latter part of the first period —— most of the decline in Latin America before 1986 was during 1975—1979 ~ and accelerated during the second period, as family planning spread through most of the developing world. Meanwhile, out—migration from rural areas was sub- stantial in most regions in both periods, though again slightly higher in the sece 0nd period. We now examine the evidence on economic, economicwdernographic and demographic changes and tradeoffs in each of the developing~country regions, and compare and contrast them for the two time periods. First, Africa, with a prima facie low rural population density (but not low when water supply limitations are taken into account), experienced a modest decline in arable land and in per capita agricultural production during the 19603-19703. During the later 1980549903 period, rural labour density increased rapidly, despite notable expansion of agricultural land and some out—migration because of continuing high fertility. The economic crisis pervading the region through both time periods, combined with low levels of education, undoubtedly contributed to the lack of increase in agricultural productivity. In the first period, the only response to rising population density was out—migration, but, as pressures on the land increased considerably, in the second period multiple responses are evident, with land extensification, higher rural outmmigration and the onset of a fertility decline. Since we expect extensification to precede intensification, in the case of Africa in the latter period, the lack of intensification and the increasing desPeration of rural populations may have led to pushing out the agricultural frontier as populations sought land to- feed their families. The Near East has little arable land that is not already used, so increased production has generally been possible only through increased irrigation. How»- ever, during the first time period, rural population density remained low and the only response observed was out-migration. But, as the rural population became denser during the second period, the full range of responses occurred-including a large increase in agricultural output per rural worker (presumably as the Green Revolution technology made its way into the region) and an acceleration in the pace of fertility decline, which began to approximate that of other regions. Looking at each period in terms of tradeoffs, in the first period the lack of signif— icant other responses put more pressure on out—migration; in the later period, continuing increases in rural density are associated with multiple responses. China experienced rapid increases in rural population density in both peri— ods, especially in the second. Agricultural land actually declined significantly in the first period and slightly in the second (the latter confirmed in Heilig, 1999). China rapidly adopted new agricultural technology in the earlier period, but the Food and Agriculture Organization of the UN (FAQ) data suggest no further intensification in the latter period (which is inconsistent with recently reported macroeconomic indicators of the agricultural sector). Also, in the first period it is known that China restricted migration to urban areas and imposed a onevehild policy, which forced a substantial fertility decline. Whether such policies should be considered completely exogenous or as being themselves induced by population pressures is a moot point, but the changes did occur. In the more recent period, with economic liberalization, marily demographic, with a continuing rapid decline i ruraluurban migration. China has evidently experie decline of all regions, in both time periods. In terms of tradeoffs, in the absence of a land response, in the first period, China responded to an increase in rural population density by land intensification and a draconian fertility control policy. In the second period, both fertility decline and out—migration occurred. In the ‘Other Asia’ region, rural density increased only slightly in the first period but rapidly in the second, so we would expect much more in the way of responses in the second period. This is indeed the case, with little change in agri— cultural land or fertility in the first period, and only modest rural—urban migra— tion. The only major change was agricultural intensification, assooiated with the Green Revolution in India and other countries in the region. In the second changes appear to be pri— n fertility and acoelerating' need the swiftest fertility- mummnw'g v.4... . ._.,. .. v. .. Us... u ...v -. A .' orn rural areas. In the first time y and the success of intensification deferred pressures to reSpond in other ways, while in the later period rapidly rising den— sity tended to stimulate multiple responses. Government policies also played a role in encouraging fertility declines in some countries such as India, lndonesia and Thailand. Furthermore, land availability in many countries was associated ~ as expected a- with continued land extensifidation in the second period «influc .____.-.. Fume... - "U :3. O “9-. M :3" (b :3" . O w. 2“ . 9-i- I?! Q (9 m o: {‘0 E. 5%“ B :. the changes observed in the suite of eco demographic reSponses to be strongly linked to population pressures in general. i: In the first time period, substantial extensification occurred, befitting the region’s plentiful land endowments. At the same time,'agricultural productivity rose and rapid rural out—migration occurred, associated with urbanization, which had begun earlier. Fertility started to drop only in the 1970s in most countries, associated with public— and private—sector familywplanning provision and the demand to reduce actual fertility to desired levels, impelled by high levels of urbanization and improved women’s education, following cumulative trends over previous decades. In the second period, associated with no increase ‘ in density, there was much less agricultural extensification and intensification, 1 x but rural out—migration and fertility decline continued. Scant economic oppor- E , tunities in rural areas, land inequality, considerably higher wages in cities, ‘ * _ and the urban bias in development policies contributed to consistently high ’ rural-urban migration: from the 19605 to the 19805 nearly 2% of rural ' Americans out-migrated annually (Chen et ad, 1998, p. 82). While the changes i - .a-— F...» .mpW-M. ....—,.—..—1v-.- '=‘ 9-W- - ' .. observed in Table 3.1 in Latin America would appear exogenous relative to changes in population density, substantial differen country which may be related ces exist from country to to changes in rural population growth and density at the country level (see next section). Although the data here illus erstand the interrelationships between trade— l we examine the agricultural sector in Latin America at empirical work is needed to und offs. In the next Section, Population, Land Use and the Environment 43 1;? n: {it ' 3 the country level, looking at trends in agricultural extensification and intensifi» cation and their possible relationships to population change. Population Density, {and Use and Deforestation in Latin America Central and South America are different from other developing regions in hav— ing much more potential arable land available, more current agricultural land in pasture, higher levels of forests being converted to agricultural land and in being far more urbanized. Amounts of land being deforested were more than double those of other continents during the first half of the 19905 (World Bank, 1998b). But differences are substantial within the region. For the present analysis, it is important to distinguish between Central and South America. First, resource endowments are very different, largely due to the presence of the world’s largest tropical forest, the Amazon, which covers parts of eight South American nations. Secondly, fertility has declined far more broadly in South America than in Central America (taken here to refer to Mexico and the Caribbean as well). Taking into account both lower rural fertility and significant net rural—urban migration, the rural population in South America as a whole actually declined slightly from 1984 to 1994. As a result, while mean rural population density in South Americawas less than half that of Central America in 1966, it has since fallen to about a third. That there is a strong relationship between population density and the per- centage of land area covered by forests is hardly surprising, and is evident at a global level in the differences in the percentage of land area covered by forests in Asia and Eumpe compared with sub—Saharan Africa and the western hemisphere. A much more sensitive - as well as contemporary and dynamic - test of the relationship is that of first differences -- that is, of the relationship betWeen population change and reduction in forest area. ' - ' “ - The difference in the two regions of Latin America is significant not only because greater population densities can be expected to be associated with greater cumulative levels of deforestation, but also because countries with small areas remaining in forests can have very high rates of deforestation even when the absolute areas of forests being lost are small (e.g. countries in Central Amer» ica). Other countries with large forest stocks may experience very low annual . rates of deforestation even when the absolute area being cleared is huge (eg. Brazil); The usual simple measure of national population density (total popula— tion/ total land area) ignores the fact that countries have large areas unusable for ' '.agriculture (e.g. mountain ranges, deserts, bodies of water and urban areas), and also confuses the issue by combining urban and rural populations. We therefore measure population density by rural population density (rural population per unit of arable land plus land under permanent crops), and use this for analysing population impacts on land use and land-cover change. The numerator is modi~ fied through population change (fertility, mortality and net migration), while the'denominator is affected by changes in the amount of arable and permanently cropped land. we exclude pastureland from the denominator since it has little: direct relationship to population density. Rural population dynamics, forest~cover change and agricultural extensification Urbanization and rural fertility decline occurred earlier in South America than in Central America. In fact, during the last three decades, Central America gained approximately 8 million rural residents, while South America lost 3 million. With the exception of pioneering countries, such as Costa Rica and Cuba, where fertility decline began in earnest in the 1960s, it was not until the 19705 that fertility began to fall generally in Latin America. None the less, fertility decline has been delayed in a number of Central American countries (Guatemala, Honduras, Nicaragua and Haiti). Table 3.2 shows that in Central America, from the 19603 to the mide19903, rural population density grew (from 2.2 to 2.4 persons ha” arable and cropped land), as modest increases in the area of land in crops (29%) were still outpaced by rural population growth. Only in the Dominican Republic and Honduras did growth in agricultural land exceed population growth. In South America, in r ' contrast, greater agricultural expansion, earlier and greater rates of rural~urban migration and earlier fertility decline together led to a slight decrease in overall rural population density, from 1.0 to 0.9 persons per hectare. Latin America harbours the greatest area of closed tropical forests in the world, with Brazil alone containing a third of the world’s tropical forests. The Amazon basin also contains 45% of all the fresh water on the earth, and is the planet’s largest carbon sink. Its preservation is thought to be crucial for moderating global warming and regulating global weather systems (e.g. Adger and Brown, 1994; rlinker et al, 1996). Yet Latin Americans have been clearing much more forest than those elsewhere: during the first half of the 19905, Latin 40 times more than Asians (derived from data in FAQ, 1997b). Furthermore, reforestation has only} minimally compensated for forest losses in Latin America, and far less than in other regions. Thus, in the Asia—Pacific region, reforestation compensated for over 50% of the forest loss from 1981 to 1992, while in Latin America it was around 6% (FAQ, 1997b). panied by a 5% increase in rural population density. But in the ensuing decade L or so, from 1976 to 1984, deforestation accelerated sharply in both Central l America and South America. High rates of deforestation were accompanied by l increases in rural population density of 40% and 8%, respectively. Finally, deforestation in the latest decade (1984 to 1994) was much lower in both regions compared with the previous decade, and was accompanied by increased agricultural intensification, particularly in Central America. L‘- «'x "IAN .75; n .. ‘- Qipuq P1. I1 U1 ua up u :5 Population, Land Use and the Environment 45 Table 3.2. Latin America: rural popuiation density and percentage of land forested. Rural populatiori per hectare of agricultural land Forest land/total land 1%)3 Country 1966 1976 1984 1994 1966 1976 1984 1994 Central America Costa Rica 1 .9 2.4 5.0 3.2 54 SO 32 24 Cuba 1.9 1.1 1.1 0.8 14 11 24 a. 22 Dominican Repubiic 2.8 2.8 2.6 1.9 23 23 13 10 El Salvador 2.9 3.7 4.8 4.3 11 13 6 4 Guatemala 2.1 2.3 3.7 3.2 56 54 42 34 Haiti 4.6 4.5 7.8 5.3 11 7 5 3 Henduras 2.1 2.4 1.6 1.5 63 63 54 43 Mexico 0.8 . 0.8 1.0 0.9 40 ‘36 25 22 Nicaragua 0.7 0.8 1.3 1.3 49 53 33 26 Panama 1.3 1.6 2.3 1.8 49 55 51 41 C. Am. totaE 2.? 2.2 3.1 2.4 37 36 28 23 C. Am iota:b 1.1 1.1 1.4 1.3 40 38 27 24 South America Argentina 0.2 0.1 0.2 0.2 22 22 19 18‘ Bolivia 1.3 0.8 1.3 1.2 55 52 53 47 Brazil 1.3 1.1 0.8 0.7 62 60 60 57 Chile 0.5 0.4 0.5 0.5 28 28 22 21c Coiombia 1.7 1.8 1.9 1.8 74 74 50 47 Ecuador 1.0 0.8 1.8 1.6 66 65 56 47 Paraguay 1.4 1.6 1.0 1.0 52 51 46 35 Peru 2.1 1.8 t] 1.6 58 58 53 53C Uruguay 0.3 0.2 0.3 0.2 3 3 5 5c Venezuela 0.5 0.5 0.6 0.4 54 54 36 32 S. Amtotai _ 1.0 0.9 1.0 I 0.9 47 47 41 37 S.Am.totaib 0.8 0.7 0.7 0.7 53 52' ' 48' '45 aUnless otherwise mentioned, all 1994 data on forested land were caiculated averag— . Eng deforestation rates from 1981 to 1990 from FAO. bWeighted totals. CEstimated from deforestation rates from 1990 to $995 in FAQ (1997a). crops for human consumption. A correlation between land in crop production and population density is strikingly evident at the regional levei. Thus, when rural pepulation densities are computed with the total land area in the denomi— nator, rural Central America was seven times more densely populated than rural ‘ South America in 1966 and 11 times more in 1994. When rural population density was measured using only arable and permanently cropped land in the denominator, it was never more than three times as high in Central America ‘ during the period. As is evident from Table 3.3, the difference is because Central i = America had much more arable and permanently cropped land (15% compared with 6% in 1966, and 21% compared with 6% in 1994). ‘ my. mail Table 3.3. Extensificatlon: percentage of land in pasture and arable and cropped land. A8113 land and Pastureland as % pastureland as % of A&F’ land as % of of total land total land total land Wm M W Country 1966 1994 1966 1994 1966 1994 Central America Costa Rica 23 46 32 56 1O 10 Cuba ‘ 24 27 39 58 15 31 Dominican Republic 24 43 43 74 18 31 El Salvador 29 29 59 65 3O 35 Guatemala 9 24 23 42 14 18 Haiti 22 18 49 51 2 7 33 Honduras 18 14 25 32 7 18 Mexico 37 39 50 52 13 13 Nicaragua 28 45 40 56 12 10 Panama '13 20 21 29 7 9 Total (weighted) ‘ 32 36 46 51 13 15 Total (unweighted) . 23 31 38 51 15 21 South America Argentina 53 52 64 62 1 1 1O Bolivia 26 24 2 7 27 2 2 Brazil 17 22 20 28 4 6 Chile ‘34 l 8 20 24 6 6 Colombia 16 39 21 44 5 5 Ecuador 8 18 20 29 12 11 Paraguay 35 55 38 6O 2 6 Peru 2‘! 21 23 24 2 3 Uruguay 78 7 7 89 85 10 7 Venezuela 17 - 20 23 25 6 4 Total (weighted) 24 29 _29 35 5 6 - Total {unweighted} 29 35 35 41 6 6 A&P, arable and permanently cropped. As illustrated In F1g 3 2, most agricultural land 1n Latin Amerxca is 1:1 pasture The conversron of forest to pasture cannot be drrectly baked to rural ; population growth. Thus, the more densely populated an area, the more likely it 7“ is to have land in crops rather than pasture. Indeed, as the rural population grew ‘ by 47% from 1966 to 1994 in the already densely populated region of Central , America, cropland expansion kept pace with growth in pastureland. Con— ‘; versely, in less densely populated South America, the rural population increased by only 13% per decade on average and the percentage of cropland remained u-Lée-‘cx-aée-Maaas was»? Population, Land Use and the Environment 47 W (a) 60 50 ,3 40 Pasture 2‘; m 30 9 ‘1 20 Arabia and permanentiy ‘EO cropped tancf {13) 45 40 35 30 .s s? astute 91: 25 3 g 20 &. Arable and 1 5 permanently cropped 10 land iseé 1976 ' 1984 1994 Fig; 3.2. Pasture and arable and permanently cropped land as a percentage of total land (a) in Central America and (b) in South America. However, it is important to note that much pastureland in Latin America is converted from cropland and not from forests directly (see, for example, Hecht and Cockburn, 1990; Nations, 1992). This is most clear in the case of Brazil, where the total growth in pastureland from 1966 to 1994 equalled the total area of forest eliminated. Below we review landmuse changes in more detail, first in Central America and secondly in South America. ' ' I -‘ CentralAmerica "-."-"‘1 - The percentage of forests cleared in Central America in recent decades is astounding. The area in forest extant in 1966 that was cleared by 1994 surpassed the area of forestland remaining in 1994 in all but three countries in the region. Still, the percentage of forests lost was less than the percentage growth in the ‘- rural population in every country. For example, Guatemala cleared 38% of its : forests, and was left with but 35% of its initial level, while the rural population "grew by 97%. Between 1966 and 1994, more than 20% of the total national ‘ -‘ territory was deforested in Guatemala, Nicaragua and Costa Rica; at over 7% ‘per annum, the deforestation rate in Costa Rica during the 19703 was probably ' the highest in the world, reducing its stock of forests by over a half in a single "decade. Three nations in the region cleared more forests in the 28 years than they had remaining in 1994: El Salvador, the Dominican Republic and Haiti. In these nations, deforestation rates must fall in the future, since there is little left to clear. South America “-muuuw «NU U-L. Lam As in Central America, forest loss in South America rapid pace in recent decades. But rural population densit lower than in Central America, and many of the South had no increase in the size of t period. As a consequence, chan rural pepulation growth cannot has proceeded at a y is generally much ' American countries heir rural populations over the observation ges in rural population density due to net explain much of the deforestation in South I America. Government policies favouring cattle, especially including subsidies ' for ranching in Brazil in the 19803, strong export markets and rising domestic demand from growing and increasingly prosperous urban populations help to explain the large growth of pastureland in South America. In addition, policies to keep the prices of basic grains low to benefit urban consumers at the expense . of rural producers have provided incentives for farmers to switch from crops to cattle. Over the 1966w1 994 period, Brazil had the lowest percentage rate of deforestation in Latin America (8% , while its rural population declined by 16%. But, in absolute terms, its area of forest loss, primarily in the Amazon basin, far exceeded that of any country in the world. In industrialized countries such as Argentina and Chile, rural populations also declined but deforestation rose (associated with an increase in timber harvesting and an expansion of export agriculture, rather than with population growth). On the other hand, in a few South American nations — for example, Ecuador and Paraguay m rural able and permanently cropped land by nearly a half and a more than doubling of pastureland. By 1994, Ecuador’s share of total land in cropland (11%) was the highest in South America, while its share in pastureland remained the lowest, albeit rapidly growing. Even more than in Central America, however, it is the conversion of crop— land and forest to pasture that has been the most dramatic land—cover change. Thus, while Colombia’s rural population grew slowly, by only 10% over three decades, pasture grew by 144%, while its crop area remained constant at only 5% of its territory — barely one—ninth the area in pasture in 1994. In Venezuela, the rural population declined 44%, and yet 41% of its forested land in 1966 was still cut by 1994. Evidently, even where rural populations are declining, agricultural frontiers have still expanded. Although rural—urban migration continues to be significant and to contribute to urban growth,4 ruralwrural migration is of growing importance, with" people migrating from rural areas characterized by high poverty and unequal land distributions to other, more remote, rural areas in search of land to continue agricultural livelihoods. Thus, contrary to what some scholars expect, even in con ' ' substantial deforestation can still occur. On the remote frentiers, colonists are migrating along newly established roads, convertin g forests to cropland and pasture (cg. Rudel, 1983; Rudel and Richards, 1990; Pichén, 1997a,b; Sader et 4.1., 1997). Population, Land Use and the Environment - 49 W A comparison. Despite considerable country—to-country variation, some patterns do emerge. First, where population density is lower, the level of deforestation is lower, but the rate of deforestation per capita is higher (e.g. South American nations compared with Central America; on a global scale, the same comparisons hold for Latin America relative to Asia). Secondly, countries with the highest rural population densities (eg. Haiti and El Salvador) deforested their land earlier. Deforestation rates remain high because of a shrinking base, but these countries have increasingly lower deforestation rates per rural dweller as forest reserves approach extinction. Thirdly, pastureland accounts for a greater proportion of agricultural land than cropland in Latin America, and its share has risen in recent decades. To maintain current per capita food consumption levels in coming decades, Latin America will need to greatly increase food imports or increase domestic production through intensification of current agricultural land and/ or expan— siou of the land area in food production. In countries where little forest remains, as in many Central American nations, intensification will be the main option for increasing food production. On the other hand, among South American nations with large endowments of Amazon jungle, extensification is still occurring on a large scale, and could continue. But the sustainability of Amazonian agriculture is dubious as most of the region’s biomass is rooted in nutriennimpoverished soils (Buol, 1995). Agricultural intensification Starting in the 19503, international agencies promoted technological advances in agriculture in developing countries in what became known as the Green Rew olution. During these years, new seeds were developed for basic grains, and irri— gation, pesticide and fertilizer use all grew dramatically. Most of the growth in agricultural production of the past-30years is attributable to increases inyields-. per hectare: from 1961 to 1996, global cereal yields increased 107%, while the area harvested increased by only 10% (Bender and Smith, 1997). Only in Latin America has the expansion of the agricultural land area been an important contributor to the expansion of agricultural output, but this share has been declining and is now less than that of increases in yields. 0 3, In developing countries, agriculture accounts for 80% of water use (IFPRI, d 1996). The growing scarcity of fresh water limits its contribution to increasing 1f agricultural yields in much of the world, though this is not a pervasive problem .e in Latin America. The percentage of agricultural land irrigated rose from 9% to a, 14% in Central America and from 11% to 14% in South America from 1966 to ‘e 1994. As is evident from Table 3.4, the countries that increased irrigation most d ' were not those with the highest rural population densities or the most rapidly growing populations, but, rather, those with large export operations based on semi-«arid and arid lands (Mexico and Peru). mt. mrsoorrow and D. L,- Cay, indicators of agricultural intensification in Latin America. Percentage of A&P land Fertilizer use (kg her1 of irrigated cropland} ' . l W W 1964~ 1974— Country 1966 1976 1984 1994 1966 1976 1983 1993 Central America ‘ Costa Rica 5 5 21 24 59 137 171 208 Cuba 29 23 25 27 134 103 164 52 Dominican Republic 13 14 13 17 12 68 39 61 El Salvador 3 5 15 16 71 147 113 106 Guatemala 3 4 6 7 14 36 38 87 | ., Haiti 6 8 8 8 0 2 4 5 Honduras 8 9 4 a 6 12 16 32 . Mexico 14 17 20 25 15 42 60 71 Nicaragua 1 S 2 7 17 22 56 21 Panama 3 4 5 5 1 7 47 4O 48 Averagea 9 9 12 14 34.5 61.6 70.1 69,1 - South America -- I I . Argentina 5 5 6 6 1 2 5 1 1 Bolivia 4 4 6 4 1 1 4 6 Brazil 2 3 4 6 5 31 45 85 Chile 24 22 29 30 26 25 18 58 Colombia 5 6 9 14 28 44 '61 94 Ecuador 14 10 21 18 6 14 30 31 Paraguay 4 5 3 3 2 1 5 14 Peru 41 35 33 41 29 38 22 44 Uruguay 2 3 7 11 29 40 ' 31 72 Venezuela 5 6 5 5 1O 38 41 65 Average.“ ‘11 1O 12 14 13.7 23.4 26.2 418 aUrlweighted. A811”, arable and permanently cropped. A Sampling of Country Case Studies The studies summarized below are selected to represent the diversity of experiences across countries and the fundamental effects of contextual factors, such as natural resource endowments and government policy. This is by no means intended to indicate that all cases are equally likely. Population growth and agricultural extensification In Ecuador (as well as many other countries with tropical forests), population growth in the Amazon region has led to a Widespread process of land clearing and deforestation by agricultural colonists in recent decades. With the discovery of significant petroleum deposits in 1967, oil companies built roads to extract Population, Land Use and the Environment 51 and transport the oil from the northern Amazon starting in the 1970s. This facilitated an influx of migrant colonists, coming largely from the highlands (Pichon, 1997a; Pichon and Bilsborrow, 1999). Population grew at annual rates of 8% in 19744982 and 6% in 19824990 - '11 both cases more than double the national average -— while deforestation was widespread in the region and led to Ecuador’s overall annual deforestation rate of 1.8%, the highest of any Amazon basin country. By the time of a detailed household survey of migrant colonists in 1990, 44% of the colonists’ plots (most 40-60 ha) had been deforested. Ongoing work in the regionby Bilsborrow and others indicates continuing rapid population growth andagricultural extensification, due as much to high fertility in the region as to new immigration. Population decline and agricultural extensification In apparent contradiction to theory, in the Brazilian Amazon during recent decades, deforestation has accompanied a declining rural population; the rural population decreased by 16% from 1966 to 1994 and, since the early 19803, even the rural population in the Amazon has not grown. Poor tropical soils, underdeveloped transportation infrastructure, titling delays, tenure insecurity and lack of credit severely constrained the agricultural development prospects of many migrant farmers within a few years of their arrival. Many migrants abandoned their farms to work in mining or timber harvesting or to move to cities. Large—scale ranchers engaged in speculation, buying up land and sometimes forcefully removing farmers (Hecht and Cockburn, 1990; Stewart, 1994). Much of the abandoned land was converted to pastureland, which is a far more extensive use of land than growing crops, contributing to ongoing deforestation, despite a decreasing rural population. I While demographic factors appear absent as proximate causes of this pro- cess, both high (earlier) fertility and-migration areamongtheundcrlying factors, Many of the migrant settlers initially came from north~eastern Brazil, where high fertility, extreme land inequality and recurrent droughts led to population pressures on the land and poverty, which impelled the population to migrate in search of solutions, mostly to cities but many to the Amazon frontier as well. It density is associated with rapid deforestation at the national level, due, in large part, to rural—rural migration from densely populated areas to the Amazon. Population growth and agricultural intensification A direct link between rural population change and land-use change can be ascertained better when confounding economic and political variables are absent. Such is the case in traditional (non—monetized) agrarian societies, which are rare today. However, some scattered evidence providing a quantitative test of Boserup’s intensification hypothesis has been brought together based on data from, studies of 29 tropical subsistence agricultural societies around the world is striking that, in both Brazil and Ecuador, a decline in overall rural population I w m a...“ 14¢. Lani (Turner et 41., 1977). Turner and colleagues observed a significant positive" cross-sectional relationship between population density and agricultural intensity, with the latter measured as the proportion of time the agricultural land was in use.5 The best fit was from an exponential model, suggesting that agricultural intensity is much higher as population density rises m evidence in support of diminishing returns. A later study by Turner and All (1996) examined the link between growing population density in six villages (265 households) in Bangladesh from 1950 to 1986. They noted that the lack of available unused land dictated intensification as the only agricultural response (91% of the land was cultivated at least once per year). Yet population density differed greatly across the six villages, with three much more densely populated than the others (means of about Finally, in a study on three districts of Tanzania, Meertens et 4!. (1996) observed changes over a century based on historical records recent field studiesf‘They concluded that increasing population densities have led to intensi~ fication, but that the process has been shaped by agroecological conditions (such as differences in rainfall and type of land). They also noted that the land frontier was exhausted first, followed by out—migration, before intensification occurred. 22:22:" Population growth and out~migration Among the several recent studies on Africa, that of Netting er a]. (1989) stands out. They looked at how the Koyfar of the Joe Plateau in Nigeria adapted to population growth between the 1950s and the 19803. For decades before, much of the population migrated seasonally to the nearby Benue valley (lowlands) to grow yams to supplement their subsistence production in the plateau homeland. The rapidly growing internal market for yams led to this migration eventually linkages are directly related to the atypical advantages of having a large area of unused, unowned, fertile land nearby, as well as a growing market for yams. Finally, it was noted that certain cultural factors also played important roles in the positiVe adaptation — household autonomy, work ethic and lack of local autocratic rulers and institutions. Population growth and multiphasic responses Research which has attracted considerable attention recently is that of Tiffen and colleagues on the Machakos district, near Nairobi, Kenya (Tiffen and 'r v H. nag. (D r- a Population, Land Use and the Environment - 53 Mortimore, 1994; Tiffen, 1995). Even as population grew fourfold over the period 1930—4990, the value of agricultural output per capita tripled, mainly from huge increases in the production of cash crops in the 19605—19703 and in horticulture in the 19803, while livestock production declined. This reflects a more intensive use of land, as expected by Boserup (1965). Meanwhile, environ— mental conditions actually improved, with reforestation and the creation of broad labour~intensive bench terraces, both of which saved water and reduced erosion. Still, other changes also occurred, together comprising a multiphasic response, with the first in the 19305 being extensification of agriculture to inferior land and substantial out-migration of males for wage labour, both prior to intensification. indeed, remittances and savings of the out-migrants later financed an increase in education (human capital) and the creation of non—farm businesses, which probably contributed to the later intensification. More recently, increases in local employment opportunities (including non—farm) have led to a decline in male out—migration. Finally, a demographic transition now appears to have begun since the 19803, which may reflect increasing. difficulties in achieving further extensiiication (there being no unclaimed land by 1980 (Tiffen, 1995)) or intensification (the value of output per hectare failed to rise as much as population during the 19803 (Tiffen, 1995)). Nevertheless, it should be noted that in Kenya, generally, fertility is now declining, as policies initiated in the 19703 begin to find ‘fertile ground’. :‘éfiufiéu-tt-zs-a-m,-.;.c.':. a, .. ll: Summary, Gaps in Knowledge and Research Needs The data presented here on linkages between population dynamics and changes in land use in Latin America and other developing regions of the world suggest that neither neo—Malthusian nor Boserupian models offer satisfactory explana~ tions of population-«environment linkages. Complex processes are difficult to ’ disentangle on a country scale. 'Su'bnationai-data-for' provinces or even smaller - political units are better suited for testing the interrelationships because they permit taking into account additional factors, including contextual factors that differ from one area to another. Evidence of linkages between population dynamics and the various eco~ nomic, economic—demographic and demographic responses appears stronger in other regions than in Latin America for the 1960s to the mid—19903. This is because, unlike in the other regions, rural population density in Latin America rose only slightly during the first half of that period and actually fell during the latter half. Thus, at a continent—wide level, it is not plausible to expect a relation— ship between changes in rural density and induced responses, since there was nothing inducing the response. However, this does not mean that such relation— ships did not obtain for some countries in the region, as discussed above. Unfortunately, a major shortcoming of cross-continent and cross—country (Latin America) studies, such as those summarized here, is the egregious lack _ of reliable and comparable data, with the result that many of the data used ‘ are based only on ‘guestimates’ by officials in the countries or at the FAQ. For example, few countries in Latin America or Africa carried out national !\.E. DHSUONOW and 0.1.. Car agricultural censuses in the 19803 or 1990s. Although studies of the same relationships (using appropriately different measures) based upon farmdevel data could be even more useful than those based on cenSus data for testing population—extensification/intensification linkages, few data sets are available. With these caveats in mind, let us briefly review sorne patterns that . emerge on population and landnuse/landwcover patterns during recent decades - (1966*1 994) in Latin America, focusing on the differences between Central and South America. Rural population densities in South America were already Start» ing to decline by the early 19805 or before. This was due both to the opening up of new lands for agriculture in frontier areas (mainly forests) and also to declines in fertility and outwmigration from rural areas by people searching for better opportunities in urban areas and frontier regions. In Central America, in contrast, the rural population continued to grow throughout the 19664994 period in virtually all countries. Rural Central America was seven times more densely populated than rural South America in 1966, but 11 times more in 1994. Asa result, the connection between population dynamics and land—use change should be greater there. However, even in countries such as Brazil, where rural population density is,,declining, demographic factors are important underlying factors in “the expansion of the agricultural frontier and deforestation. One key factor is rural—rural migration, rather than high or rising rural population density due to high population growth and high fertility. Thus, it is crucial to ascertain the origins of rural—rural migrants — and conditions in the places of origin —~ to determine the ultimate causes of that migration and hence of the forest degradation that occurs in other parts of the country. The fact that extensification has continued unabated where land is available is sobering, given diminishing forest reserves worldwide and the ecological imperative of tropical forest conservation. Economic and demographic changes have contributed to the widening gap between those regions moving towards rapid rural and overall development (China, South~east Asia and some countries in Latin America) and others which are stagnant (Africa and much of the rest of Asia and Latin America). While demographic responses, especially further fertility decline towards replacement fertility, may help in confronting I 7.; these problems, and while some further extensification will prove necessary, mechanisms need to be found to achieve much more successful intensification of agriculture, especially in the poorer regions of the Third World. . . Unfortunately, there are growing signs of limitations to intensification. The FAQ estimates that two—thirds of the global increase in agricultural production from 1990 to 2010 will need to come from increasing yields, with only 21% from expanding the land area and 13% from increasing crop intensity in the sense of shorter fallow periods (WRI, 1996). But the cost of expanding production is rising. Of the remaining potentially usable land, half is already in protected forests, while three—quarters has serious soil or terrain (topography, drainage, etc.) constraints; meanwhile 10% of current agricultural land is already moderately or severely degraded (WRI, 1996). By the early 19903, three~quarters of the world’s wheat and rice acreage and half the land in maize were already in highmyielding varieties (HYV). And irrigated land, accounting Population, Land Use and the Environment 55 for one—third of the world’s agricultural production (with 60% of the irrigated land in Asia, and only 5% in Africa and 4% in Latin America), is increasing at a slower rate, growing 2% per annum in the 19703 and 19803 but at less than 1% in the 19903 (WRI, 1996). The main inputs expected to contribute to further intensification are chemical fertilizers, the usage of which is expected to double- during the period 1990—42010. Chemical fertilizer usage levels in Africa are currently only one—seventh of those in Asia. Still, increases in fertilizer use have led to smaller increments in production in recent decades (Brown, 1997). With continued population growth, diminishing available land and future intensification constraints, policy—makers at all levels will be challenged to improve the agriculture—population nexus in developing countries. Successful policies may include: (i) helping rural farmers intensify production through more technical assistance and credit targeted to raising crops rather than cattle; (ii) improving access to reproductive health and familywplanning services in rural areas; and (iii) better conserving what is left of the precious vestiges of tropical forest through higher royalties for logging concessions, more rational road—building policies and involving local populations in protecting conservation areas. 1 Among the nations of the temperate regions, only Finland and Japan (at approxi— mately 0.2% annually) reported positive deforestation rates in the 19905 (World Bank, 1998b). In contrast, all tropical nations are experiencing deforestation at a rate of 0.5% or greater annually. 2 A more extreme version of Boserup’s hypothesis, which is discredited by empirical evidence, is that population growth is automatically a boon to the developing world by inducing technological innovation, enabling output to rise faster than population (Simon, 1977, 1981). . 3 The use of the same five regions continues to present difficulties for the Near East, " which we intended to comprise all countries. from Morocco across northernAfrica and I} western Asia to Afghanistan. However, because of the almost total dependence on oil of many countries and/or the lack of adequate data (cg. Syria, Algeria, Morocco, Tunisia), u. m" V} m uz‘ ; the final sample comprised only four countries — Egypt, Iran, Iraq and Jordan. In these r countries, a couple of values seemed implausible and were omitted, namely, the increase ’ in agricultural productivity in Iraq of 158% in 19804 995 implied by FAO data. 3 4 Chen et all. (1998) find that rural—urban migration accounts for about two~fiiths of urban population growth in the region, with the rest due to natural population growth [3 (the difference between the fertility and mortality of the urban population). In 5 i 'The standard definition was used. Land in use one year followed by 4 years in fallow V would yield an intensification measure of 20 (20% of the time), while land used for :1 doubleucropping ina year leads to a measure of 200. l 1% in ,Y, _ is 33, ze ...
View Full Document

This note was uploaded on 02/03/2011 for the course EEP 153 taught by Professor Marsh during the Spring '11 term at University of California, Berkeley.

Page1 / 21

lecture 8 reading---Bilsborrow,+2001+_CAB+Intl_ -...

This preview shows document pages 1 - 21. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online